Global Biomarkers Consortium/ World Cutaneous Malignancies Congress: PMO Live

SEATTLE – Research in colorectal cancer should focus on part on a handful of high-priority needs for prognostic and predictive biomarkers, according to Dr. William M. Grady, an oncologist at the Fred Hutchinson Cancer Research Center and section chief of gastroenterology at the University of Washington, Seattle.

“These needs are in the selection of optimal agents for the stage II and III cancers, the ones for which we add adjuvant therapy,” he told attendees of a joint meeting by the Global Biomarkers Consortium and the World Cutaneous Malignancies Congress. Additionally, “in the clinic, the main issue is selecting stage II patients who are at high risk for recurrence who may benefit from adjuvant therapy, and then stage III patients who will not benefit from therapy.”

There are some high-priority needs in the metastatic setting as well. “For stage IV disease, really the issue is selecting candidates for conventional chemotherapy, and here we are trying to balance toxicity versus survival and quality of life. Also, there is the issue of identifying people who will be optimal responders to some of the targeted therapies,” Dr. Grady elaborated.

The quest for biomarkers has been aided by increasing recognition of the molecular heterogeneity of colorectal cancer. Research has thus far identified three predominant molecular groups – a microsatellite unstable group, a microsatellite stable group, and a CIMP (CpG island methylator phenotype) group – having alterations of different signaling pathways. “This is clearly an oversimplification, and where I think we are going is that we will continue to parse out hopefully finer and finer groups that will be used for directing clinical care based on the response to agents that are being developed,” he said.

Microsatellite instability

Microsatellite instability is already used clinically to screen colorectal cancers in patients younger than 50 for hereditary Lynch syndrome, as recommended in national guidelines, according to Dr. Grady.

It can also be used to assess the prognosis of stage II and III disease, as patients with tumors showing microsatellite instability have better survival (J Clin Oncol. 2005;23[3]:609-18). But studies have been mixed on whether this biomarker is predictive for the response to adjuvant 5-fluorouracil–based chemotherapy (Nat Rev Clin Oncol. 2010;7[4]:197-208).

“Right now, the consensus is that for stage II disease, if there is microsatellite instability, that correlates with a very low-risk group and there doesn’t appear to be any benefit from adding adjuvant therapy; in fact, one study even suggested harm,” he noted. “For stage III disease [with microsatellite instability], there is no change … indicated at this time for 5-[fluorouracil]-based therapy; potentially, some people even argue that you might want to avoid this therapy, although I don’t think that’s the general opinion. And there may be a benefit in these tumors from the use of oxaliplatin, which one study last year suggested.”

In patients with stage IV disease, recent data suggest microsatellite instability may be a marker for benefit from therapies that target the programmed death–1 (PD1) pathway such as pembrolizumab (Keytruda) (N Engl J Med. 2015;372[26]:2509-20). “The challenge here is that only about 2% to 3% of metastatic colon cancers have microsatellite instability, so unfortunately it’s a small group that will benefit,” Dr. Grady commented.

Multigene signatures

A pair of clinically available multigene signatures, Oncotype DX–Colon and ColoPrint, are being used to estimate prognosis in patients with early disease. However, differences between adjacent assay-defined risk strata are fairly small, on the order of an absolute 10% or so. “This raises a question about what the clinical utility of these scores is,” he said.

“The real question we want to know, though, is can we use [these signatures] to choose people who will get adjuvant therapy for stage II disease,” he commented. At present, the answer appears to be a qualified yes for Oncotype DX, with some data suggesting that the survival benefit of chemotherapy may be greatest in patients with early disease who fall into the assay-defined high-risk category (J Clin Oncol. 2013;31[36]:4512-9)

“Where the ColoPrint assay seems to potentially have the highest benefit is in identifying a low- and high-risk group in the patients who have T3N0 disease but that’s microsatellite-stable. It doesn’t add much to those who have tumors that are already identified as being microsatellite-unstable or who have large [T4N0] disease that is microsatellite-stable,” he noted (Oncologist. 2015;20[2]:127-33).

Ultimately, use of Oncotype DX and ColoPrint should be individualized, according to Dr. Grady. “They certainly can be used, but they really need to be discussed on a case-by-case basis with the patient and physician because they have unclear clinical utility,” he said.

CIMP biomarker

Research suggests that CIMP provides prognostic information in patients with colorectal cancer, according to Dr. Grady. But it also may predict benefit from the addition of irinotecan to 5-fluorouracil and leucovorin in patients with stage III disease having intact mismatch repair (Gastroenterology. 2014;147[3]:637-45).

“CIMP is a potentially emerging marker for the use of cytotoxic agents, probably the most promising one today,” he commented.

Extended RAS assay

The extended RAS assay, which evaluates the whole exome for mutations in both KRAS and NRAS, is useful for guiding decisions about therapies that target signaling through the epidermal growth factor receptor (EGFR), such as panitumumab (Vectibix). However, its value is mainly limited to identifying patients who will not respond to this therapy, according to Dr. Grady.

By incorporating NRAS, this combination assay picks up an additional 10% of patients who will not respond to anti-EGFR therapy (J Clin Oncol. 2015;33[7]:682-5). But even so, it identifies only about a third of all nonresponders.

“We really need better biomarkers to identify patients who will respond, as well as identifying those patients who will not respond,” he said. “There is still a huge need for positive predictive biomarkers for the anti-VEGF [vascular endothelial growth factor] therapies and the anti-EGFR therapies, as well as for the cytotoxic therapies and the immune checkpoint inhibitors.”

Emerging biomarkers

The emerging colorectal cancer biomarkers include a variety of cancer genome mutation panels for assessing treatment options in patients with metastatic disease: OncoPlex, Foundation Medicine panels, and SNaPshot.

The clinical utility of these assays at present is still unclear. For example, the OncoPlex assay assesses tumors for 194 genes, only a subset of which are currently actionable or may be in the near future.

“Where you go with this is the real challenge,” Dr. Grady said. He referred clinicians to My Cancer Genome, a website managed by the Vanderbilt-Ingram Cancer Center that provides information about the specific mutations found with these panels.

The site additionally offers a concierge-like service called DIRECT (DNA-mutation Inventory to Refine and Enhance Cancer Treatment) that may be helpful in individual cases. “That [service] will identify what the response for tumors that have specific mutations has been in previous clinical trials and any clinical trials that are ongoing for which your patient may be a candidate,” he explained.

Dr. Grady disclosed that he is a patent holder of methylated MLH1 and receives consulting fees from Myriad Genetics.

SEATTLE – Research in colorectal cancer should focus on part on a handful of high-priority needs for prognostic and predictive biomarkers, according to Dr. William M. Grady, an oncologist at the Fred Hutchinson Cancer Research Center and section chief of gastroenterology at the University of Washington, Seattle.

“These needs are in the selection of optimal agents for the stage II and III cancers, the ones for which we add adjuvant therapy,” he told attendees of a joint meeting by the Global Biomarkers Consortium and the World Cutaneous Malignancies Congress. Additionally, “in the clinic, the main issue is selecting stage II patients who are at high risk for recurrence who may benefit from adjuvant therapy, and then stage III patients who will not benefit from therapy.”

There are some high-priority needs in the metastatic setting as well. “For stage IV disease, really the issue is selecting candidates for conventional chemotherapy, and here we are trying to balance toxicity versus survival and quality of life. Also, there is the issue of identifying people who will be optimal responders to some of the targeted therapies,” Dr. Grady elaborated.

The quest for biomarkers has been aided by increasing recognition of the molecular heterogeneity of colorectal cancer. Research has thus far identified three predominant molecular groups – a microsatellite unstable group, a microsatellite stable group, and a CIMP (CpG island methylator phenotype) group – having alterations of different signaling pathways. “This is clearly an oversimplification, and where I think we are going is that we will continue to parse out hopefully finer and finer groups that will be used for directing clinical care based on the response to agents that are being developed,” he said.

Microsatellite instability

Microsatellite instability is already used clinically to screen colorectal cancers in patients younger than 50 for hereditary Lynch syndrome, as recommended in national guidelines, according to Dr. Grady.

It can also be used to assess the prognosis of stage II and III disease, as patients with tumors showing microsatellite instability have better survival (J Clin Oncol. 2005;23[3]:609-18). But studies have been mixed on whether this biomarker is predictive for the response to adjuvant 5-fluorouracil–based chemotherapy (Nat Rev Clin Oncol. 2010;7[4]:197-208).

“Right now, the consensus is that for stage II disease, if there is microsatellite instability, that correlates with a very low-risk group and there doesn’t appear to be any benefit from adding adjuvant therapy; in fact, one study even suggested harm,” he noted. “For stage III disease [with microsatellite instability], there is no change … indicated at this time for 5-[fluorouracil]-based therapy; potentially, some people even argue that you might want to avoid this therapy, although I don’t think that’s the general opinion. And there may be a benefit in these tumors from the use of oxaliplatin, which one study last year suggested.”

In patients with stage IV disease, recent data suggest microsatellite instability may be a marker for benefit from therapies that target the programmed death–1 (PD1) pathway such as pembrolizumab (Keytruda) (N Engl J Med. 2015;372[26]:2509-20). “The challenge here is that only about 2% to 3% of metastatic colon cancers have microsatellite instability, so unfortunately it’s a small group that will benefit,” Dr. Grady commented.

Multigene signatures

A pair of clinically available multigene signatures, Oncotype DX–Colon and ColoPrint, are being used to estimate prognosis in patients with early disease. However, differences between adjacent assay-defined risk strata are fairly small, on the order of an absolute 10% or so. “This raises a question about what the clinical utility of these scores is,” he said.

“The real question we want to know, though, is can we use [these signatures] to choose people who will get adjuvant therapy for stage II disease,” he commented. At present, the answer appears to be a qualified yes for Oncotype DX, with some data suggesting that the survival benefit of chemotherapy may be greatest in patients with early disease who fall into the assay-defined high-risk category (J Clin Oncol. 2013;31[36]:4512-9)

“Where the ColoPrint assay seems to potentially have the highest benefit is in identifying a low- and high-risk group in the patients who have T3N0 disease but that’s microsatellite-stable. It doesn’t add much to those who have tumors that are already identified as being microsatellite-unstable or who have large [T4N0] disease that is microsatellite-stable,” he noted (Oncologist. 2015;20[2]:127-33).

Ultimately, use of Oncotype DX and ColoPrint should be individualized, according to Dr. Grady. “They certainly can be used, but they really need to be discussed on a case-by-case basis with the patient and physician because they have unclear clinical utility,” he said.

CIMP biomarker

Research suggests that CIMP provides prognostic information in patients with colorectal cancer, according to Dr. Grady. But it also may predict benefit from the addition of irinotecan to 5-fluorouracil and leucovorin in patients with stage III disease having intact mismatch repair (Gastroenterology. 2014;147[3]:637-45).

“CIMP is a potentially emerging marker for the use of cytotoxic agents, probably the most promising one today,” he commented.

Extended RAS assay

The extended RAS assay, which evaluates the whole exome for mutations in both KRAS and NRAS, is useful for guiding decisions about therapies that target signaling through the epidermal growth factor receptor (EGFR), such as panitumumab (Vectibix). However, its value is mainly limited to identifying patients who will not respond to this therapy, according to Dr. Grady.

By incorporating NRAS, this combination assay picks up an additional 10% of patients who will not respond to anti-EGFR therapy (J Clin Oncol. 2015;33[7]:682-5). But even so, it identifies only about a third of all nonresponders.

“We really need better biomarkers to identify patients who will respond, as well as identifying those patients who will not respond,” he said. “There is still a huge need for positive predictive biomarkers for the anti-VEGF [vascular endothelial growth factor] therapies and the anti-EGFR therapies, as well as for the cytotoxic therapies and the immune checkpoint inhibitors.”

Emerging biomarkers

The emerging colorectal cancer biomarkers include a variety of cancer genome mutation panels for assessing treatment options in patients with metastatic disease: OncoPlex, Foundation Medicine panels, and SNaPshot.

The clinical utility of these assays at present is still unclear. For example, the OncoPlex assay assesses tumors for 194 genes, only a subset of which are currently actionable or may be in the near future.

“Where you go with this is the real challenge,” Dr. Grady said. He referred clinicians to My Cancer Genome, a website managed by the Vanderbilt-Ingram Cancer Center that provides information about the specific mutations found with these panels.

The site additionally offers a concierge-like service called DIRECT (DNA-mutation Inventory to Refine and Enhance Cancer Treatment) that may be helpful in individual cases. “That [service] will identify what the response for tumors that have specific mutations has been in previous clinical trials and any clinical trials that are ongoing for which your patient may be a candidate,” he explained.

Dr. Grady disclosed that he is a patent holder of methylated MLH1 and receives consulting fees from Myriad Genetics.

SEATTLE – Research in colorectal cancer should focus on part on a handful of high-priority needs for prognostic and predictive biomarkers, according to Dr. William M. Grady, an oncologist at the Fred Hutchinson Cancer Research Center and section chief of gastroenterology at the University of Washington, Seattle.

“These needs are in the selection of optimal agents for the stage II and III cancers, the ones for which we add adjuvant therapy,” he told attendees of a joint meeting by the Global Biomarkers Consortium and the World Cutaneous Malignancies Congress. Additionally, “in the clinic, the main issue is selecting stage II patients who are at high risk for recurrence who may benefit from adjuvant therapy, and then stage III patients who will not benefit from therapy.”

There are some high-priority needs in the metastatic setting as well. “For stage IV disease, really the issue is selecting candidates for conventional chemotherapy, and here we are trying to balance toxicity versus survival and quality of life. Also, there is the issue of identifying people who will be optimal responders to some of the targeted therapies,” Dr. Grady elaborated.

The quest for biomarkers has been aided by increasing recognition of the molecular heterogeneity of colorectal cancer. Research has thus far identified three predominant molecular groups – a microsatellite unstable group, a microsatellite stable group, and a CIMP (CpG island methylator phenotype) group – having alterations of different signaling pathways. “This is clearly an oversimplification, and where I think we are going is that we will continue to parse out hopefully finer and finer groups that will be used for directing clinical care based on the response to agents that are being developed,” he said.

Microsatellite instability

Microsatellite instability is already used clinically to screen colorectal cancers in patients younger than 50 for hereditary Lynch syndrome, as recommended in national guidelines, according to Dr. Grady.

It can also be used to assess the prognosis of stage II and III disease, as patients with tumors showing microsatellite instability have better survival (J Clin Oncol. 2005;23[3]:609-18). But studies have been mixed on whether this biomarker is predictive for the response to adjuvant 5-fluorouracil–based chemotherapy (Nat Rev Clin Oncol. 2010;7[4]:197-208).

“Right now, the consensus is that for stage II disease, if there is microsatellite instability, that correlates with a very low-risk group and there doesn’t appear to be any benefit from adding adjuvant therapy; in fact, one study even suggested harm,” he noted. “For stage III disease [with microsatellite instability], there is no change … indicated at this time for 5-[fluorouracil]-based therapy; potentially, some people even argue that you might want to avoid this therapy, although I don’t think that’s the general opinion. And there may be a benefit in these tumors from the use of oxaliplatin, which one study last year suggested.”

In patients with stage IV disease, recent data suggest microsatellite instability may be a marker for benefit from therapies that target the programmed death–1 (PD1) pathway such as pembrolizumab (Keytruda) (N Engl J Med. 2015;372[26]:2509-20). “The challenge here is that only about 2% to 3% of metastatic colon cancers have microsatellite instability, so unfortunately it’s a small group that will benefit,” Dr. Grady commented.

Multigene signatures

A pair of clinically available multigene signatures, Oncotype DX–Colon and ColoPrint, are being used to estimate prognosis in patients with early disease. However, differences between adjacent assay-defined risk strata are fairly small, on the order of an absolute 10% or so. “This raises a question about what the clinical utility of these scores is,” he said.

“The real question we want to know, though, is can we use [these signatures] to choose people who will get adjuvant therapy for stage II disease,” he commented. At present, the answer appears to be a qualified yes for Oncotype DX, with some data suggesting that the survival benefit of chemotherapy may be greatest in patients with early disease who fall into the assay-defined high-risk category (J Clin Oncol. 2013;31[36]:4512-9)

“Where the ColoPrint assay seems to potentially have the highest benefit is in identifying a low- and high-risk group in the patients who have T3N0 disease but that’s microsatellite-stable. It doesn’t add much to those who have tumors that are already identified as being microsatellite-unstable or who have large [T4N0] disease that is microsatellite-stable,” he noted (Oncologist. 2015;20[2]:127-33).

Ultimately, use of Oncotype DX and ColoPrint should be individualized, according to Dr. Grady. “They certainly can be used, but they really need to be discussed on a case-by-case basis with the patient and physician because they have unclear clinical utility,” he said.

