Article Type
Changed
Tue, 05/03/2022 - 15:08

Individualized 10-year and lifetime risks of cancer can now for the first time be estimated in patients with established cardiovascular disease, Cilie C. van ’t Klooster, MD, reported at the virtual annual congress of the European Society of Cardiology.

©sripfoto/Thinkstock.com

She and her coinvestigators have developed an easy-to-use predictive model that generates individualized risk estimates for total cancer, lung cancer, and colorectal cancer. The tool relies on nine readily available clinical variables: age, sex, smoking, weight, height, alcohol use, diabetes, antiplatelet drug use, and C-reactive protein level. The cancer risk calculator factors in an individual’s competing risk of death because of cardiovascular disease (CVD).

The risk calculator was developed using data on 7,280 patients with established CVD enrolled in the ongoing long-term Dutch UCC-SMART (Utrecht Cardiovascular Cohort – Second Manifestations of Arterial Disease) study, then independently validated in 9,322 patients in the double-blind CANTOS (Canakinumab Anti-Inflammatory Thrombosis Outcomes) trial, explained Dr. van ’t Klooster of Utrecht (the Netherlands) University.

Several other prediction models estimate the risk of a specific type of cancer, most commonly breast cancer or lung cancer. But the new Utrecht prediction tool is the first one to estimate total cancer risk. It’s also the first to apply specifically to patients with known CVD, thus filling an unmet need, because patients with established CVD are known to be on average at 19% increased risk of total cancer and 56% greater risk for lung cancer, compared with the general population. This is thought to be caused mainly by shared risk factors, including smoking, obesity, and low-grade systemic inflammation.

As the Utrecht/CANTOS analysis shows, however, that 19% increased relative risk for cancer in patients with CVD doesn’t tell the whole story. While the median lifetime and 10-year risks of total cancer in CANTOS were 26% and 10%, respectively, the individual patient risks for total cancer estimated using the Dutch prediction model ranged from 1% to 52% for lifetime and from 1% to 31% for 10-year risk. The same was true for lung cancer risk: median 5% lifetime and 2% 10-year risks, with individual patient risks ranging from 0% to 37% and from 0% to 24%. Likewise for colorectal cancer: a median 4% lifetime risk, ranging from 0% to 6%, and a median 2% risk over the next 10 years, with personalized risks ranging as high as 13% for lifetime risk and 6% for 10-year colorectal cancer risk.

The risk calculator performed “reasonably well,” according to Dr. van ’t Klooster. She pointed to a C-statistic of 0.74 for lung cancer, 0.63 for total cancer, and 0.64 for colorectal cancer. It’s possible the risk predictor’s performance could be further enhanced by incorporation of several potentially important factors that weren’t available in the UCC-SMART derivation cohort, including race, education level, and socioeconomic status, she added.

Potential applications for the risk calculator in clinical practice require further study, but include using the lifetime risk prediction for cancer as a motivational aid in conversations with patients about the importance of behavioral change in support of a healthier lifestyle. Also, a high predicted 10-year lung cancer risk could potentially be used to lower the threshold for a screening chest CT, resulting in earlier detection and treatment of lung cancer, Dr. van ’t Klooster noted.

In an interview, Bonnie Ky, MD, MSCE, praised the risk prediction study as rigorously executed, topical, and clinically significant.

“This paper signifies the overlap between our two disciplines of cancer and cardiovascular disease in terms of the risks that we face together when we care for this patient population,” said Dr. Ky, a cardiologist at the University of Pennsylvania, Philadelphia.

“Many of us in medicine believe in the importance of risk prediction: identifying who’s at high risk and doing everything we can to mitigate that risk. This paper speaks to that and moves us one step closer to accomplishing that aim,” added Dr. Ky, who is editor in chief of JACC: CardioOncology, which published the study simultaneously with Dr. van ’t Klooster’s presentation at ESC 2020. The paper provides direct access to the risk calculator.

Dr. van ’t Klooster reported having no financial conflicts regarding her study. UCC-SMART is funded by a Utrecht University grant, and CANTOS was funded by Novartis.

SOURCE: van ’t Klooster CC. ESC 2020 and JACC CardioOncol. 2020 Aug. doi: 10.1016/j.jaccao.2020.07.001.

Meeting/Event
Publications
Topics
Sections
Meeting/Event
Meeting/Event

Individualized 10-year and lifetime risks of cancer can now for the first time be estimated in patients with established cardiovascular disease, Cilie C. van ’t Klooster, MD, reported at the virtual annual congress of the European Society of Cardiology.

©sripfoto/Thinkstock.com

She and her coinvestigators have developed an easy-to-use predictive model that generates individualized risk estimates for total cancer, lung cancer, and colorectal cancer. The tool relies on nine readily available clinical variables: age, sex, smoking, weight, height, alcohol use, diabetes, antiplatelet drug use, and C-reactive protein level. The cancer risk calculator factors in an individual’s competing risk of death because of cardiovascular disease (CVD).

