More than 1,000 patients with TNBC, ER+ breast cancer, or ovarian cancer from five cohorts were examined for genomic instability scores (GIS) and the presence of BRCA deficiency, which showed that, while GIS was similar in BRCA-deficient TNBC and ovarian cancer, it was significantly different in ER+ breast cancer.
The analysis, presented at the San Antonio Breast Cancer Symposium, showed that the genomic instability scores threshold, which could be used to dictate a patient’s treatment, should be lower for ER+ breast cancer than for TNBC.
“This indicates that different GIS thresholds are appropriate for breast cancer subtypes, and that the GIS threshold developed for ovarian caner is not appropriate for ER+ breast cancer,” said lead author Kirsten Timms, PhD, from Myriad Genetics.
This, she noted, is “consistent with the fact that ovarian cancer and TNBC are known to have similar molecular signatures.”
The researchers suggest that the “more inclusive” thresholds assessed in the study should be examined in further studies “to determine whether these cutoffs are associated with a benefit from treatment with DNA-targeting agents,” such as poly (ADP-ribose) polymerase (PARP) inhibitors.
Thomas P. Slavin, MD, chief medical officer at Myriad Genetics, said in an interview that there is “not a one size fits all” for GIS thresholds.
“When you look at ER+ breast cancer you see we need a different cutoff because it’s probably not as driven by homologous recombination deficiency [HRD], as least as a whole, compared to the other two,” he said. “There’s a little less genomic instability.”
He continued that their results suggest around half of TNBC patients have a GIS score that indicates the presence of significant HRD, which is “spot on for what we see with ovarian cancer” and “those people should respond pretty well to PARP-inhibitor therapies,” which is currently being investigated in clinical trials.
“But even in the ER+ group, when we look at the thresholds we used in this research, still about a third have what looks like a substantial amount of HRD, so that’s a huge biomarker,” Dr. Slavin said.
He explained that the importance of their score is that, rather than looking for the causes of HRD, they are looking for the consequences.
“We don’t know all the causes of why, all of a sudden, a tumor cell looks like it can’t replicate through homologous combination [but] what this test does is it says: ‘We don’t really care what the cause is ... we can just look at the genomic scarring and the consequences.’ ”
Elena Provenzano, MD, CRUK Therapeutic Discovery Laboratories, Cambridge (England) University Hospitals NHS Foundation Trust, who was not involved in the study, said in an interview it is “interesting work.”
“We have a personalized breast cancer program here in Cambridge and we’re running trials where we use PARP inhibitors and platinum-based therapy, and what we’re using to make these sorts of decisions is COSMIC mutational signatures associated with genetic instability. And I guess we also look at the total mutational burden,” Dr. Provenzano said.
She continued that the GIS is one of several ways of measuring HRD. “So the question is how it compares with the other measures that are being used to assess whether or not patients are suitable for PARP inhibitor and platinum-based therapy.”
Dr. Provenzano underlined that it has been known since the “early 2000s” that breast cancer is a group of different diseases. “Even within those categories there’s quite a lot of tumor types,” so it “makes sense you need to adjust the threshold slightly for it to become relevant to types of breast cancer.”
She added that the “holy grail in oncology is this concept of personalized medicine, so all these tests help us make sure that the right patient is getting the right treatment.
“At the moment TNBC is often getting treated in a similar way, although we know that there are different biological subtypes, so while there’s a significant group that falls into this BRCA-deficient group that are going to respond to PARPs there are other types that don’t.
“So these sorts of tests help us decide which subset are going to help us the most, and for the others ones we potentially need to identify other treatments as being optimal,” Dr. Provenzano said.
Previous studies have shown that HR-deficient tumors may benefit from treatment with DNA-damaging agents, and that tools such as the three-biomarker GIS can be used to identify HR deficiency.
The Food and Drug Administration has already approved a GIS threshold for identifying HR deficiency in ovarian cancer of 42, which was set as the 5th percentile for BRCA-deficient tumors. However, a recent published in Molecular Cancer Research, and a second published on MDPI Open Access Journals, indicated that a lower, first percentile, cutoff of at least 33 was associated with improved outcomes after platinum-based treatment.
As TNBC is known to have a similar molecular profile to ovarian cancer, the researchers investigated whether it has a different GIS threshold to that in ER+ breast cancer, gathering data on patients newly diagnosed with ovarian cancer, TNBC, or ER+ breast cancer from across five cohorts.
They included 127 ovarian cancer patients from Nature, 434 ovarian cancer, 44 TNBC, and 213 ER+ breast cancer patients from The Cancer Genome Atlas, 55 TNBC and 112 ER+ breast cancer patients from Breast Cancer Research, 19 TNBC and 25 ER+ breast cancer patients from TBCRC 008, and 56 ER+ breast cancer patients from OlympiAD.
GIS was defined as a combination of loss of heterozygosity, telomeric-allellic imbalance, and large-scale state transitions, identified through next-generation sequencing, and GIS distributions were compared between cancer types and subtypes.
The team also determined the presence of BRCA deficiency, finding that, among BRCA deficient tumors, the GIS distribution among patients with ER+ breast cancer was significantly different from that seen in both ovarian cancer (P = 9.6 x 10–5) and TNBC (P = 2.1 x 10–4).
The first percentile of a normal distribution of BRCA-deficient ER+ breast cancers indicated a GIS threshold of 24, with 45.1% of all ER+ tumors at or above this threshold found to be GIS positive. This translated into 98.7% of BRCA-deficient tumors and 32.7% that were BRCA intact.
The results also showed, however, that the GIS distribution for TNBC was not significantly different from that seen in ovarian cancer (P = .72), with the threshold of at least 33 Identifying 64.4% of TNBC tumors as GIS positive. This equated to 100% of BRCA-deficient tumors and 41.7% that were BRCA intact.
Dr. Timms and Dr. Slavin are employed by Myriad Genetics, who funded the study.