CIMP biomarker

Research suggests that CIMP provides prognostic information in patients with colorectal cancer, according to Dr. Grady. But it also may predict benefit from the addition of irinotecan to 5-fluorouracil and leucovorin in patients with stage III disease having intact mismatch repair (Gastroenterology. 2014;147[3]:637-45).

“CIMP is a potentially emerging marker for the use of cytotoxic agents, probably the most promising one today,” he commented.

Extended RAS assay

The extended RAS assay, which evaluates the whole exome for mutations in both KRAS and NRAS, is useful for guiding decisions about therapies that target signaling through the epidermal growth factor receptor (EGFR), such as panitumumab (Vectibix). However, its value is mainly limited to identifying patients who will not respond to this therapy, according to Dr. Grady.

By incorporating NRAS, this combination assay picks up an additional 10% of patients who will not respond to anti-EGFR therapy (J Clin Oncol. 2015;33[7]:682-5). But even so, it identifies only about a third of all nonresponders.

“We really need better biomarkers to identify patients who will respond, as well as identifying those patients who will not respond,” he said. “There is still a huge need for positive predictive biomarkers for the anti-VEGF [vascular endothelial growth factor] therapies and the anti-EGFR therapies, as well as for the cytotoxic therapies and the immune checkpoint inhibitors.”

Emerging biomarkers

The emerging colorectal cancer biomarkers include a variety of cancer genome mutation panels for assessing treatment options in patients with metastatic disease: OncoPlex, Foundation Medicine panels, and SNaPshot.

The clinical utility of these assays at present is still unclear. For example, the OncoPlex assay assesses tumors for 194 genes, only a subset of which are currently actionable or may be in the near future.

“Where you go with this is the real challenge,” Dr. Grady said. He referred clinicians to My Cancer Genome, a website managed by the Vanderbilt-Ingram Cancer Center that provides information about the specific mutations found with these panels.

The site additionally offers a concierge-like service called DIRECT (DNA-mutation Inventory to Refine and Enhance Cancer Treatment) that may be helpful in individual cases. “That [service] will identify what the response for tumors that have specific mutations has been in previous clinical trials and any clinical trials that are ongoing for which your patient may be a candidate,” he explained.

Dr. Grady disclosed that he is a patent holder of methylated MLH1 and receives consulting fees from Myriad Genetics.

SEATTLE – Many advanced non–small cell lung cancer (NSCLC) adenocarcinomas can now be managed with therapies that target driving mutations, but oncologists must look for these mutations and be aware that they can change over time, Dr. Mark A. Socinski said at a joint meeting by Global Biomarkers Consortium and World Cutaneous Malignancies Congress.

The 2013 College of American Pathology guideline for the molecular testing of lung cancer “was a monumental publication and a beachhead, if you will, for establishing a standard of care [for NSCLC], much like we have in breast cancer for ER, PR, and HER2 measurement,” he said. Furthermore, “we are now in an era where doing subsequent or sequential biopsies with repeat molecular testing is a standard of care in this population.”

Although lung adenocarcinomas are homogeneous histologically, they are diverse with respect to oncogenic drivers (JAMA 2014;311[19]:1998-2006), noted Dr. Socinski, director of the lung cancer section at the University of Pittsburgh Medical Center; clinical associate director of the University of Pittsburgh Lung Cancer SPORE; codirector of the UPMC Lung Cancer Center of Excellence; and coleader of the lung cancer program at the University of Pittsburgh.

“Our major nemesis is KRAS. We still don’t have a good answer for that,” he said. But roughly a third of lung adenocarcinomas have actionable mutations in the genes for EGFR [epidermal growth factor receptor], ALK, ROS1, BRAF, MET, or RET.

“In the year 2015, these are what I look for in our patient population … We test routinely to identify these populations,” he said. “In my clinic this week, I might have had almost all of these patients on targeted TKIs [tyrosine kinase inhibitors] with these sorts of things, getting clinical benefit in this particular setting.”

Common mutations

“The EGFR mutation story really transformed lung cancer,” Dr. Socinski said. In patients whose adenocarcinomas harbor these mutations, targeted therapy with an EGFR inhibitor commonly nets a dramatic response. “If you see this a number of times and you’re a lung cancer doc, you become addicted to oncotype testing. And you certainly don’t want to ever miss this,” he said.

The IPASS trial (First-Line Iressa Versus Carboplatin/Paclitaxel in Asia) comparing the targeted agent gefitinib (Iressa) with chemotherapy in advanced NSCLC adenocarcinoma among never or light smokers was “a transformational trial in lung cancer,” according to Dr. Socinski (N Engl J Med. 2009;361[10]:947-57).

“The lesson from IPASS: phenotype we threw out the door, it’s really about genotype. And if you didn’t have the genotype [EGFR mutation], a TKI was very poor treatment. And if you had the genotype, the TKI was superior to chemotherapy,” with a 52% reduction in the risk of progression or death.

Trials testing the EGFR inhibitors erlotinib (Tarceva) and afatinib (Gilotrif) have likewise shown a progression-free survival benefit in this patient population.

“One of the issues that we struggled with for some time was whether there is any survival benefit,” Dr. Socinski said. A recent combined analysis of two afatinib trials has answered that question affirmatively (Lancet Oncol. 2015;16[2]:141-151), and these agents have therefore become standard of care for EGFR-mutant adenocarcinoma.

“Interestingly, as we say, all EGFR mutants are not created equal, because in the exon 21[–mutated tumors], actually there was no difference relative to chemotherapy, and that survival advantage is really driven by exon 19. So the nature of your mutation is important in this particular analysis,” he cautioned.

When patients on EGFR targeted therapy develop resistance, the cause in about half of cases is emergence of a secondary mutation in exon 20, the T790M mutation (Sci Transl Med. 2011;3(75):75ra26).

“The standard of care is to biopsy at the time of progression,” Dr. Socinski maintained. “The reason why rebiopsy is important and it’s important to diagnose that [new mutation] is that we have a couple of drugs close to [Food and Drug Administration] approval that are highly active in patients with T790M-positive disease after a first- or second-generation TKI.”

Specifically, the investigational third-generation TKIs rociletinib (ASCO 2015. Abstract 8001) and AZD9291 (ASCO 2014. Abstract 8009) have response rates of about 48% and 53%, respectively, in this setting. “This looks quite promising. And these drugs will likely be commercially available between now and the holidays at the end of the year,” he predicted.

The T790M mutation can appear at different times, he said. “I’ve even got several patients whom we’ve rebiopsied three or four times, and there has been T790M negativity and then emergence of positivity on subsequent biopsy. Given the activity of these drugs, that’s important to know.”

Another fairly common actionable mutation in lung adenocarcinoma is ALK, for which oncologists now have crizotinib (Xalkori). Crizotinib has likewise been tested against combination chemotherapy in a phase III trial in which it yielded superior progression-free survival in patients with advanced nonsquamous NSCLC harboring ALK mutations (ASCO 2014. Abstract 8002).

“This is now a second example with a molecular biomarker in which we’ve replaced the standard of care chemotherapy with a molecularly targeted agent,” Dr. Socinski noted.

Second-generation ALK inhibitors such as the investigational agent alectinib are showing promise (ASCO 2015. Abstract 8008). “Even in previously crizotinib-exposed patients, these have a great deal of activity and allow another option for sequential therapy in this population of patients,” he said.

Uncommon mutations

Driving mutations of ROS1 are found in about 1%-2% of lung cancers, most often in younger never-smokers with adenocarcinomas, according to Dr. Socinski.

These tumors respond to crizotinib, which is also a ROS1 inhibitor. “In fact I think it may actually be a better ROS1 drug than an ALK drug,” he said.

The drug yields an impressive median progression-free survival of 19.2 months and overall response rate of 72% in this setting (N Engl J Med. 2014;371[21]:1963-1971), “so ROS1 is another biomarker that we go hunting for in this population, even though you won’t see it very commonly.”

Mutations of BRAF are found in about 2% of metastatic adenocarcinomas (Cancer. 2015;121[3]:448-456). The large majority, about fourth-fifths, are of the V600E type.

The BRAF inhibitor dabrafenib (Tafinlar) has been associated with an overall response rate of 32% in patients with this specific mutation (abstract LBA38, Ann Oncol. 2014;25[Suppl 4]. doi:10.1093/annonc/mdu438.46). And preliminary data suggest efficacy increases when it is combined with the Mek inhibitor trametinib (Mekinist) (ASCO 2015. Abstract 8006), as has been seen in melanoma.

About 4% of lung cancers have an intermediate or high level of MET amplification. In a small sample of patients with these tumors, treatment with crizotinib appeared to be active (ASCO 2014. Abstract 8001). In addition, this agent has efficacy against lung cancers having exon 14 splice mutations in MET (ASCO 2015. Abstract 8021). “So this is another genotype not to miss,” Dr. Socinski said.

Finally, mutation of RET is seen about 1%-2% of unselected NSCLCs, also typically in young never-smokers or former smokers with adenocarcinoma. Cabozantinib (Cometriq), a multitargeted TKI having activity against RET, yields a 28% response rate in RET-rearranged adenocarcinomas (ASCO 2015. Abstract 8007).

A controversial topic for these uncommon mutations in lung adenocarcinomas is how much evidence should be required for new targeted agents to gain FDA approval, Dr. Socinski said.

“For instance, the ROS1 experience: Do we really need a randomized trial in a rare genotype to approve this drug [crizotinib] for ROS1-positive patients? I would say, absolutely not,” he concluded.

Dr. Socinski disclosed that he receives fees from Celgene and Genentech, and performs contracted research for Celgene, Clovis, Genentech, GlaxoSmithKline, Pfizer, and Synta.

SEATTLE – Many advanced non–small cell lung cancer (NSCLC) adenocarcinomas can now be managed with therapies that target driving mutations, but oncologists must look for these mutations and be aware that they can change over time, Dr. Mark A. Socinski said at a joint meeting by Global Biomarkers Consortium and World Cutaneous Malignancies Congress.

The 2013 College of American Pathology guideline for the molecular testing of lung cancer “was a monumental publication and a beachhead, if you will, for establishing a standard of care [for NSCLC], much like we have in breast cancer for ER, PR, and HER2 measurement,” he said. Furthermore, “we are now in an era where doing subsequent or sequential biopsies with repeat molecular testing is a standard of care in this population.”

Although lung adenocarcinomas are homogeneous histologically, they are diverse with respect to oncogenic drivers (JAMA 2014;311[19]:1998-2006), noted Dr. Socinski, director of the lung cancer section at the University of Pittsburgh Medical Center; clinical associate director of the University of Pittsburgh Lung Cancer SPORE; codirector of the UPMC Lung Cancer Center of Excellence; and coleader of the lung cancer program at the University of Pittsburgh.

“Our major nemesis is KRAS. We still don’t have a good answer for that,” he said. But roughly a third of lung adenocarcinomas have actionable mutations in the genes for EGFR [epidermal growth factor receptor], ALK, ROS1, BRAF, MET, or RET.

“In the year 2015, these are what I look for in our patient population … We test routinely to identify these populations,” he said. “In my clinic this week, I might have had almost all of these patients on targeted TKIs [tyrosine kinase inhibitors] with these sorts of things, getting clinical benefit in this particular setting.”

Common mutations

“The EGFR mutation story really transformed lung cancer,” Dr. Socinski said. In patients whose adenocarcinomas harbor these mutations, targeted therapy with an EGFR inhibitor commonly nets a dramatic response. “If you see this a number of times and you’re a lung cancer doc, you become addicted to oncotype testing. And you certainly don’t want to ever miss this,” he said.

The IPASS trial (First-Line Iressa Versus Carboplatin/Paclitaxel in Asia) comparing the targeted agent gefitinib (Iressa) with chemotherapy in advanced NSCLC adenocarcinoma among never or light smokers was “a transformational trial in lung cancer,” according to Dr. Socinski (N Engl J Med. 2009;361[10]:947-57).

“The lesson from IPASS: phenotype we threw out the door, it’s really about genotype. And if you didn’t have the genotype [EGFR mutation], a TKI was very poor treatment. And if you had the genotype, the TKI was superior to chemotherapy,” with a 52% reduction in the risk of progression or death.

Trials testing the EGFR inhibitors erlotinib (Tarceva) and afatinib (Gilotrif) have likewise shown a progression-free survival benefit in this patient population.

“One of the issues that we struggled with for some time was whether there is any survival benefit,” Dr. Socinski said. A recent combined analysis of two afatinib trials has answered that question affirmatively (Lancet Oncol. 2015;16[2]:141-151), and these agents have therefore become standard of care for EGFR-mutant adenocarcinoma.

“Interestingly, as we say, all EGFR mutants are not created equal, because in the exon 21[–mutated tumors], actually there was no difference relative to chemotherapy, and that survival advantage is really driven by exon 19. So the nature of your mutation is important in this particular analysis,” he cautioned.

When patients on EGFR targeted therapy develop resistance, the cause in about half of cases is emergence of a secondary mutation in exon 20, the T790M mutation (Sci Transl Med. 2011;3(75):75ra26).

“The standard of care is to biopsy at the time of progression,” Dr. Socinski maintained. “The reason why rebiopsy is important and it’s important to diagnose that [new mutation] is that we have a couple of drugs close to [Food and Drug Administration] approval that are highly active in patients with T790M-positive disease after a first- or second-generation TKI.”

Specifically, the investigational third-generation TKIs rociletinib (ASCO 2015. Abstract 8001) and AZD9291 (ASCO 2014. Abstract 8009) have response rates of about 48% and 53%, respectively, in this setting. “This looks quite promising. And these drugs will likely be commercially available between now and the holidays at the end of the year,” he predicted.

The T790M mutation can appear at different times, he said. “I’ve even got several patients whom we’ve rebiopsied three or four times, and there has been T790M negativity and then emergence of positivity on subsequent biopsy. Given the activity of these drugs, that’s important to know.”

Another fairly common actionable mutation in lung adenocarcinoma is ALK, for which oncologists now have crizotinib (Xalkori). Crizotinib has likewise been tested against combination chemotherapy in a phase III trial in which it yielded superior progression-free survival in patients with advanced nonsquamous NSCLC harboring ALK mutations (ASCO 2014. Abstract 8002).

“This is now a second example with a molecular biomarker in which we’ve replaced the standard of care chemotherapy with a molecularly targeted agent,” Dr. Socinski noted.

Second-generation ALK inhibitors such as the investigational agent alectinib are showing promise (ASCO 2015. Abstract 8008). “Even in previously crizotinib-exposed patients, these have a great deal of activity and allow another option for sequential therapy in this population of patients,” he said.

Uncommon mutations

Driving mutations of ROS1 are found in about 1%-2% of lung cancers, most often in younger never-smokers with adenocarcinomas, according to Dr. Socinski.

These tumors respond to crizotinib, which is also a ROS1 inhibitor. “In fact I think it may actually be a better ROS1 drug than an ALK drug,” he said.

The drug yields an impressive median progression-free survival of 19.2 months and overall response rate of 72% in this setting (N Engl J Med. 2014;371[21]:1963-1971), “so ROS1 is another biomarker that we go hunting for in this population, even though you won’t see it very commonly.”

Mutations of BRAF are found in about 2% of metastatic adenocarcinomas (Cancer. 2015;121[3]:448-456). The large majority, about fourth-fifths, are of the V600E type.

The BRAF inhibitor dabrafenib (Tafinlar) has been associated with an overall response rate of 32% in patients with this specific mutation (abstract LBA38, Ann Oncol. 2014;25[Suppl 4]. doi:10.1093/annonc/mdu438.46). And preliminary data suggest efficacy increases when it is combined with the Mek inhibitor trametinib (Mekinist) (ASCO 2015. Abstract 8006), as has been seen in melanoma.

About 4% of lung cancers have an intermediate or high level of MET amplification. In a small sample of patients with these tumors, treatment with crizotinib appeared to be active (ASCO 2014. Abstract 8001). In addition, this agent has efficacy against lung cancers having exon 14 splice mutations in MET (ASCO 2015. Abstract 8021). “So this is another genotype not to miss,” Dr. Socinski said.

Finally, mutation of RET is seen about 1%-2% of unselected NSCLCs, also typically in young never-smokers or former smokers with adenocarcinoma. Cabozantinib (Cometriq), a multitargeted TKI having activity against RET, yields a 28% response rate in RET-rearranged adenocarcinomas (ASCO 2015. Abstract 8007).

A controversial topic for these uncommon mutations in lung adenocarcinomas is how much evidence should be required for new targeted agents to gain FDA approval, Dr. Socinski said.

“For instance, the ROS1 experience: Do we really need a randomized trial in a rare genotype to approve this drug [crizotinib] for ROS1-positive patients? I would say, absolutely not,” he concluded.

Dr. Socinski disclosed that he receives fees from Celgene and Genentech, and performs contracted research for Celgene, Clovis, Genentech, GlaxoSmithKline, Pfizer, and Synta.

SEATTLE – Many advanced non–small cell lung cancer (NSCLC) adenocarcinomas can now be managed with therapies that target driving mutations, but oncologists must look for these mutations and be aware that they can change over time, Dr. Mark A. Socinski said at a joint meeting by Global Biomarkers Consortium and World Cutaneous Malignancies Congress.

The 2013 College of American Pathology guideline for the molecular testing of lung cancer “was a monumental publication and a beachhead, if you will, for establishing a standard of care [for NSCLC], much like we have in breast cancer for ER, PR, and HER2 measurement,” he said. Furthermore, “we are now in an era where doing subsequent or sequential biopsies with repeat molecular testing is a standard of care in this population.”

Although lung adenocarcinomas are homogeneous histologically, they are diverse with respect to oncogenic drivers (JAMA 2014;311[19]:1998-2006), noted Dr. Socinski, director of the lung cancer section at the University of Pittsburgh Medical Center; clinical associate director of the University of Pittsburgh Lung Cancer SPORE; codirector of the UPMC Lung Cancer Center of Excellence; and coleader of the lung cancer program at the University of Pittsburgh.

“Our major nemesis is KRAS. We still don’t have a good answer for that,” he said. But roughly a third of lung adenocarcinomas have actionable mutations in the genes for EGFR [epidermal growth factor receptor], ALK, ROS1, BRAF, MET, or RET.

“In the year 2015, these are what I look for in our patient population … We test routinely to identify these populations,” he said. “In my clinic this week, I might have had almost all of these patients on targeted TKIs [tyrosine kinase inhibitors] with these sorts of things, getting clinical benefit in this particular setting.”

Common mutations

“The EGFR mutation story really transformed lung cancer,” Dr. Socinski said. In patients whose adenocarcinomas harbor these mutations, targeted therapy with an EGFR inhibitor commonly nets a dramatic response. “If you see this a number of times and you’re a lung cancer doc, you become addicted to oncotype testing. And you certainly don’t want to ever miss this,” he said.

The IPASS trial (First-Line Iressa Versus Carboplatin/Paclitaxel in Asia) comparing the targeted agent gefitinib (Iressa) with chemotherapy in advanced NSCLC adenocarcinoma among never or light smokers was “a transformational trial in lung cancer,” according to Dr. Socinski (N Engl J Med. 2009;361[10]:947-57).

“The lesson from IPASS: phenotype we threw out the door, it’s really about genotype. And if you didn’t have the genotype [EGFR mutation], a TKI was very poor treatment. And if you had the genotype, the TKI was superior to chemotherapy,” with a 52% reduction in the risk of progression or death.

Trials testing the EGFR inhibitors erlotinib (Tarceva) and afatinib (Gilotrif) have likewise shown a progression-free survival benefit in this patient population.

“One of the issues that we struggled with for some time was whether there is any survival benefit,” Dr. Socinski said. A recent combined analysis of two afatinib trials has answered that question affirmatively (Lancet Oncol. 2015;16[2]:141-151), and these agents have therefore become standard of care for EGFR-mutant adenocarcinoma.

“Interestingly, as we say, all EGFR mutants are not created equal, because in the exon 21[–mutated tumors], actually there was no difference relative to chemotherapy, and that survival advantage is really driven by exon 19. So the nature of your mutation is important in this particular analysis,” he cautioned.

When patients on EGFR targeted therapy develop resistance, the cause in about half of cases is emergence of a secondary mutation in exon 20, the T790M mutation (Sci Transl Med. 2011;3(75):75ra26).

“The standard of care is to biopsy at the time of progression,” Dr. Socinski maintained. “The reason why rebiopsy is important and it’s important to diagnose that [new mutation] is that we have a couple of drugs close to [Food and Drug Administration] approval that are highly active in patients with T790M-positive disease after a first- or second-generation TKI.”

Specifically, the investigational third-generation TKIs rociletinib (ASCO 2015. Abstract 8001) and AZD9291 (ASCO 2014. Abstract 8009) have response rates of about 48% and 53%, respectively, in this setting. “This looks quite promising. And these drugs will likely be commercially available between now and the holidays at the end of the year,” he predicted.

The T790M mutation can appear at different times, he said. “I’ve even got several patients whom we’ve rebiopsied three or four times, and there has been T790M negativity and then emergence of positivity on subsequent biopsy. Given the activity of these drugs, that’s important to know.”

Another fairly common actionable mutation in lung adenocarcinoma is ALK, for which oncologists now have crizotinib (Xalkori). Crizotinib has likewise been tested against combination chemotherapy in a phase III trial in which it yielded superior progression-free survival in patients with advanced nonsquamous NSCLC harboring ALK mutations (ASCO 2014. Abstract 8002).

“This is now a second example with a molecular biomarker in which we’ve replaced the standard of care chemotherapy with a molecularly targeted agent,” Dr. Socinski noted.

Second-generation ALK inhibitors such as the investigational agent alectinib are showing promise (ASCO 2015. Abstract 8008). “Even in previously crizotinib-exposed patients, these have a great deal of activity and allow another option for sequential therapy in this population of patients,” he said.

Uncommon mutations

Driving mutations of ROS1 are found in about 1%-2% of lung cancers, most often in younger never-smokers with adenocarcinomas, according to Dr. Socinski.

These tumors respond to crizotinib, which is also a ROS1 inhibitor. “In fact I think it may actually be a better ROS1 drug than an ALK drug,” he said.

The drug yields an impressive median progression-free survival of 19.2 months and overall response rate of 72% in this setting (N Engl J Med. 2014;371[21]:1963-1971), “so ROS1 is another biomarker that we go hunting for in this population, even though you won’t see it very commonly.”

Mutations of BRAF are found in about 2% of metastatic adenocarcinomas (Cancer. 2015;121[3]:448-456). The large majority, about fourth-fifths, are of the V600E type.

The BRAF inhibitor dabrafenib (Tafinlar) has been associated with an overall response rate of 32% in patients with this specific mutation (abstract LBA38, Ann Oncol. 2014;25[Suppl 4]. doi:10.1093/annonc/mdu438.46). And preliminary data suggest efficacy increases when it is combined with the Mek inhibitor trametinib (Mekinist) (ASCO 2015. Abstract 8006), as has been seen in melanoma.

About 4% of lung cancers have an intermediate or high level of MET amplification. In a small sample of patients with these tumors, treatment with crizotinib appeared to be active (ASCO 2014. Abstract 8001). In addition, this agent has efficacy against lung cancers having exon 14 splice mutations in MET (ASCO 2015. Abstract 8021). “So this is another genotype not to miss,” Dr. Socinski said.

Finally, mutation of RET is seen about 1%-2% of unselected NSCLCs, also typically in young never-smokers or former smokers with adenocarcinoma. Cabozantinib (Cometriq), a multitargeted TKI having activity against RET, yields a 28% response rate in RET-rearranged adenocarcinomas (ASCO 2015. Abstract 8007).

A controversial topic for these uncommon mutations in lung adenocarcinomas is how much evidence should be required for new targeted agents to gain FDA approval, Dr. Socinski said.

“For instance, the ROS1 experience: Do we really need a randomized trial in a rare genotype to approve this drug [crizotinib] for ROS1-positive patients? I would say, absolutely not,” he concluded.

Dr. Socinski disclosed that he receives fees from Celgene and Genentech, and performs contracted research for Celgene, Clovis, Genentech, GlaxoSmithKline, Pfizer, and Synta.

SEATTLE – Biomarkers may soon join other modalities for the early detection of breast cancer and identification of women at high risk for the disease, Susan L. Neuhausen, Ph.D., told attendees of the Global Biomarkers Consortium.

“There is an urgent need for biomarkers for breast cancer,” she maintained, citing its high prevalence and considerable mortality, coupled with better outcomes and lesser treatment requirements when it is caught early.

“I really think that for a lot of these biomarkers [in development], maybe their best use will be to augment mammographic screening, especially in the short term as they are being developed,” she speculated. Ultimately, “the hope – similar to PSA [prostate-specific antigen], where you can use it to detect cancer as well as use it as a marker of recurrence – is that these biomarkers of early detection will be developed to have the same attributes.”

Genetic biomarkers

“To me, the current best biomarker is actually a genetic biomarker,” said Dr. Neuhausen, who is the Morris and Horowitz Families Professor in Cancer Etiology and Outcomes Research, Population Sciences, and also coleader of the cancer control and population sciences program, Comprehensive Cancer Center, at the Beckman Research Institute, City of Hope, in Duarte, Calif.

Identifying the breast cancer susceptibility genes BRCA 1 and BRCA 2 paved the way for targeted chemoprevention and prophylactic surgery, which have been highly effective in reducing the incidence of breast cancer as well as ovarian cancer. However, mutations in these genes account for no more than approximately 5% of all breast cancers.

“There has been identification of additional moderate- to high-risk genes, and the strategies that work to prevent cancer in BRCA 1 and 2 carriers are the same that can used for these other high-risk genes,” Dr. Neuhausen noted. For example, mutations of partner and localizer of BRCA2 (PALB2) have been linked to heightened breast cancer risk (N Engl J Med. 2014;371:497-506). Additionally, many companies now offer multigene clinical risk panels.

“The problem or the issue at the current time is that we really don’t know what the risk of cancer in unaffected women who are carrying mutations in these genes is,” she commented. “The good news is there is a large European, United States, and Canadian study that is actually going to be screening for mutations in a total of about 25,000 women, so that similar to what’s been done in BRCA 1 and 2, we will actually have better risk estimates that one can then use for these women.”

A U.S. genome-wide association research effort, the Breast Cancer Surveillance Consortium (BCSC), has thus far identified about 100 loci linked to an increased risk of developing breast cancer, according to Dr. Neuhausen. “On an individual level, these really don’t account for much risk. However, if you combine them, they actually do,” she said.

Applied clinically, a polygenic risk score incorporating 77 single-nucleotide polymorphisms can stratify women into quintiles of risk (J Natl Cancer Inst. 2015;107(5):djv036). Among women without a family history of the disease, lifetime risk ranges from 5.2% for those in the lowest quintile to 16.6% for those in the highest.

Moreover, adding the polygenic risk score to the commonly used BCSC model increases the area under the curve from 0.66 to 0.69 for breast cancer risk prediction. It also results in reclassification of some women, in particular, identifying an additional 3% as having a greater than 3% risk of developing the disease over 5 years. “So this actually does have good discrimination and is something that can be useful moving forward,” Dr. Neuhausen commented.

“We have these low-, moderate-, and high-risk genes, and they really all need to be incorporated into models. There is research ongoing to try to incorporate all of them into models along with lifestyle factors and mammography, family history, that kind of thing,” she said. “But I really think that these genetic markers are very important because rather than early detection, if we can prevent breast cancer, that’s actually the real goal.”

Other biomarkers

A variety of other, nongenetic biomarkers for early breast cancer detection and risk stratification are generally less far along in development, but several have shown promise, according to Dr. Neuhausen.

In the realm of proteomics, a three–amino acid profile detectable in saliva that exploits the hypermetabolic and hypercatabolic nature of cancer was found to have an area under the curve of 0.916 for differentiating women with early breast cancer from unaffected peers (Clin Chim Acta. 2015;447:23-31). “It was only a very small study and they really need to look further, but I think it was an intriguing idea,” she commented.

Infrared spectroscopy of plasma identified a protein fingerprint that had roughly 90% sensitivity and 80% specificity for detecting breast cancer in a cohort that included women with the disease, women with benign breast conditions, and healthy women (BMC Cancer. 2015;15:408). And a pair of proteins in serum—HSP27 and 14-3-3 sigma—accurately distinguished breast cancer cases from controls in a Chinese population (Proteomics. 2003;3:433-9).

One study has suggested how proteomic biomarkers might be integrated with conventional modalities, showing that a five-protein signature in serum (dtectDx Breast, Provista) performed well among women younger than age 50 years for differentiating benign breast lesions from invasive breast cancer in those with a BI-RADS 3 or 4 mammogram (J Clin Oncol. 2014;32(26 Suppl 20). However, the signature had a fairly high false-positive rate, Dr. Neuhausen noted.

As for yet other types of biomarkers, a panel of circulating cell-free methylated DNA of eight tumor suppressor genes was found to have sensitivity and specificity exceeding 90% for differentiating between samples from breast cancer patients and from unaffected women (PLoS One. 2011;6:e16080).MicroRNA profiles of breast cancer have been identified, but findings have generally been inconsistent across studies (Breast Cancer. 2015;7:59-79). Somatic mutations in p53 and PI3KCA, present in about a third of breast cancers, currently have issues when applied to screening. “Those kinds of things are maybe better for determining response to treatment and that kind of thing. Are they ready for early detection? Not really,” Dr. Neuhausen said.

The situation is similar for DNA copy number, although intriguingly, a recent study of a prenatal test looking for fetal copy number aberrations in maternal plasma incidentally discovered maternal cell-free DNA that had copy number changes (JAMA Oncol. 2015 June 5.doi: 10.1001/jamaoncol.2015.1883). Imaging ultimately found various cancers in three women out of about 4,000 tested.

Dr. Neuhausen disclosed that she had no relevant conflicts of interest.

SEATTLE – Biomarkers may soon join other modalities for the early detection of breast cancer and identification of women at high risk for the disease, Susan L. Neuhausen, Ph.D., told attendees of the Global Biomarkers Consortium.

“There is an urgent need for biomarkers for breast cancer,” she maintained, citing its high prevalence and considerable mortality, coupled with better outcomes and lesser treatment requirements when it is caught early.

“I really think that for a lot of these biomarkers [in development], maybe their best use will be to augment mammographic screening, especially in the short term as they are being developed,” she speculated. Ultimately, “the hope – similar to PSA [prostate-specific antigen], where you can use it to detect cancer as well as use it as a marker of recurrence – is that these biomarkers of early detection will be developed to have the same attributes.”

Genetic biomarkers

“To me, the current best biomarker is actually a genetic biomarker,” said Dr. Neuhausen, who is the Morris and Horowitz Families Professor in Cancer Etiology and Outcomes Research, Population Sciences, and also coleader of the cancer control and population sciences program, Comprehensive Cancer Center, at the Beckman Research Institute, City of Hope, in Duarte, Calif.

Identifying the breast cancer susceptibility genes BRCA 1 and BRCA 2 paved the way for targeted chemoprevention and prophylactic surgery, which have been highly effective in reducing the incidence of breast cancer as well as ovarian cancer. However, mutations in these genes account for no more than approximately 5% of all breast cancers.

“There has been identification of additional moderate- to high-risk genes, and the strategies that work to prevent cancer in BRCA 1 and 2 carriers are the same that can used for these other high-risk genes,” Dr. Neuhausen noted. For example, mutations of partner and localizer of BRCA2 (PALB2) have been linked to heightened breast cancer risk (N Engl J Med. 2014;371:497-506). Additionally, many companies now offer multigene clinical risk panels.

“The problem or the issue at the current time is that we really don’t know what the risk of cancer in unaffected women who are carrying mutations in these genes is,” she commented. “The good news is there is a large European, United States, and Canadian study that is actually going to be screening for mutations in a total of about 25,000 women, so that similar to what’s been done in BRCA 1 and 2, we will actually have better risk estimates that one can then use for these women.”

A U.S. genome-wide association research effort, the Breast Cancer Surveillance Consortium (BCSC), has thus far identified about 100 loci linked to an increased risk of developing breast cancer, according to Dr. Neuhausen. “On an individual level, these really don’t account for much risk. However, if you combine them, they actually do,” she said.

Applied clinically, a polygenic risk score incorporating 77 single-nucleotide polymorphisms can stratify women into quintiles of risk (J Natl Cancer Inst. 2015;107(5):djv036). Among women without a family history of the disease, lifetime risk ranges from 5.2% for those in the lowest quintile to 16.6% for those in the highest.

Moreover, adding the polygenic risk score to the commonly used BCSC model increases the area under the curve from 0.66 to 0.69 for breast cancer risk prediction. It also results in reclassification of some women, in particular, identifying an additional 3% as having a greater than 3% risk of developing the disease over 5 years. “So this actually does have good discrimination and is something that can be useful moving forward,” Dr. Neuhausen commented.

“We have these low-, moderate-, and high-risk genes, and they really all need to be incorporated into models. There is research ongoing to try to incorporate all of them into models along with lifestyle factors and mammography, family history, that kind of thing,” she said. “But I really think that these genetic markers are very important because rather than early detection, if we can prevent breast cancer, that’s actually the real goal.”

Other biomarkers

A variety of other, nongenetic biomarkers for early breast cancer detection and risk stratification are generally less far along in development, but several have shown promise, according to Dr. Neuhausen.

In the realm of proteomics, a three–amino acid profile detectable in saliva that exploits the hypermetabolic and hypercatabolic nature of cancer was found to have an area under the curve of 0.916 for differentiating women with early breast cancer from unaffected peers (Clin Chim Acta. 2015;447:23-31). “It was only a very small study and they really need to look further, but I think it was an intriguing idea,” she commented.

Infrared spectroscopy of plasma identified a protein fingerprint that had roughly 90% sensitivity and 80% specificity for detecting breast cancer in a cohort that included women with the disease, women with benign breast conditions, and healthy women (BMC Cancer. 2015;15:408). And a pair of proteins in serum—HSP27 and 14-3-3 sigma—accurately distinguished breast cancer cases from controls in a Chinese population (Proteomics. 2003;3:433-9).

One study has suggested how proteomic biomarkers might be integrated with conventional modalities, showing that a five-protein signature in serum (dtectDx Breast, Provista) performed well among women younger than age 50 years for differentiating benign breast lesions from invasive breast cancer in those with a BI-RADS 3 or 4 mammogram (J Clin Oncol. 2014;32(26 Suppl 20). However, the signature had a fairly high false-positive rate, Dr. Neuhausen noted.

As for yet other types of biomarkers, a panel of circulating cell-free methylated DNA of eight tumor suppressor genes was found to have sensitivity and specificity exceeding 90% for differentiating between samples from breast cancer patients and from unaffected women (PLoS One. 2011;6:e16080).MicroRNA profiles of breast cancer have been identified, but findings have generally been inconsistent across studies (Breast Cancer. 2015;7:59-79). Somatic mutations in p53 and PI3KCA, present in about a third of breast cancers, currently have issues when applied to screening. “Those kinds of things are maybe better for determining response to treatment and that kind of thing. Are they ready for early detection? Not really,” Dr. Neuhausen said.

The situation is similar for DNA copy number, although intriguingly, a recent study of a prenatal test looking for fetal copy number aberrations in maternal plasma incidentally discovered maternal cell-free DNA that had copy number changes (JAMA Oncol. 2015 June 5.doi: 10.1001/jamaoncol.2015.1883). Imaging ultimately found various cancers in three women out of about 4,000 tested.

Dr. Neuhausen disclosed that she had no relevant conflicts of interest.

SEATTLE – Biomarkers may soon join other modalities for the early detection of breast cancer and identification of women at high risk for the disease, Susan L. Neuhausen, Ph.D., told attendees of the Global Biomarkers Consortium.

“There is an urgent need for biomarkers for breast cancer,” she maintained, citing its high prevalence and considerable mortality, coupled with better outcomes and lesser treatment requirements when it is caught early.

“I really think that for a lot of these biomarkers [in development], maybe their best use will be to augment mammographic screening, especially in the short term as they are being developed,” she speculated. Ultimately, “the hope – similar to PSA [prostate-specific antigen], where you can use it to detect cancer as well as use it as a marker of recurrence – is that these biomarkers of early detection will be developed to have the same attributes.”

Genetic biomarkers

“To me, the current best biomarker is actually a genetic biomarker,” said Dr. Neuhausen, who is the Morris and Horowitz Families Professor in Cancer Etiology and Outcomes Research, Population Sciences, and also coleader of the cancer control and population sciences program, Comprehensive Cancer Center, at the Beckman Research Institute, City of Hope, in Duarte, Calif.

Identifying the breast cancer susceptibility genes BRCA 1 and BRCA 2 paved the way for targeted chemoprevention and prophylactic surgery, which have been highly effective in reducing the incidence of breast cancer as well as ovarian cancer. However, mutations in these genes account for no more than approximately 5% of all breast cancers.

“There has been identification of additional moderate- to high-risk genes, and the strategies that work to prevent cancer in BRCA 1 and 2 carriers are the same that can used for these other high-risk genes,” Dr. Neuhausen noted. For example, mutations of partner and localizer of BRCA2 (PALB2) have been linked to heightened breast cancer risk (N Engl J Med. 2014;371:497-506). Additionally, many companies now offer multigene clinical risk panels.

“The problem or the issue at the current time is that we really don’t know what the risk of cancer in unaffected women who are carrying mutations in these genes is,” she commented. “The good news is there is a large European, United States, and Canadian study that is actually going to be screening for mutations in a total of about 25,000 women, so that similar to what’s been done in BRCA 1 and 2, we will actually have better risk estimates that one can then use for these women.”

A U.S. genome-wide association research effort, the Breast Cancer Surveillance Consortium (BCSC), has thus far identified about 100 loci linked to an increased risk of developing breast cancer, according to Dr. Neuhausen. “On an individual level, these really don’t account for much risk. However, if you combine them, they actually do,” she said.

Applied clinically, a polygenic risk score incorporating 77 single-nucleotide polymorphisms can stratify women into quintiles of risk (J Natl Cancer Inst. 2015;107(5):djv036). Among women without a family history of the disease, lifetime risk ranges from 5.2% for those in the lowest quintile to 16.6% for those in the highest.

Moreover, adding the polygenic risk score to the commonly used BCSC model increases the area under the curve from 0.66 to 0.69 for breast cancer risk prediction. It also results in reclassification of some women, in particular, identifying an additional 3% as having a greater than 3% risk of developing the disease over 5 years. “So this actually does have good discrimination and is something that can be useful moving forward,” Dr. Neuhausen commented.

“We have these low-, moderate-, and high-risk genes, and they really all need to be incorporated into models. There is research ongoing to try to incorporate all of them into models along with lifestyle factors and mammography, family history, that kind of thing,” she said. “But I really think that these genetic markers are very important because rather than early detection, if we can prevent breast cancer, that’s actually the real goal.”

Other biomarkers

A variety of other, nongenetic biomarkers for early breast cancer detection and risk stratification are generally less far along in development, but several have shown promise, according to Dr. Neuhausen.

In the realm of proteomics, a three–amino acid profile detectable in saliva that exploits the hypermetabolic and hypercatabolic nature of cancer was found to have an area under the curve of 0.916 for differentiating women with early breast cancer from unaffected peers (Clin Chim Acta. 2015;447:23-31). “It was only a very small study and they really need to look further, but I think it was an intriguing idea,” she commented.

Infrared spectroscopy of plasma identified a protein fingerprint that had roughly 90% sensitivity and 80% specificity for detecting breast cancer in a cohort that included women with the disease, women with benign breast conditions, and healthy women (BMC Cancer. 2015;15:408). And a pair of proteins in serum—HSP27 and 14-3-3 sigma—accurately distinguished breast cancer cases from controls in a Chinese population (Proteomics. 2003;3:433-9).

One study has suggested how proteomic biomarkers might be integrated with conventional modalities, showing that a five-protein signature in serum (dtectDx Breast, Provista) performed well among women younger than age 50 years for differentiating benign breast lesions from invasive breast cancer in those with a BI-RADS 3 or 4 mammogram (J Clin Oncol. 2014;32(26 Suppl 20). However, the signature had a fairly high false-positive rate, Dr. Neuhausen noted.

As for yet other types of biomarkers, a panel of circulating cell-free methylated DNA of eight tumor suppressor genes was found to have sensitivity and specificity exceeding 90% for differentiating between samples from breast cancer patients and from unaffected women (PLoS One. 2011;6:e16080).MicroRNA profiles of breast cancer have been identified, but findings have generally been inconsistent across studies (Breast Cancer. 2015;7:59-79). Somatic mutations in p53 and PI3KCA, present in about a third of breast cancers, currently have issues when applied to screening. “Those kinds of things are maybe better for determining response to treatment and that kind of thing. Are they ready for early detection? Not really,” Dr. Neuhausen said.

The situation is similar for DNA copy number, although intriguingly, a recent study of a prenatal test looking for fetal copy number aberrations in maternal plasma incidentally discovered maternal cell-free DNA that had copy number changes (JAMA Oncol. 2015 June 5.doi: 10.1001/jamaoncol.2015.1883). Imaging ultimately found various cancers in three women out of about 4,000 tested.

Dr. Neuhausen disclosed that she had no relevant conflicts of interest.

SEATTLE – It’s time to consider an alternative strategy for evaluating new biomarkers that focuses on the predictive information they add, Michael W. Kattan, Ph.D., said at a joint meeting by Global Biomarkers Consortium and World Cutaneous Malignancies Congress.

“In my view, we get too fixated on P values or hazard ratios and odds ratios. Instead, we need to step back and think more about what the goal of any new marker is, and often, it’s to improve our ability to predict a patient outcome,” said Dr. Kattan, professor of medicine, epidemiology, and biostatistics at Case Western Reserve University, Cleveland, and chair of quantitative health sciences at the Cleveland Clinic. “If that’s the case, why not worry more about something like incremental predictive accuracy or incremental predictive ability associated with that new marker, and make our decisions and our modeling steps toward that?”

The long-used, conventional three-step approach to evaluating a new biomarker – assessing its correlation with an established biomarker, its association with an outcome in univariate analysis, and finally its performance in a multivariate analysis (J Natl Cancer Inst. 2003;95:634-5) – has considerable issues, according to Dr. Kattan.

In particular, the multivariate analysis is problematic. “My P value is testing whether my hazard ratio is 1, it’s not per se an improvement in predictive accuracy, which is what I’m going to argue that the new marker should do,” he said. But more concerning is the fact that the hazard ratio is affected by factors the investigators control, such as whether the new biomarker is coded as a continuous or categorical variable, which established biomarkers are included, and any data transformations done.

“At the end of the day, things are getting a little bit subjective because I have a bunch of knobs under my control as the keeper of the data. I can turn all of these knobs, and unfortunately, I don’t have excellent arguments to defend how I would do that, and they may very well affect the [hazard ratio] that has everyone’s attention,” Dr. Kattan elaborated.

Thus, an alternative approach is needed, one that tests the new biomarker as part of a model and addresses the central question of whether it improves predictive accuracy, he maintained. “It’s [comparing] a model of markers that lacks the new marker versus a model of markers that contains the new marker. So it’s a model versus model comparison, it’s not simply looking at the marker in isolation, which is where we get in trouble with the typical way.”

Furthermore, aiming for the most accurate model removes much of the subjectivity of the conventional approach, he added. “Remember, I said there were knobs I could turn that might change the hazard ratio and I didn’t have a good defense for how I would turn these knobs. … Now I do, now I have an explicit goal: I want to have a prediction model that predicts patient outcome as accurately as I can. So whatever I’m doing with my knobs and stuff, that ought to be delivering a more accurate prediction model.”

Dr. Kattan outlined a four-step alternative approach to evaluating new biomarkers. The first step entails calculating the improvement in the concordance index, similar to an area under the receiver operating characteristic (ROC) curve, with the new biomarker. Ideally, that number will increase in a model that contains the marker, indicating an improvement in predictive accuracy.

In the second step, which assesses model calibration, established and new biomarkers are entered into a multivariate model predicting the outcome of interest (Clin Cancer Res. 2004;10:822-4). If the concordance index drops by a clinically significant degree when the new biomarker is omitted, indicating a loss of predictive accuracy, it advances.

The third step is to construct scatterplots comparing results obtained with prediction models of the outcome, say, 10-year progression-free survival, that do and do not contain the new marker, say, surgeon experience with prostatectomy (Cancer. 2009;115:1005-10). If the improvement in accuracy here is clinically significant, the marker again advances.

In the fourth and final step, decision curve analysis, the net benefit is plotted as a function of the threshold for clinical action (Epidemiology. 2010;21:128-38). “This gets at, should I be making clinical decisions based on the prediction model, or should I just treat everyone or treat no one. It’s a way of looking at what the net benefit is of the prediction model across the spectrum of predictions,” Dr. Kattan explained. “So you would first decide what’s my threshold for action … where’s it going to change what I do, and then read upwards [in the plot] and see what the net benefit is.”

Dr. Kattan disclosed that he receives consulting fees from Bayer, Exosome, GlaxoSmithKline, HistoSonics, and Merck.

SEATTLE – It’s time to consider an alternative strategy for evaluating new biomarkers that focuses on the predictive information they add, Michael W. Kattan, Ph.D., said at a joint meeting by Global Biomarkers Consortium and World Cutaneous Malignancies Congress.

“In my view, we get too fixated on P values or hazard ratios and odds ratios. Instead, we need to step back and think more about what the goal of any new marker is, and often, it’s to improve our ability to predict a patient outcome,” said Dr. Kattan, professor of medicine, epidemiology, and biostatistics at Case Western Reserve University, Cleveland, and chair of quantitative health sciences at the Cleveland Clinic. “If that’s the case, why not worry more about something like incremental predictive accuracy or incremental predictive ability associated with that new marker, and make our decisions and our modeling steps toward that?”

The long-used, conventional three-step approach to evaluating a new biomarker – assessing its correlation with an established biomarker, its association with an outcome in univariate analysis, and finally its performance in a multivariate analysis (J Natl Cancer Inst. 2003;95:634-5) – has considerable issues, according to Dr. Kattan.

In particular, the multivariate analysis is problematic. “My P value is testing whether my hazard ratio is 1, it’s not per se an improvement in predictive accuracy, which is what I’m going to argue that the new marker should do,” he said. But more concerning is the fact that the hazard ratio is affected by factors the investigators control, such as whether the new biomarker is coded as a continuous or categorical variable, which established biomarkers are included, and any data transformations done.

“At the end of the day, things are getting a little bit subjective because I have a bunch of knobs under my control as the keeper of the data. I can turn all of these knobs, and unfortunately, I don’t have excellent arguments to defend how I would do that, and they may very well affect the [hazard ratio] that has everyone’s attention,” Dr. Kattan elaborated.

Thus, an alternative approach is needed, one that tests the new biomarker as part of a model and addresses the central question of whether it improves predictive accuracy, he maintained. “It’s [comparing] a model of markers that lacks the new marker versus a model of markers that contains the new marker. So it’s a model versus model comparison, it’s not simply looking at the marker in isolation, which is where we get in trouble with the typical way.”

Furthermore, aiming for the most accurate model removes much of the subjectivity of the conventional approach, he added. “Remember, I said there were knobs I could turn that might change the hazard ratio and I didn’t have a good defense for how I would turn these knobs. … Now I do, now I have an explicit goal: I want to have a prediction model that predicts patient outcome as accurately as I can. So whatever I’m doing with my knobs and stuff, that ought to be delivering a more accurate prediction model.”

Dr. Kattan outlined a four-step alternative approach to evaluating new biomarkers. The first step entails calculating the improvement in the concordance index, similar to an area under the receiver operating characteristic (ROC) curve, with the new biomarker. Ideally, that number will increase in a model that contains the marker, indicating an improvement in predictive accuracy.

In the second step, which assesses model calibration, established and new biomarkers are entered into a multivariate model predicting the outcome of interest (Clin Cancer Res. 2004;10:822-4). If the concordance index drops by a clinically significant degree when the new biomarker is omitted, indicating a loss of predictive accuracy, it advances.

The third step is to construct scatterplots comparing results obtained with prediction models of the outcome, say, 10-year progression-free survival, that do and do not contain the new marker, say, surgeon experience with prostatectomy (Cancer. 2009;115:1005-10). If the improvement in accuracy here is clinically significant, the marker again advances.

In the fourth and final step, decision curve analysis, the net benefit is plotted as a function of the threshold for clinical action (Epidemiology. 2010;21:128-38). “This gets at, should I be making clinical decisions based on the prediction model, or should I just treat everyone or treat no one. It’s a way of looking at what the net benefit is of the prediction model across the spectrum of predictions,” Dr. Kattan explained. “So you would first decide what’s my threshold for action … where’s it going to change what I do, and then read upwards [in the plot] and see what the net benefit is.”

Dr. Kattan disclosed that he receives consulting fees from Bayer, Exosome, GlaxoSmithKline, HistoSonics, and Merck.

SEATTLE – It’s time to consider an alternative strategy for evaluating new biomarkers that focuses on the predictive information they add, Michael W. Kattan, Ph.D., said at a joint meeting by Global Biomarkers Consortium and World Cutaneous Malignancies Congress.

“In my view, we get too fixated on P values or hazard ratios and odds ratios. Instead, we need to step back and think more about what the goal of any new marker is, and often, it’s to improve our ability to predict a patient outcome,” said Dr. Kattan, professor of medicine, epidemiology, and biostatistics at Case Western Reserve University, Cleveland, and chair of quantitative health sciences at the Cleveland Clinic. “If that’s the case, why not worry more about something like incremental predictive accuracy or incremental predictive ability associated with that new marker, and make our decisions and our modeling steps toward that?”

The long-used, conventional three-step approach to evaluating a new biomarker – assessing its correlation with an established biomarker, its association with an outcome in univariate analysis, and finally its performance in a multivariate analysis (J Natl Cancer Inst. 2003;95:634-5) – has considerable issues, according to Dr. Kattan.

In particular, the multivariate analysis is problematic. “My P value is testing whether my hazard ratio is 1, it’s not per se an improvement in predictive accuracy, which is what I’m going to argue that the new marker should do,” he said. But more concerning is the fact that the hazard ratio is affected by factors the investigators control, such as whether the new biomarker is coded as a continuous or categorical variable, which established biomarkers are included, and any data transformations done.

“At the end of the day, things are getting a little bit subjective because I have a bunch of knobs under my control as the keeper of the data. I can turn all of these knobs, and unfortunately, I don’t have excellent arguments to defend how I would do that, and they may very well affect the [hazard ratio] that has everyone’s attention,” Dr. Kattan elaborated.

Thus, an alternative approach is needed, one that tests the new biomarker as part of a model and addresses the central question of whether it improves predictive accuracy, he maintained. “It’s [comparing] a model of markers that lacks the new marker versus a model of markers that contains the new marker. So it’s a model versus model comparison, it’s not simply looking at the marker in isolation, which is where we get in trouble with the typical way.”

Furthermore, aiming for the most accurate model removes much of the subjectivity of the conventional approach, he added. “Remember, I said there were knobs I could turn that might change the hazard ratio and I didn’t have a good defense for how I would turn these knobs. … Now I do, now I have an explicit goal: I want to have a prediction model that predicts patient outcome as accurately as I can. So whatever I’m doing with my knobs and stuff, that ought to be delivering a more accurate prediction model.”

Dr. Kattan outlined a four-step alternative approach to evaluating new biomarkers. The first step entails calculating the improvement in the concordance index, similar to an area under the receiver operating characteristic (ROC) curve, with the new biomarker. Ideally, that number will increase in a model that contains the marker, indicating an improvement in predictive accuracy.

In the second step, which assesses model calibration, established and new biomarkers are entered into a multivariate model predicting the outcome of interest (Clin Cancer Res. 2004;10:822-4). If the concordance index drops by a clinically significant degree when the new biomarker is omitted, indicating a loss of predictive accuracy, it advances.

The third step is to construct scatterplots comparing results obtained with prediction models of the outcome, say, 10-year progression-free survival, that do and do not contain the new marker, say, surgeon experience with prostatectomy (Cancer. 2009;115:1005-10). If the improvement in accuracy here is clinically significant, the marker again advances.

In the fourth and final step, decision curve analysis, the net benefit is plotted as a function of the threshold for clinical action (Epidemiology. 2010;21:128-38). “This gets at, should I be making clinical decisions based on the prediction model, or should I just treat everyone or treat no one. It’s a way of looking at what the net benefit is of the prediction model across the spectrum of predictions,” Dr. Kattan explained. “So you would first decide what’s my threshold for action … where’s it going to change what I do, and then read upwards [in the plot] and see what the net benefit is.”

Dr. Kattan disclosed that he receives consulting fees from Bayer, Exosome, GlaxoSmithKline, HistoSonics, and Merck.

SEATTLE – “Biomarkers have a promising role in the early detection and risk stratification of colonic neoplasia,” Dr. Hemant K. Roy said at a joint meeting by Global Biomarkers Consortium and World Cutaneous Malignancies Congress.

A better, personalized strategy for screening the population is needed given the current limitations of colonoscopy, he said. “Unfortunately, we know that the endoscopic capacity and funding are really inadequate to perform colonoscopy on the entire average at-risk population, which is greater than 100 million Americans over the age of 50. Juxtaposed with this is the fact that most screenings by colonoscopy are negative for a significant lesion.

“In retrospect, most colonoscopies are squandered if you just look at colonoscopy as a cancer prevention tool. The solution may be a better risk stratification tool, to develop a prescreen that can detect patients most likely to benefit from colonoscopy,” he said.

Biomarker tests

Numerous colorectal cancer biomarkers are being evaluated in various physiologic compartments, according to Dr. Roy, professor of medicine and chief of the section of gastroenterology, Boston University Medical Center. The development of these biomarkers is complicated by several factors, such as the heterogeneity of the neoplasia and the large number of mutations found in this cancer.

Most of those being investigated for screening are still in the earlier phases of development. One now on the market is a blood test for methylated SEPT9 (ColoVantage). Its sensitivity is 48% for colorectal cancer but only 11% for advanced adenomas (Gut. 2014 Feb;63(2):317-25). “This represents a small step forward, but it is available,” Dr. Roy said.

Among other promising blood tests is the microRNA MiR-21 assay. This assay was found to have an area under the curve of 0.92 for the detection of colorectal cancer and 0.81 for the detection of adenomas (J Natl Cancer Inst. 2013 Jun 19;105(12):849-59).

A stool DNA test on the market (Cologuard) outperformed the fecal immunochemical test (FIT) for detecting colon cancer, with a sensitivity of 92% and a specificity of 87% (N Engl J Med. 2014 Apr 3;370(14):1287-97). But the sensitivity was only 42% for advanced adenomas. “This is our real target, and this is a problem, especially as the Cologuard test, while good, is expensive at about $500 a test,” he said.

Fecal microRNA biomarkers have also shown some promise, with a pair found to perform well for differentiating individuals with advanced adenomas and cancers from unaffected peers (Cancer Epidemiol Biomarkers Prev. 2010 Jul;19(7):1766-74). “While this needs to be repeated, it has some potential,” Dr. Roy commented.

Tissue tests exploit the fact that, in general, the entire large intestine is exposed to the same genetic and environmental influences. “The corollary is that we can potentially look in the rectum or at other lesions in the colon and predict risk throughout the colon,” he explained.

Promising tissue biomarkers include nanoarchitectural changes in endoscopically normal-appearing rectal mucosa that appear to reflect the risk of carcinogenesis throughout the colon (Cancer Res. Jun 1;72(11):2720-7) and microsatellite instability, which can be used as part of an algorithm to identify Lynch syndrome (Gastroenterology. Aug;147(2):502-26).

Clinical applications

“Our long-term goal is to take the general population and find these biomarkers, to narrow down our conventional cancer-screening group,” Dr. Roy explained. An additional goal is developing noninvasive tests that are more acceptable to patients than colonoscopy.

Although a randomized trial showed colonoscopy performs better than FIT at detecting advanced adenomas among patients who actually undergo screening, the difference is attenuated among all patients offered screening, likely due in part to poorer compliance with colonoscopy (N Engl J Med. 2012 Feb 23;366(8):697-706).

“You could say, we should do colonoscopy on everyone [eligible], and I’m a gastroenterologist, so that would make me very happy,” Dr. Roy commented. “But if we can’t get people to undergo the test, and with the costs associated with it and the really very small number of people in whom we actually find a significant polyp, we need a better strategy. And that is where I think a biomarker approach is going to be very helpful.”

He predicted risk-stratified use of screening tests in the future. “I think [we will see] a combination approach, with the highest-risk patients going directly to colonoscopy and the lowest-risk patients getting less sensitive biomarker-based tests,” he said.

Biomarkers may also be applied clinically to help inform colorectal cancer chemoprevention, for example, by identifying patients most likely to benefit from aspirin therapy and by helping to assess whether the dose should be altered, according to Dr. Roy. To this end, research on tissue biomarkers suggests that aspirin’s risk-reducing benefit is limited to individuals with higher 15-PDGH expression in normal colonic mucosa (Sci Transl Med. 2014 Apr 23;6(233):233re2).

Dr. Roy disclosed that he receives a salary from NanoCytomics, and has an ownership interest in American BioOptics, NanoCytomics, and Pegasus Biosolutions.

SEATTLE – “Biomarkers have a promising role in the early detection and risk stratification of colonic neoplasia,” Dr. Hemant K. Roy said at a joint meeting by Global Biomarkers Consortium and World Cutaneous Malignancies Congress.

A better, personalized strategy for screening the population is needed given the current limitations of colonoscopy, he said. “Unfortunately, we know that the endoscopic capacity and funding are really inadequate to perform colonoscopy on the entire average at-risk population, which is greater than 100 million Americans over the age of 50. Juxtaposed with this is the fact that most screenings by colonoscopy are negative for a significant lesion.

“In retrospect, most colonoscopies are squandered if you just look at colonoscopy as a cancer prevention tool. The solution may be a better risk stratification tool, to develop a prescreen that can detect patients most likely to benefit from colonoscopy,” he said.

Biomarker tests

Numerous colorectal cancer biomarkers are being evaluated in various physiologic compartments, according to Dr. Roy, professor of medicine and chief of the section of gastroenterology, Boston University Medical Center. The development of these biomarkers is complicated by several factors, such as the heterogeneity of the neoplasia and the large number of mutations found in this cancer.

Most of those being investigated for screening are still in the earlier phases of development. One now on the market is a blood test for methylated SEPT9 (ColoVantage). Its sensitivity is 48% for colorectal cancer but only 11% for advanced adenomas (Gut. 2014 Feb;63(2):317-25). “This represents a small step forward, but it is available,” Dr. Roy said.

Among other promising blood tests is the microRNA MiR-21 assay. This assay was found to have an area under the curve of 0.92 for the detection of colorectal cancer and 0.81 for the detection of adenomas (J Natl Cancer Inst. 2013 Jun 19;105(12):849-59).

A stool DNA test on the market (Cologuard) outperformed the fecal immunochemical test (FIT) for detecting colon cancer, with a sensitivity of 92% and a specificity of 87% (N Engl J Med. 2014 Apr 3;370(14):1287-97). But the sensitivity was only 42% for advanced adenomas. “This is our real target, and this is a problem, especially as the Cologuard test, while good, is expensive at about $500 a test,” he said.

Fecal microRNA biomarkers have also shown some promise, with a pair found to perform well for differentiating individuals with advanced adenomas and cancers from unaffected peers (Cancer Epidemiol Biomarkers Prev. 2010 Jul;19(7):1766-74). “While this needs to be repeated, it has some potential,” Dr. Roy commented.

Tissue tests exploit the fact that, in general, the entire large intestine is exposed to the same genetic and environmental influences. “The corollary is that we can potentially look in the rectum or at other lesions in the colon and predict risk throughout the colon,” he explained.

Promising tissue biomarkers include nanoarchitectural changes in endoscopically normal-appearing rectal mucosa that appear to reflect the risk of carcinogenesis throughout the colon (Cancer Res. Jun 1;72(11):2720-7) and microsatellite instability, which can be used as part of an algorithm to identify Lynch syndrome (Gastroenterology. Aug;147(2):502-26).

Clinical applications

“Our long-term goal is to take the general population and find these biomarkers, to narrow down our conventional cancer-screening group,” Dr. Roy explained. An additional goal is developing noninvasive tests that are more acceptable to patients than colonoscopy.

Although a randomized trial showed colonoscopy performs better than FIT at detecting advanced adenomas among patients who actually undergo screening, the difference is attenuated among all patients offered screening, likely due in part to poorer compliance with colonoscopy (N Engl J Med. 2012 Feb 23;366(8):697-706).

“You could say, we should do colonoscopy on everyone [eligible], and I’m a gastroenterologist, so that would make me very happy,” Dr. Roy commented. “But if we can’t get people to undergo the test, and with the costs associated with it and the really very small number of people in whom we actually find a significant polyp, we need a better strategy. And that is where I think a biomarker approach is going to be very helpful.”

He predicted risk-stratified use of screening tests in the future. “I think [we will see] a combination approach, with the highest-risk patients going directly to colonoscopy and the lowest-risk patients getting less sensitive biomarker-based tests,” he said.

Biomarkers may also be applied clinically to help inform colorectal cancer chemoprevention, for example, by identifying patients most likely to benefit from aspirin therapy and by helping to assess whether the dose should be altered, according to Dr. Roy. To this end, research on tissue biomarkers suggests that aspirin’s risk-reducing benefit is limited to individuals with higher 15-PDGH expression in normal colonic mucosa (Sci Transl Med. 2014 Apr 23;6(233):233re2).

Dr. Roy disclosed that he receives a salary from NanoCytomics, and has an ownership interest in American BioOptics, NanoCytomics, and Pegasus Biosolutions.

SEATTLE – “Biomarkers have a promising role in the early detection and risk stratification of colonic neoplasia,” Dr. Hemant K. Roy said at a joint meeting by Global Biomarkers Consortium and World Cutaneous Malignancies Congress.

A better, personalized strategy for screening the population is needed given the current limitations of colonoscopy, he said. “Unfortunately, we know that the endoscopic capacity and funding are really inadequate to perform colonoscopy on the entire average at-risk population, which is greater than 100 million Americans over the age of 50. Juxtaposed with this is the fact that most screenings by colonoscopy are negative for a significant lesion.

“In retrospect, most colonoscopies are squandered if you just look at colonoscopy as a cancer prevention tool. The solution may be a better risk stratification tool, to develop a prescreen that can detect patients most likely to benefit from colonoscopy,” he said.

Biomarker tests

Numerous colorectal cancer biomarkers are being evaluated in various physiologic compartments, according to Dr. Roy, professor of medicine and chief of the section of gastroenterology, Boston University Medical Center. The development of these biomarkers is complicated by several factors, such as the heterogeneity of the neoplasia and the large number of mutations found in this cancer.

Most of those being investigated for screening are still in the earlier phases of development. One now on the market is a blood test for methylated SEPT9 (ColoVantage). Its sensitivity is 48% for colorectal cancer but only 11% for advanced adenomas (Gut. 2014 Feb;63(2):317-25). “This represents a small step forward, but it is available,” Dr. Roy said.

Among other promising blood tests is the microRNA MiR-21 assay. This assay was found to have an area under the curve of 0.92 for the detection of colorectal cancer and 0.81 for the detection of adenomas (J Natl Cancer Inst. 2013 Jun 19;105(12):849-59).

A stool DNA test on the market (Cologuard) outperformed the fecal immunochemical test (FIT) for detecting colon cancer, with a sensitivity of 92% and a specificity of 87% (N Engl J Med. 2014 Apr 3;370(14):1287-97). But the sensitivity was only 42% for advanced adenomas. “This is our real target, and this is a problem, especially as the Cologuard test, while good, is expensive at about $500 a test,” he said.

Fecal microRNA biomarkers have also shown some promise, with a pair found to perform well for differentiating individuals with advanced adenomas and cancers from unaffected peers (Cancer Epidemiol Biomarkers Prev. 2010 Jul;19(7):1766-74). “While this needs to be repeated, it has some potential,” Dr. Roy commented.

Tissue tests exploit the fact that, in general, the entire large intestine is exposed to the same genetic and environmental influences. “The corollary is that we can potentially look in the rectum or at other lesions in the colon and predict risk throughout the colon,” he explained.

Promising tissue biomarkers include nanoarchitectural changes in endoscopically normal-appearing rectal mucosa that appear to reflect the risk of carcinogenesis throughout the colon (Cancer Res. Jun 1;72(11):2720-7) and microsatellite instability, which can be used as part of an algorithm to identify Lynch syndrome (Gastroenterology. Aug;147(2):502-26).

Clinical applications

“Our long-term goal is to take the general population and find these biomarkers, to narrow down our conventional cancer-screening group,” Dr. Roy explained. An additional goal is developing noninvasive tests that are more acceptable to patients than colonoscopy.

Although a randomized trial showed colonoscopy performs better than FIT at detecting advanced adenomas among patients who actually undergo screening, the difference is attenuated among all patients offered screening, likely due in part to poorer compliance with colonoscopy (N Engl J Med. 2012 Feb 23;366(8):697-706).

“You could say, we should do colonoscopy on everyone [eligible], and I’m a gastroenterologist, so that would make me very happy,” Dr. Roy commented. “But if we can’t get people to undergo the test, and with the costs associated with it and the really very small number of people in whom we actually find a significant polyp, we need a better strategy. And that is where I think a biomarker approach is going to be very helpful.”

He predicted risk-stratified use of screening tests in the future. “I think [we will see] a combination approach, with the highest-risk patients going directly to colonoscopy and the lowest-risk patients getting less sensitive biomarker-based tests,” he said.

Biomarkers may also be applied clinically to help inform colorectal cancer chemoprevention, for example, by identifying patients most likely to benefit from aspirin therapy and by helping to assess whether the dose should be altered, according to Dr. Roy. To this end, research on tissue biomarkers suggests that aspirin’s risk-reducing benefit is limited to individuals with higher 15-PDGH expression in normal colonic mucosa (Sci Transl Med. 2014 Apr 23;6(233):233re2).

Dr. Roy disclosed that he receives a salary from NanoCytomics, and has an ownership interest in American BioOptics, NanoCytomics, and Pegasus Biosolutions.

SEATTLE – Recent developments in the field of melanoma may help clinicians refine their approach to this disease, according to Dr. Suraj S. Venna, a dermatologist and medical director of the Inova Melanoma and Skin Cancer Center in Fairfax, Va. He discussed several key studies and guideline revisions at a joint meeting of the Global Biomarkers Consortium and World Cutaneous Malignancies Congress.

Sentinel node biopsy for thin melanomas

Guidance regarding staging of and use of sentinel lymph node (SLN) biopsy in cases of thin melanomas, those no more than 1 mm thick, is in flux, and these gray areas pose challenges to management.

In its seventh edition of the TNM Staging System published in 2010, the American Joint Committee on Cancer started incorporating mitotic rate to substage T1 melanomas, according to Dr. Venna. Those having a mitotic rate of at least 1 mitosis/mm² are now categorized as T1b. Yet, even within this subset, there is a spectrum of 10-year survival according to mitotic rate (J Clin Oncol. 2011;29:2199-205). Also, the association of mitotic rate with regional nodal metastasis is unclear.

Therefore, “T1b is not an automatic recommendation for lymph node biopsy,” he said. Decisions about SLN in patients with such tumors should consider all available clinical and pathologic factors.

The National Comprehensive Cancer Network has also revised its guidelines in this area. In 2011, they advised that SLN could be discussed and considered in stage IA melanomas with various high-risk features and in selected stage IB melanomas based on mitotic rate and thickness.

But in their 2015 update, they gave priority to lesion depth. “What they say is other than Breslow depth, there remains little consensus regarding the significance of ulceration, mitotic rate, or lymphovascular invasion,” Dr. Venna elaborated. Now, SLN biopsy can be considered on a case-by-case basis for stage 1A melanomas that are thicker or have uncertain microstaging, and can be discussed and offered for thicker stage 1B melanomas with ulceration or an elevated mitotic rate.

“At the end of the day, when we are all in our multidisciplinary clinics trying to talk to patients about this, these are nuances that really cause a lot of angst for us and the patient in terms of making decisions,” commented Dr. Venna.

“Acknowledge the prognostic gap in the T1b cohort and that not all T1b melanomas are created equal. We all anticipate that molecular techniques may better identify who among that thin group may benefit from lymph node testing,” Dr. Venna said.

Screening strategies

Population-based screening for melanoma and other skin cancers remains controversial, and this practice is generally limited to areas of the world having high incidence, according to Dr. Venna. Germany was the first country to implement a nationwide program, starting in 2008.

“There is evidence that screening may lead to the prevention of a substantial proportion of melanoma deaths,” he maintained. However, the U.S. National Cancer Institute’s position is that current evidence is inadequate and population-based screening has potential drawbacks such as overdiagnosis.

In the meantime, many opportunities to detect melanoma during routine medical care may be missed, according to Dr. Venna; for example, auscultating a patients’ heart or lungs without pulling up their shirt may miss truncal lesions. “Use every opportunity to screen the patient – that’s opportunistic screening,” he recommended.

He also endorsed listening to patients’ concerns and balancing their requests with judicious removal of lesions, as a large share of melanomas are self-detected. In one study of note, 9 of 166 lesions removed and found to be melanoma were removed only to reassure the patient (Arch Dermatol. 2005;141:434-8).

Barriers to and imperatives for early detection

Data suggest that access to specialist care is problematic for the early detection of melanoma. In a survey of dermatologists, the median wait time for an appointment was 8 days for botulinum toxin type A (Botox) injections, compared with 26 days for evaluation of a suspicious mole (J Am Acad Dermatol. 2007;57:985-9).

“The concern of course is a delay in diagnosis and the fact that the nondermatologists will have an opportunity likely to capture melanoma before we do,” Dr. Venna commented.

Imperatives for early detection include not only better outcomes but also reduced costs, given the hefty price tag of novel targeted therapies being used to treat advanced disease, he noted.

A recent phase III trial found that progression-free survival in advanced melanoma was roughly two to four times longer with the combination of the targeted agents nivolumab (Opdivo) and ipilimumab (Yervoy) than with either agent alone (ASCO 2015, Abstract LBA1). But a discussion touching on costs noted that for the average-weight American, this combination would run about $300,000 annually (ASCO 2015, Saltz L).

Risk and protective factors

New data have implicated medications used to treat erectile dysfunction (ED) as a risk factor for melanoma, raising concern given their widespread use, Dr. Venna said.

A cohort study found that melanoma risk was 84% higher with recent use of sildenafil (Viagra) and 92% higher with ever use (JAMA Intern Med. 2014;174:964-70). But the absolute number of excess cases was fairly small.

A case-control study looking at all types of ED drugs found that receipt of a single prescription was associated with a 32% higher risk of melanoma (JAMA. 2015;313:2449-55). Use increased risk of in situ and stage I melanoma, but not more advanced disease. There was no difference between short- and long-acting agents.

“Prospective studies with clearly defined inclusions and exclusions are needed, with dosing in particular,” Dr. Venna summed up. “There appears to be a modest association, but it certainly is not enough to call for widespread discontinuation. In the patients who are on these drugs, it’s good to document [use] until we have more clarity about the biologic effects of this ultimately.”

A study using dietary surveys and having an average follow-up of more than a decade showed that coffee drinkers had a 20% lower risk of melanoma relative to nondrinkers (J Natl Cancer Inst. 2015 Jan 20;107(2)). But benefit was significant only among the subset drinking at least 4 cups a day.

“It’s premature to advise coffee intake based on this paper, although it does build upon an earlier Norwegian study from the 1990s which showed a similar trend,” Dr. Venna said. “At the end of the day, these data are intriguing, but obviously there are a lot of side effects of trying to consume 4 cups of coffee a day, so further study is certainly warranted.”

Dr. Venna disclosed that he had no relevant conflicts of interest.

SEATTLE – Recent developments in the field of melanoma may help clinicians refine their approach to this disease, according to Dr. Suraj S. Venna, a dermatologist and medical director of the Inova Melanoma and Skin Cancer Center in Fairfax, Va. He discussed several key studies and guideline revisions at a joint meeting of the Global Biomarkers Consortium and World Cutaneous Malignancies Congress.

Sentinel node biopsy for thin melanomas

Guidance regarding staging of and use of sentinel lymph node (SLN) biopsy in cases of thin melanomas, those no more than 1 mm thick, is in flux, and these gray areas pose challenges to management.

In its seventh edition of the TNM Staging System published in 2010, the American Joint Committee on Cancer started incorporating mitotic rate to substage T1 melanomas, according to Dr. Venna. Those having a mitotic rate of at least 1 mitosis/mm² are now categorized as T1b. Yet, even within this subset, there is a spectrum of 10-year survival according to mitotic rate (J Clin Oncol. 2011;29:2199-205). Also, the association of mitotic rate with regional nodal metastasis is unclear.

Therefore, “T1b is not an automatic recommendation for lymph node biopsy,” he said. Decisions about SLN in patients with such tumors should consider all available clinical and pathologic factors.

The National Comprehensive Cancer Network has also revised its guidelines in this area. In 2011, they advised that SLN could be discussed and considered in stage IA melanomas with various high-risk features and in selected stage IB melanomas based on mitotic rate and thickness.

But in their 2015 update, they gave priority to lesion depth. “What they say is other than Breslow depth, there remains little consensus regarding the significance of ulceration, mitotic rate, or lymphovascular invasion,” Dr. Venna elaborated. Now, SLN biopsy can be considered on a case-by-case basis for stage 1A melanomas that are thicker or have uncertain microstaging, and can be discussed and offered for thicker stage 1B melanomas with ulceration or an elevated mitotic rate.

“At the end of the day, when we are all in our multidisciplinary clinics trying to talk to patients about this, these are nuances that really cause a lot of angst for us and the patient in terms of making decisions,” commented Dr. Venna.

“Acknowledge the prognostic gap in the T1b cohort and that not all T1b melanomas are created equal. We all anticipate that molecular techniques may better identify who among that thin group may benefit from lymph node testing,” Dr. Venna said.

Screening strategies

Population-based screening for melanoma and other skin cancers remains controversial, and this practice is generally limited to areas of the world having high incidence, according to Dr. Venna. Germany was the first country to implement a nationwide program, starting in 2008.

“There is evidence that screening may lead to the prevention of a substantial proportion of melanoma deaths,” he maintained. However, the U.S. National Cancer Institute’s position is that current evidence is inadequate and population-based screening has potential drawbacks such as overdiagnosis.

In the meantime, many opportunities to detect melanoma during routine medical care may be missed, according to Dr. Venna; for example, auscultating a patients’ heart or lungs without pulling up their shirt may miss truncal lesions. “Use every opportunity to screen the patient – that’s opportunistic screening,” he recommended.

He also endorsed listening to patients’ concerns and balancing their requests with judicious removal of lesions, as a large share of melanomas are self-detected. In one study of note, 9 of 166 lesions removed and found to be melanoma were removed only to reassure the patient (Arch Dermatol. 2005;141:434-8).

Barriers to and imperatives for early detection

Data suggest that access to specialist care is problematic for the early detection of melanoma. In a survey of dermatologists, the median wait time for an appointment was 8 days for botulinum toxin type A (Botox) injections, compared with 26 days for evaluation of a suspicious mole (J Am Acad Dermatol. 2007;57:985-9).

“The concern of course is a delay in diagnosis and the fact that the nondermatologists will have an opportunity likely to capture melanoma before we do,” Dr. Venna commented.

Imperatives for early detection include not only better outcomes but also reduced costs, given the hefty price tag of novel targeted therapies being used to treat advanced disease, he noted.

A recent phase III trial found that progression-free survival in advanced melanoma was roughly two to four times longer with the combination of the targeted agents nivolumab (Opdivo) and ipilimumab (Yervoy) than with either agent alone (ASCO 2015, Abstract LBA1). But a discussion touching on costs noted that for the average-weight American, this combination would run about $300,000 annually (ASCO 2015, Saltz L).

Risk and protective factors

New data have implicated medications used to treat erectile dysfunction (ED) as a risk factor for melanoma, raising concern given their widespread use, Dr. Venna said.

A cohort study found that melanoma risk was 84% higher with recent use of sildenafil (Viagra) and 92% higher with ever use (JAMA Intern Med. 2014;174:964-70). But the absolute number of excess cases was fairly small.

A case-control study looking at all types of ED drugs found that receipt of a single prescription was associated with a 32% higher risk of melanoma (JAMA. 2015;313:2449-55). Use increased risk of in situ and stage I melanoma, but not more advanced disease. There was no difference between short- and long-acting agents.

“Prospective studies with clearly defined inclusions and exclusions are needed, with dosing in particular,” Dr. Venna summed up. “There appears to be a modest association, but it certainly is not enough to call for widespread discontinuation. In the patients who are on these drugs, it’s good to document [use] until we have more clarity about the biologic effects of this ultimately.”

A study using dietary surveys and having an average follow-up of more than a decade showed that coffee drinkers had a 20% lower risk of melanoma relative to nondrinkers (J Natl Cancer Inst. 2015 Jan 20;107(2)). But benefit was significant only among the subset drinking at least 4 cups a day.

“It’s premature to advise coffee intake based on this paper, although it does build upon an earlier Norwegian study from the 1990s which showed a similar trend,” Dr. Venna said. “At the end of the day, these data are intriguing, but obviously there are a lot of side effects of trying to consume 4 cups of coffee a day, so further study is certainly warranted.”

Dr. Venna disclosed that he had no relevant conflicts of interest.

SEATTLE – Recent developments in the field of melanoma may help clinicians refine their approach to this disease, according to Dr. Suraj S. Venna, a dermatologist and medical director of the Inova Melanoma and Skin Cancer Center in Fairfax, Va. He discussed several key studies and guideline revisions at a joint meeting of the Global Biomarkers Consortium and World Cutaneous Malignancies Congress.

Sentinel node biopsy for thin melanomas

Guidance regarding staging of and use of sentinel lymph node (SLN) biopsy in cases of thin melanomas, those no more than 1 mm thick, is in flux, and these gray areas pose challenges to management.

In its seventh edition of the TNM Staging System published in 2010, the American Joint Committee on Cancer started incorporating mitotic rate to substage T1 melanomas, according to Dr. Venna. Those having a mitotic rate of at least 1 mitosis/mm² are now categorized as T1b. Yet, even within this subset, there is a spectrum of 10-year survival according to mitotic rate (J Clin Oncol. 2011;29:2199-205). Also, the association of mitotic rate with regional nodal metastasis is unclear.

Therefore, “T1b is not an automatic recommendation for lymph node biopsy,” he said. Decisions about SLN in patients with such tumors should consider all available clinical and pathologic factors.

The National Comprehensive Cancer Network has also revised its guidelines in this area. In 2011, they advised that SLN could be discussed and considered in stage IA melanomas with various high-risk features and in selected stage IB melanomas based on mitotic rate and thickness.

But in their 2015 update, they gave priority to lesion depth. “What they say is other than Breslow depth, there remains little consensus regarding the significance of ulceration, mitotic rate, or lymphovascular invasion,” Dr. Venna elaborated. Now, SLN biopsy can be considered on a case-by-case basis for stage 1A melanomas that are thicker or have uncertain microstaging, and can be discussed and offered for thicker stage 1B melanomas with ulceration or an elevated mitotic rate.

“At the end of the day, when we are all in our multidisciplinary clinics trying to talk to patients about this, these are nuances that really cause a lot of angst for us and the patient in terms of making decisions,” commented Dr. Venna.

“Acknowledge the prognostic gap in the T1b cohort and that not all T1b melanomas are created equal. We all anticipate that molecular techniques may better identify who among that thin group may benefit from lymph node testing,” Dr. Venna said.

Screening strategies

Population-based screening for melanoma and other skin cancers remains controversial, and this practice is generally limited to areas of the world having high incidence, according to Dr. Venna. Germany was the first country to implement a nationwide program, starting in 2008.

“There is evidence that screening may lead to the prevention of a substantial proportion of melanoma deaths,” he maintained. However, the U.S. National Cancer Institute’s position is that current evidence is inadequate and population-based screening has potential drawbacks such as overdiagnosis.

In the meantime, many opportunities to detect melanoma during routine medical care may be missed, according to Dr. Venna; for example, auscultating a patients’ heart or lungs without pulling up their shirt may miss truncal lesions. “Use every opportunity to screen the patient – that’s opportunistic screening,” he recommended.

He also endorsed listening to patients’ concerns and balancing their requests with judicious removal of lesions, as a large share of melanomas are self-detected. In one study of note, 9 of 166 lesions removed and found to be melanoma were removed only to reassure the patient (Arch Dermatol. 2005;141:434-8).

Barriers to and imperatives for early detection

Data suggest that access to specialist care is problematic for the early detection of melanoma. In a survey of dermatologists, the median wait time for an appointment was 8 days for botulinum toxin type A (Botox) injections, compared with 26 days for evaluation of a suspicious mole (J Am Acad Dermatol. 2007;57:985-9).

“The concern of course is a delay in diagnosis and the fact that the nondermatologists will have an opportunity likely to capture melanoma before we do,” Dr. Venna commented.

Imperatives for early detection include not only better outcomes but also reduced costs, given the hefty price tag of novel targeted therapies being used to treat advanced disease, he noted.

A recent phase III trial found that progression-free survival in advanced melanoma was roughly two to four times longer with the combination of the targeted agents nivolumab (Opdivo) and ipilimumab (Yervoy) than with either agent alone (ASCO 2015, Abstract LBA1). But a discussion touching on costs noted that for the average-weight American, this combination would run about $300,000 annually (ASCO 2015, Saltz L).

Risk and protective factors

New data have implicated medications used to treat erectile dysfunction (ED) as a risk factor for melanoma, raising concern given their widespread use, Dr. Venna said.

A cohort study found that melanoma risk was 84% higher with recent use of sildenafil (Viagra) and 92% higher with ever use (JAMA Intern Med. 2014;174:964-70). But the absolute number of excess cases was fairly small.

A case-control study looking at all types of ED drugs found that receipt of a single prescription was associated with a 32% higher risk of melanoma (JAMA. 2015;313:2449-55). Use increased risk of in situ and stage I melanoma, but not more advanced disease. There was no difference between short- and long-acting agents.

“Prospective studies with clearly defined inclusions and exclusions are needed, with dosing in particular,” Dr. Venna summed up. “There appears to be a modest association, but it certainly is not enough to call for widespread discontinuation. In the patients who are on these drugs, it’s good to document [use] until we have more clarity about the biologic effects of this ultimately.”

A study using dietary surveys and having an average follow-up of more than a decade showed that coffee drinkers had a 20% lower risk of melanoma relative to nondrinkers (J Natl Cancer Inst. 2015 Jan 20;107(2)). But benefit was significant only among the subset drinking at least 4 cups a day.

“It’s premature to advise coffee intake based on this paper, although it does build upon an earlier Norwegian study from the 1990s which showed a similar trend,” Dr. Venna said. “At the end of the day, these data are intriguing, but obviously there are a lot of side effects of trying to consume 4 cups of coffee a day, so further study is certainly warranted.”

Dr. Venna disclosed that he had no relevant conflicts of interest.

SEATTLE – Recent developments in the field of melanoma may help clinicians refine their approach to this disease, according to Dr. Suraj S. Venna, a dermatologist and medical director of the Inova Melanoma and Skin Cancer Center in Fairfax, Va. He discussed several key studies and guideline revisions at a joint meeting of the Global Biomarkers Consortium and World Cutaneous Malignancies Congress.

Sentinel node biopsy for thin melanomas

Guidance regarding staging of and use of sentinel lymph node (SLN) biopsy in cases of thin melanomas, those no more than 1 mm thick, is in flux, and these gray areas pose challenges to management.

In its seventh edition of the TNM Staging System published in 2010, the American Joint Committee on Cancer started incorporating mitotic rate to substage T1 melanomas, according to Dr. Venna. Those having a mitotic rate of at least 1 mitosis/mm² are now categorized as T1b. Yet, even within this subset, there is a spectrum of 10-year survival according to mitotic rate (J Clin Oncol. 2011;29:2199-205). Also, the association of mitotic rate with regional nodal metastasis is unclear.

Therefore, “T1b is not an automatic recommendation for lymph node biopsy,” he said. Decisions about SLN in patients with such tumors should consider all available clinical and pathologic factors.

The National Comprehensive Cancer Network has also revised its guidelines in this area. In 2011, they advised that SLN could be discussed and considered in stage IA melanomas with various high-risk features and in selected stage IB melanomas based on mitotic rate and thickness.

But in their 2015 update, they gave priority to lesion depth. “What they say is other than Breslow depth, there remains little consensus regarding the significance of ulceration, mitotic rate, or lymphovascular invasion,” Dr. Venna elaborated. Now, SLN biopsy can be considered on a case-by-case basis for stage 1A melanomas that are thicker or have uncertain microstaging, and can be discussed and offered for thicker stage 1B melanomas with ulceration or an elevated mitotic rate.

“At the end of the day, when we are all in our multidisciplinary clinics trying to talk to patients about this, these are nuances that really cause a lot of angst for us and the patient in terms of making decisions,” commented Dr. Venna.

“Acknowledge the prognostic gap in the T1b cohort and that not all T1b melanomas are created equal. We all anticipate that molecular techniques may better identify who among that thin group may benefit from lymph node testing,” Dr. Venna said.

Screening strategies

Population-based screening for melanoma and other skin cancers remains controversial, and this practice is generally limited to areas of the world having high incidence, according to Dr. Venna. Germany was the first country to implement a nationwide program, starting in 2008.

“There is evidence that screening may lead to the prevention of a substantial proportion of melanoma deaths,” he maintained. However, the U.S. National Cancer Institute’s position is that current evidence is inadequate and population-based screening has potential drawbacks such as overdiagnosis.

In the meantime, many opportunities to detect melanoma during routine medical care may be missed, according to Dr. Venna; for example, auscultating a patients’ heart or lungs without pulling up their shirt may miss truncal lesions. “Use every opportunity to screen the patient – that’s opportunistic screening,” he recommended.

He also endorsed listening to patients’ concerns and balancing their requests with judicious removal of lesions, as a large share of melanomas are self-detected. In one study of note, 9 of 166 lesions removed and found to be melanoma were removed only to reassure the patient (Arch Dermatol. 2005;141:434-8).

Barriers to and imperatives for early detection

Data suggest that access to specialist care is problematic for the early detection of melanoma. In a survey of dermatologists, the median wait time for an appointment was 8 days for botulinum toxin type A (Botox) injections, compared with 26 days for evaluation of a suspicious mole (J Am Acad Dermatol. 2007;57:985-9).

“The concern of course is a delay in diagnosis and the fact that the nondermatologists will have an opportunity likely to capture melanoma before we do,” Dr. Venna commented.

Imperatives for early detection include not only better outcomes but also reduced costs, given the hefty price tag of novel targeted therapies being used to treat advanced disease, he noted.

A recent phase III trial found that progression-free survival in advanced melanoma was roughly two to four times longer with the combination of the targeted agents nivolumab (Opdivo) and ipilimumab (Yervoy) than with either agent alone (ASCO 2015, Abstract LBA1). But a discussion touching on costs noted that for the average-weight American, this combination would run about $300,000 annually (ASCO 2015, Saltz L).

Risk and protective factors

New data have implicated medications used to treat erectile dysfunction (ED) as a risk factor for melanoma, raising concern given their widespread use, Dr. Venna said.

A cohort study found that melanoma risk was 84% higher with recent use of sildenafil (Viagra) and 92% higher with ever use (JAMA Intern Med. 2014;174:964-70). But the absolute number of excess cases was fairly small.

A case-control study looking at all types of ED drugs found that receipt of a single prescription was associated with a 32% higher risk of melanoma (JAMA. 2015;313:2449-55). Use increased risk of in situ and stage I melanoma, but not more advanced disease. There was no difference between short- and long-acting agents.

“Prospective studies with clearly defined inclusions and exclusions are needed, with dosing in particular,” Dr. Venna summed up. “There appears to be a modest association, but it certainly is not enough to call for widespread discontinuation. In the patients who are on these drugs, it’s good to document [use] until we have more clarity about the biologic effects of this ultimately.”

A study using dietary surveys and having an average follow-up of more than a decade showed that coffee drinkers had a 20% lower risk of melanoma relative to nondrinkers (J Natl Cancer Inst. 2015 Jan 20;107(2)). But benefit was significant only among the subset drinking at least 4 cups a day.

“It’s premature to advise coffee intake based on this paper, although it does build upon an earlier Norwegian study from the 1990s which showed a similar trend,” Dr. Venna said. “At the end of the day, these data are intriguing, but obviously there are a lot of side effects of trying to consume 4 cups of coffee a day, so further study is certainly warranted.”

Dr. Venna disclosed that he had no relevant conflicts of interest.

SEATTLE – Recent developments in the field of melanoma may help clinicians refine their approach to this disease, according to Dr. Suraj S. Venna, a dermatologist and medical director of the Inova Melanoma and Skin Cancer Center in Fairfax, Va. He discussed several key studies and guideline revisions at a joint meeting of the Global Biomarkers Consortium and World Cutaneous Malignancies Congress.

Sentinel node biopsy for thin melanomas

Guidance regarding staging of and use of sentinel lymph node (SLN) biopsy in cases of thin melanomas, those no more than 1 mm thick, is in flux, and these gray areas pose challenges to management.

In its seventh edition of the TNM Staging System published in 2010, the American Joint Committee on Cancer started incorporating mitotic rate to substage T1 melanomas, according to Dr. Venna. Those having a mitotic rate of at least 1 mitosis/mm² are now categorized as T1b. Yet, even within this subset, there is a spectrum of 10-year survival according to mitotic rate (J Clin Oncol. 2011;29:2199-205). Also, the association of mitotic rate with regional nodal metastasis is unclear.

Therefore, “T1b is not an automatic recommendation for lymph node biopsy,” he said. Decisions about SLN in patients with such tumors should consider all available clinical and pathologic factors.

The National Comprehensive Cancer Network has also revised its guidelines in this area. In 2011, they advised that SLN could be discussed and considered in stage IA melanomas with various high-risk features and in selected stage IB melanomas based on mitotic rate and thickness.

But in their 2015 update, they gave priority to lesion depth. “What they say is other than Breslow depth, there remains little consensus regarding the significance of ulceration, mitotic rate, or lymphovascular invasion,” Dr. Venna elaborated. Now, SLN biopsy can be considered on a case-by-case basis for stage 1A melanomas that are thicker or have uncertain microstaging, and can be discussed and offered for thicker stage 1B melanomas with ulceration or an elevated mitotic rate.

“At the end of the day, when we are all in our multidisciplinary clinics trying to talk to patients about this, these are nuances that really cause a lot of angst for us and the patient in terms of making decisions,” commented Dr. Venna.

“Acknowledge the prognostic gap in the T1b cohort and that not all T1b melanomas are created equal. We all anticipate that molecular techniques may better identify who among that thin group may benefit from lymph node testing,” Dr. Venna said.

Screening strategies

Population-based screening for melanoma and other skin cancers remains controversial, and this practice is generally limited to areas of the world having high incidence, according to Dr. Venna. Germany was the first country to implement a nationwide program, starting in 2008.

“There is evidence that screening may lead to the prevention of a substantial proportion of melanoma deaths,” he maintained. However, the U.S. National Cancer Institute’s position is that current evidence is inadequate and population-based screening has potential drawbacks such as overdiagnosis.

In the meantime, many opportunities to detect melanoma during routine medical care may be missed, according to Dr. Venna; for example, auscultating a patients’ heart or lungs without pulling up their shirt may miss truncal lesions. “Use every opportunity to screen the patient – that’s opportunistic screening,” he recommended.

He also endorsed listening to patients’ concerns and balancing their requests with judicious removal of lesions, as a large share of melanomas are self-detected. In one study of note, 9 of 166 lesions removed and found to be melanoma were removed only to reassure the patient (Arch Dermatol. 2005;141:434-8).

Barriers to and imperatives for early detection

Data suggest that access to specialist care is problematic for the early detection of melanoma. In a survey of dermatologists, the median wait time for an appointment was 8 days for botulinum toxin type A (Botox) injections, compared with 26 days for evaluation of a suspicious mole (J Am Acad Dermatol. 2007;57:985-9).

“The concern of course is a delay in diagnosis and the fact that the nondermatologists will have an opportunity likely to capture melanoma before we do,” Dr. Venna commented.

Imperatives for early detection include not only better outcomes but also reduced costs, given the hefty price tag of novel targeted therapies being used to treat advanced disease, he noted.

A recent phase III trial found that progression-free survival in advanced melanoma was roughly two to four times longer with the combination of the targeted agents nivolumab (Opdivo) and ipilimumab (Yervoy) than with either agent alone (ASCO 2015, Abstract LBA1). But a discussion touching on costs noted that for the average-weight American, this combination would run about $300,000 annually (ASCO 2015, Saltz L).

Risk and protective factors

New data have implicated medications used to treat erectile dysfunction (ED) as a risk factor for melanoma, raising concern given their widespread use, Dr. Venna said.

A cohort study found that melanoma risk was 84% higher with recent use of sildenafil (Viagra) and 92% higher with ever use (JAMA Intern Med. 2014;174:964-70). But the absolute number of excess cases was fairly small.

A case-control study looking at all types of ED drugs found that receipt of a single prescription was associated with a 32% higher risk of melanoma (JAMA. 2015;313:2449-55). Use increased risk of in situ and stage I melanoma, but not more advanced disease. There was no difference between short- and long-acting agents.

“Prospective studies with clearly defined inclusions and exclusions are needed, with dosing in particular,” Dr. Venna summed up. “There appears to be a modest association, but it certainly is not enough to call for widespread discontinuation. In the patients who are on these drugs, it’s good to document [use] until we have more clarity about the biologic effects of this ultimately.”

A study using dietary surveys and having an average follow-up of more than a decade showed that coffee drinkers had a 20% lower risk of melanoma relative to nondrinkers (J Natl Cancer Inst. 2015 Jan 20;107(2)). But benefit was significant only among the subset drinking at least 4 cups a day.

“It’s premature to advise coffee intake based on this paper, although it does build upon an earlier Norwegian study from the 1990s which showed a similar trend,” Dr. Venna said. “At the end of the day, these data are intriguing, but obviously there are a lot of side effects of trying to consume 4 cups of coffee a day, so further study is certainly warranted.”

Dr. Venna disclosed that he had no relevant conflicts of interest.

SEATTLE – Recent developments in the field of melanoma may help clinicians refine their approach to this disease, according to Dr. Suraj S. Venna, a dermatologist and medical director of the Inova Melanoma and Skin Cancer Center in Fairfax, Va. He discussed several key studies and guideline revisions at a joint meeting of the Global Biomarkers Consortium and World Cutaneous Malignancies Congress.

Sentinel node biopsy for thin melanomas

Guidance regarding staging of and use of sentinel lymph node (SLN) biopsy in cases of thin melanomas, those no more than 1 mm thick, is in flux, and these gray areas pose challenges to management.

In its seventh edition of the TNM Staging System published in 2010, the American Joint Committee on Cancer started incorporating mitotic rate to substage T1 melanomas, according to Dr. Venna. Those having a mitotic rate of at least 1 mitosis/mm² are now categorized as T1b. Yet, even within this subset, there is a spectrum of 10-year survival according to mitotic rate (J Clin Oncol. 2011;29:2199-205). Also, the association of mitotic rate with regional nodal metastasis is unclear.

Therefore, “T1b is not an automatic recommendation for lymph node biopsy,” he said. Decisions about SLN in patients with such tumors should consider all available clinical and pathologic factors.

The National Comprehensive Cancer Network has also revised its guidelines in this area. In 2011, they advised that SLN could be discussed and considered in stage IA melanomas with various high-risk features and in selected stage IB melanomas based on mitotic rate and thickness.

But in their 2015 update, they gave priority to lesion depth. “What they say is other than Breslow depth, there remains little consensus regarding the significance of ulceration, mitotic rate, or lymphovascular invasion,” Dr. Venna elaborated. Now, SLN biopsy can be considered on a case-by-case basis for stage 1A melanomas that are thicker or have uncertain microstaging, and can be discussed and offered for thicker stage 1B melanomas with ulceration or an elevated mitotic rate.

“At the end of the day, when we are all in our multidisciplinary clinics trying to talk to patients about this, these are nuances that really cause a lot of angst for us and the patient in terms of making decisions,” commented Dr. Venna.

“Acknowledge the prognostic gap in the T1b cohort and that not all T1b melanomas are created equal. We all anticipate that molecular techniques may better identify who among that thin group may benefit from lymph node testing,” Dr. Venna said.

Screening strategies

Population-based screening for melanoma and other skin cancers remains controversial, and this practice is generally limited to areas of the world having high incidence, according to Dr. Venna. Germany was the first country to implement a nationwide program, starting in 2008.

“There is evidence that screening may lead to the prevention of a substantial proportion of melanoma deaths,” he maintained. However, the U.S. National Cancer Institute’s position is that current evidence is inadequate and population-based screening has potential drawbacks such as overdiagnosis.

In the meantime, many opportunities to detect melanoma during routine medical care may be missed, according to Dr. Venna; for example, auscultating a patients’ heart or lungs without pulling up their shirt may miss truncal lesions. “Use every opportunity to screen the patient – that’s opportunistic screening,” he recommended.

He also endorsed listening to patients’ concerns and balancing their requests with judicious removal of lesions, as a large share of melanomas are self-detected. In one study of note, 9 of 166 lesions removed and found to be melanoma were removed only to reassure the patient (Arch Dermatol. 2005;141:434-8).

Barriers to and imperatives for early detection

Data suggest that access to specialist care is problematic for the early detection of melanoma. In a survey of dermatologists, the median wait time for an appointment was 8 days for botulinum toxin type A (Botox) injections, compared with 26 days for evaluation of a suspicious mole (J Am Acad Dermatol. 2007;57:985-9).

“The concern of course is a delay in diagnosis and the fact that the nondermatologists will have an opportunity likely to capture melanoma before we do,” Dr. Venna commented.

Imperatives for early detection include not only better outcomes but also reduced costs, given the hefty price tag of novel targeted therapies being used to treat advanced disease, he noted.

A recent phase III trial found that progression-free survival in advanced melanoma was roughly two to four times longer with the combination of the targeted agents nivolumab (Opdivo) and ipilimumab (Yervoy) than with either agent alone (ASCO 2015, Abstract LBA1). But a discussion touching on costs noted that for the average-weight American, this combination would run about $300,000 annually (ASCO 2015, Saltz L).

Risk and protective factors

New data have implicated medications used to treat erectile dysfunction (ED) as a risk factor for melanoma, raising concern given their widespread use, Dr. Venna said.

A cohort study found that melanoma risk was 84% higher with recent use of sildenafil (Viagra) and 92% higher with ever use (JAMA Intern Med. 2014;174:964-70). But the absolute number of excess cases was fairly small.

A case-control study looking at all types of ED drugs found that receipt of a single prescription was associated with a 32% higher risk of melanoma (JAMA. 2015;313:2449-55). Use increased risk of in situ and stage I melanoma, but not more advanced disease. There was no difference between short- and long-acting agents.

“Prospective studies with clearly defined inclusions and exclusions are needed, with dosing in particular,” Dr. Venna summed up. “There appears to be a modest association, but it certainly is not enough to call for widespread discontinuation. In the patients who are on these drugs, it’s good to document [use] until we have more clarity about the biologic effects of this ultimately.”

A study using dietary surveys and having an average follow-up of more than a decade showed that coffee drinkers had a 20% lower risk of melanoma relative to nondrinkers (J Natl Cancer Inst. 2015 Jan 20;107(2)). But benefit was significant only among the subset drinking at least 4 cups a day.

“It’s premature to advise coffee intake based on this paper, although it does build upon an earlier Norwegian study from the 1990s which showed a similar trend,” Dr. Venna said. “At the end of the day, these data are intriguing, but obviously there are a lot of side effects of trying to consume 4 cups of coffee a day, so further study is certainly warranted.”

Dr. Venna disclosed that he had no relevant conflicts of interest.

SEATTLE – A new biomarker for bronchial epithelium that helps identify smokers with suspicious lesions who have lung cancer is now ready for clinical use. And one for nasal epithelium that could be used for screening may not be far behind.

“There is clearly a critical unmet need to develop molecular biomarkers to address some of the challenges that we now face since we have instituted CT screening for lung cancer,” Dr. Avi Spira said at a joint meeting of the Global Biomarkers Consortium and World Cutaneous Malignancies Congress.

Although the National Lung Screening Trial established that annual chest CT among high-risk current and former smokers reduces their risk of death from lung cancer (N Engl J Med. 2011;365:395-409), the vast majority of those who screen positive do not have lung cancer. Also, screening only patients who meet criteria set by the trial will pick up less than half of all lung cancers in the United States.

“That leads to two critical unmet needs for molecular biomarkers in the so-called post–National Lung Screening Trial era,” said Dr. Spira, professor of medicine, pathology and laboratory medicine, and bioinformatics; chief of the division of computational biomedicine; and director of the translational bioinformatics program, Clinical and Translational Science Institute, all at Boston University.

“The first is … we desperately need molecular biomarkers that can distinguish a benign nodule found on CT versus a malignant one,” he said. “The second and arguably longer-term biomarker that we need is to distinguish which smokers would benefit from CT screening annually.”

Much of his team’s research in this area builds on the concept of field of injury. “The idea here is if you smoke, even though lung cancer tends to develop deep within the parenchyma of your lung, all of the epithelial cells that line your respiratory tract have genomic alterations that reflect the presence of that cancer,” Dr. Spira explained. Thus, profiling epithelial cells anywhere in the airway could be used for early detection and risk assessment.

He and his colleagues developed a 23-gene signature for use on bronchial epithelial cells. The biomarker was validated in the Airway Epithelium Gene Expression In the Diagnosis of Lung Cancer (AEGIS) 1 and 2 trials among 639 current and former smokers undergoing bronchoscopy for suspicious nodules seen on CT.

With 1 year of follow-up, biomarker sensitivity was 88%-89%, while specificity was 47% (N Engl J Med. 2015;373:243-251). “However the negative predictive value, which is really what drives the clinical utility of this test, is above 90%. And that’s what we believe will drive physicians to use the test – [determining] who can they avoid sending for an unnecessary [biopsy] procedure,” Dr. Spira said. Bronchoscopy alone had sensitivity of about 75%, but bronchoscopy combined with the gene signature had sensitivity of 97%.

Subgroup analyses showed the biomarker had superior sensitivity for detecting lung cancer when lesions measured no more than 3 cm or were located in the lung periphery, and when patients had early-stage disease. In addition, it performed similarly well across different types of tumors.

Of special note, among patients whose pretest probability of cancer fell in the intermediate range (10%-60%), bronchoscopy had an 83% nondiagnostic rate, but the biomarker had 88% sensitivity and a 91% negative predictive value. “That means if you have a nondiagnostic bronchoscopy in a patient who is at intermediate pretest risk for disease, a negative gene expression test would mean there is a less than 10% chance this is cancer. That’s where a physician might feel, okay, I don’t have to go on and do a biopsy, I can watch that patient serially with CT scans of the chest,” Dr. Spira said.

The biomarker test is now clinically available (Percepta, manufactured by Veracyte). “I think it’s exciting because it’s the first of what I believe are many molecular biomarkers that are going to be emerging in the clinical space for the early detection of lung cancer,” he said.

“The multimillion dollar question is why are we seeing gene expression changes in normal-appearing cells so far away from where the tumor arises? We don’t have the full answer to that yet, but based on the genes that are changing, we have developed some hypotheses,” Dr. Spira said.

Some of the down-regulated genes are involved in antioxidant and DNA repair pathways, suggesting that the smokers who ultimately get cancer have less of a protective response to smoking. And some of the up-regulated genes include ones in the PI3 kinase signaling pathway.

“I would argue that what we are seeing in the proximal airway isn’t necessarily reflecting the presence of the cancer but the susceptibility, and that’s a really important distinguishing factor because then perhaps the test could be used as a screening tool,” Dr. Spira maintained.

As not all smokers at elevated risk for lung cancer will undergo bronchoscopy, one of the investigators’ future goals is to move biomarker testing to a less invasive site. They are currently focusing on the nose, using nasal epithelium collected by brushings from the inferior turbinate.

An analysis of nasal epithelium collected at the time of bronchoscopy in the AEGIS trials has shown that a 200-gene signature performs well for distinguishing between patients with and without lung cancer, Dr. Spira reported. Furthermore, the changes in gene expression profile in the nose were similar to those seen in the bronchus.

Such a biomarker might have best clinical utility in two other settings, he proposed. The first would be in patients having nodules that are clearly not accessible by bronchoscopy, in which case the biomarker would be applied for diagnosis. The second would be in smokers being seen for routine annual exams, in which case it would be used to identify those who should have CT surveillance.

“We are hopeful that the nasal epithelium can serve as a less invasive surrogate for the bronchus and ultimately allow us to move airway profiling into the screening setting for lung cancer in the longer term,” he concluded.

Dr. Spira disclosed that he receives intellectual property rights and consulting fees from, and has an ownership interest in, Veracyte Inc.

SEATTLE – A new biomarker for bronchial epithelium that helps identify smokers with suspicious lesions who have lung cancer is now ready for clinical use. And one for nasal epithelium that could be used for screening may not be far behind.

“There is clearly a critical unmet need to develop molecular biomarkers to address some of the challenges that we now face since we have instituted CT screening for lung cancer,” Dr. Avi Spira said at a joint meeting of the Global Biomarkers Consortium and World Cutaneous Malignancies Congress.

Although the National Lung Screening Trial established that annual chest CT among high-risk current and former smokers reduces their risk of death from lung cancer (N Engl J Med. 2011;365:395-409), the vast majority of those who screen positive do not have lung cancer. Also, screening only patients who meet criteria set by the trial will pick up less than half of all lung cancers in the United States.

“That leads to two critical unmet needs for molecular biomarkers in the so-called post–National Lung Screening Trial era,” said Dr. Spira, professor of medicine, pathology and laboratory medicine, and bioinformatics; chief of the division of computational biomedicine; and director of the translational bioinformatics program, Clinical and Translational Science Institute, all at Boston University.

“The first is … we desperately need molecular biomarkers that can distinguish a benign nodule found on CT versus a malignant one,” he said. “The second and arguably longer-term biomarker that we need is to distinguish which smokers would benefit from CT screening annually.”

Much of his team’s research in this area builds on the concept of field of injury. “The idea here is if you smoke, even though lung cancer tends to develop deep within the parenchyma of your lung, all of the epithelial cells that line your respiratory tract have genomic alterations that reflect the presence of that cancer,” Dr. Spira explained. Thus, profiling epithelial cells anywhere in the airway could be used for early detection and risk assessment.

He and his colleagues developed a 23-gene signature for use on bronchial epithelial cells. The biomarker was validated in the Airway Epithelium Gene Expression In the Diagnosis of Lung Cancer (AEGIS) 1 and 2 trials among 639 current and former smokers undergoing bronchoscopy for suspicious nodules seen on CT.

With 1 year of follow-up, biomarker sensitivity was 88%-89%, while specificity was 47% (N Engl J Med. 2015;373:243-251). “However the negative predictive value, which is really what drives the clinical utility of this test, is above 90%. And that’s what we believe will drive physicians to use the test – [determining] who can they avoid sending for an unnecessary [biopsy] procedure,” Dr. Spira said. Bronchoscopy alone had sensitivity of about 75%, but bronchoscopy combined with the gene signature had sensitivity of 97%.

Subgroup analyses showed the biomarker had superior sensitivity for detecting lung cancer when lesions measured no more than 3 cm or were located in the lung periphery, and when patients had early-stage disease. In addition, it performed similarly well across different types of tumors.

Of special note, among patients whose pretest probability of cancer fell in the intermediate range (10%-60%), bronchoscopy had an 83% nondiagnostic rate, but the biomarker had 88% sensitivity and a 91% negative predictive value. “That means if you have a nondiagnostic bronchoscopy in a patient who is at intermediate pretest risk for disease, a negative gene expression test would mean there is a less than 10% chance this is cancer. That’s where a physician might feel, okay, I don’t have to go on and do a biopsy, I can watch that patient serially with CT scans of the chest,” Dr. Spira said.

The biomarker test is now clinically available (Percepta, manufactured by Veracyte). “I think it’s exciting because it’s the first of what I believe are many molecular biomarkers that are going to be emerging in the clinical space for the early detection of lung cancer,” he said.

“The multimillion dollar question is why are we seeing gene expression changes in normal-appearing cells so far away from where the tumor arises? We don’t have the full answer to that yet, but based on the genes that are changing, we have developed some hypotheses,” Dr. Spira said.

Some of the down-regulated genes are involved in antioxidant and DNA repair pathways, suggesting that the smokers who ultimately get cancer have less of a protective response to smoking. And some of the up-regulated genes include ones in the PI3 kinase signaling pathway.

“I would argue that what we are seeing in the proximal airway isn’t necessarily reflecting the presence of the cancer but the susceptibility, and that’s a really important distinguishing factor because then perhaps the test could be used as a screening tool,” Dr. Spira maintained.

As not all smokers at elevated risk for lung cancer will undergo bronchoscopy, one of the investigators’ future goals is to move biomarker testing to a less invasive site. They are currently focusing on the nose, using nasal epithelium collected by brushings from the inferior turbinate.

An analysis of nasal epithelium collected at the time of bronchoscopy in the AEGIS trials has shown that a 200-gene signature performs well for distinguishing between patients with and without lung cancer, Dr. Spira reported. Furthermore, the changes in gene expression profile in the nose were similar to those seen in the bronchus.

Such a biomarker might have best clinical utility in two other settings, he proposed. The first would be in patients having nodules that are clearly not accessible by bronchoscopy, in which case the biomarker would be applied for diagnosis. The second would be in smokers being seen for routine annual exams, in which case it would be used to identify those who should have CT surveillance.

“We are hopeful that the nasal epithelium can serve as a less invasive surrogate for the bronchus and ultimately allow us to move airway profiling into the screening setting for lung cancer in the longer term,” he concluded.

Dr. Spira disclosed that he receives intellectual property rights and consulting fees from, and has an ownership interest in, Veracyte Inc.

SEATTLE – A new biomarker for bronchial epithelium that helps identify smokers with suspicious lesions who have lung cancer is now ready for clinical use. And one for nasal epithelium that could be used for screening may not be far behind.

“There is clearly a critical unmet need to develop molecular biomarkers to address some of the challenges that we now face since we have instituted CT screening for lung cancer,” Dr. Avi Spira said at a joint meeting of the Global Biomarkers Consortium and World Cutaneous Malignancies Congress.

Although the National Lung Screening Trial established that annual chest CT among high-risk current and former smokers reduces their risk of death from lung cancer (N Engl J Med. 2011;365:395-409), the vast majority of those who screen positive do not have lung cancer. Also, screening only patients who meet criteria set by the trial will pick up less than half of all lung cancers in the United States.

“That leads to two critical unmet needs for molecular biomarkers in the so-called post–National Lung Screening Trial era,” said Dr. Spira, professor of medicine, pathology and laboratory medicine, and bioinformatics; chief of the division of computational biomedicine; and director of the translational bioinformatics program, Clinical and Translational Science Institute, all at Boston University.

“The first is … we desperately need molecular biomarkers that can distinguish a benign nodule found on CT versus a malignant one,” he said. “The second and arguably longer-term biomarker that we need is to distinguish which smokers would benefit from CT screening annually.”

Much of his team’s research in this area builds on the concept of field of injury. “The idea here is if you smoke, even though lung cancer tends to develop deep within the parenchyma of your lung, all of the epithelial cells that line your respiratory tract have genomic alterations that reflect the presence of that cancer,” Dr. Spira explained. Thus, profiling epithelial cells anywhere in the airway could be used for early detection and risk assessment.

He and his colleagues developed a 23-gene signature for use on bronchial epithelial cells. The biomarker was validated in the Airway Epithelium Gene Expression In the Diagnosis of Lung Cancer (AEGIS) 1 and 2 trials among 639 current and former smokers undergoing bronchoscopy for suspicious nodules seen on CT.

With 1 year of follow-up, biomarker sensitivity was 88%-89%, while specificity was 47% (N Engl J Med. 2015;373:243-251). “However the negative predictive value, which is really what drives the clinical utility of this test, is above 90%. And that’s what we believe will drive physicians to use the test – [determining] who can they avoid sending for an unnecessary [biopsy] procedure,” Dr. Spira said. Bronchoscopy alone had sensitivity of about 75%, but bronchoscopy combined with the gene signature had sensitivity of 97%.

Subgroup analyses showed the biomarker had superior sensitivity for detecting lung cancer when lesions measured no more than 3 cm or were located in the lung periphery, and when patients had early-stage disease. In addition, it performed similarly well across different types of tumors.

Of special note, among patients whose pretest probability of cancer fell in the intermediate range (10%-60%), bronchoscopy had an 83% nondiagnostic rate, but the biomarker had 88% sensitivity and a 91% negative predictive value. “That means if you have a nondiagnostic bronchoscopy in a patient who is at intermediate pretest risk for disease, a negative gene expression test would mean there is a less than 10% chance this is cancer. That’s where a physician might feel, okay, I don’t have to go on and do a biopsy, I can watch that patient serially with CT scans of the chest,” Dr. Spira said.

The biomarker test is now clinically available (Percepta, manufactured by Veracyte). “I think it’s exciting because it’s the first of what I believe are many molecular biomarkers that are going to be emerging in the clinical space for the early detection of lung cancer,” he said.

“The multimillion dollar question is why are we seeing gene expression changes in normal-appearing cells so far away from where the tumor arises? We don’t have the full answer to that yet, but based on the genes that are changing, we have developed some hypotheses,” Dr. Spira said.

Some of the down-regulated genes are involved in antioxidant and DNA repair pathways, suggesting that the smokers who ultimately get cancer have less of a protective response to smoking. And some of the up-regulated genes include ones in the PI3 kinase signaling pathway.

“I would argue that what we are seeing in the proximal airway isn’t necessarily reflecting the presence of the cancer but the susceptibility, and that’s a really important distinguishing factor because then perhaps the test could be used as a screening tool,” Dr. Spira maintained.

As not all smokers at elevated risk for lung cancer will undergo bronchoscopy, one of the investigators’ future goals is to move biomarker testing to a less invasive site. They are currently focusing on the nose, using nasal epithelium collected by brushings from the inferior turbinate.

An analysis of nasal epithelium collected at the time of bronchoscopy in the AEGIS trials has shown that a 200-gene signature performs well for distinguishing between patients with and without lung cancer, Dr. Spira reported. Furthermore, the changes in gene expression profile in the nose were similar to those seen in the bronchus.

Such a biomarker might have best clinical utility in two other settings, he proposed. The first would be in patients having nodules that are clearly not accessible by bronchoscopy, in which case the biomarker would be applied for diagnosis. The second would be in smokers being seen for routine annual exams, in which case it would be used to identify those who should have CT surveillance.

“We are hopeful that the nasal epithelium can serve as a less invasive surrogate for the bronchus and ultimately allow us to move airway profiling into the screening setting for lung cancer in the longer term,” he concluded.

Dr. Spira disclosed that he receives intellectual property rights and consulting fees from, and has an ownership interest in, Veracyte Inc.