The risk calculator was developed using data on 7,280 patients with established CVD enrolled in the ongoing long-term Dutch UCC-SMART (Utrecht Cardiovascular Cohort – Second Manifestations of Arterial Disease) study, then independently validated in 9,322 patients in the double-blind CANTOS (Canakinumab Anti-Inflammatory Thrombosis Outcomes) trial, explained Dr. van ’t Klooster of Utrecht (the Netherlands) University.

Several other prediction models estimate the risk of a specific type of cancer, most commonly breast cancer or lung cancer. But the new Utrecht prediction tool is the first one to estimate total cancer risk. It’s also the first to apply specifically to patients with known CVD, thus filling an unmet need, because patients with established CVD are known to be on average at 19% increased risk of total cancer and 56% greater risk for lung cancer, compared with the general population. This is thought to be caused mainly by shared risk factors, including smoking, obesity, and low-grade systemic inflammation.

As the Utrecht/CANTOS analysis shows, however, that 19% increased relative risk for cancer in patients with CVD doesn’t tell the whole story. While the median lifetime and 10-year risks of total cancer in CANTOS were 26% and 10%, respectively, the individual patient risks for total cancer estimated using the Dutch prediction model ranged from 1% to 52% for lifetime and from 1% to 31% for 10-year risk. The same was true for lung cancer risk: median 5% lifetime and 2% 10-year risks, with individual patient risks ranging from 0% to 37% and from 0% to 24%. Likewise for colorectal cancer: a median 4% lifetime risk, ranging from 0% to 6%, and a median 2% risk over the next 10 years, with personalized risks ranging as high as 13% for lifetime risk and 6% for 10-year colorectal cancer risk.

The risk calculator performed “reasonably well,” according to Dr. van ’t Klooster. She pointed to a C-statistic of 0.74 for lung cancer, 0.63 for total cancer, and 0.64 for colorectal cancer. It’s possible the risk predictor’s performance could be further enhanced by incorporation of several potentially important factors that weren’t available in the UCC-SMART derivation cohort, including race, education level, and socioeconomic status, she added.

Potential applications for the risk calculator in clinical practice require further study, but include using the lifetime risk prediction for cancer as a motivational aid in conversations with patients about the importance of behavioral change in support of a healthier lifestyle. Also, a high predicted 10-year lung cancer risk could potentially be used to lower the threshold for a screening chest CT, resulting in earlier detection and treatment of lung cancer, Dr. van ’t Klooster noted.

In an interview, Bonnie Ky, MD, MSCE, praised the risk prediction study as rigorously executed, topical, and clinically significant.

“This paper signifies the overlap between our two disciplines of cancer and cardiovascular disease in terms of the risks that we face together when we care for this patient population,” said Dr. Ky, a cardiologist at the University of Pennsylvania, Philadelphia.

“Many of us in medicine believe in the importance of risk prediction: identifying who’s at high risk and doing everything we can to mitigate that risk. This paper speaks to that and moves us one step closer to accomplishing that aim,” added Dr. Ky, who is editor in chief of JACC: CardioOncology, which published the study simultaneously with Dr. van ’t Klooster’s presentation at ESC 2020. The paper provides direct access to the risk calculator.

Dr. van ’t Klooster reported having no financial conflicts regarding her study. UCC-SMART is funded by a Utrecht University grant, and CANTOS was funded by Novartis.

SOURCE: van ’t Klooster CC. ESC 2020 and JACC CardioOncol. 2020 Aug. doi: 10.1016/j.jaccao.2020.07.001.

Individualized 10-year and lifetime risks of cancer can now for the first time be estimated in patients with established cardiovascular disease, Cilie C. van ’t Klooster, MD, reported at the virtual annual congress of the European Society of Cardiology.

©sripfoto/Thinkstock.com

She and her coinvestigators have developed an easy-to-use predictive model that generates individualized risk estimates for total cancer, lung cancer, and colorectal cancer. The tool relies on nine readily available clinical variables: age, sex, smoking, weight, height, alcohol use, diabetes, antiplatelet drug use, and C-reactive protein level. The cancer risk calculator factors in an individual’s competing risk of death because of cardiovascular disease (CVD).

The risk calculator was developed using data on 7,280 patients with established CVD enrolled in the ongoing long-term Dutch UCC-SMART (Utrecht Cardiovascular Cohort – Second Manifestations of Arterial Disease) study, then independently validated in 9,322 patients in the double-blind CANTOS (Canakinumab Anti-Inflammatory Thrombosis Outcomes) trial, explained Dr. van ’t Klooster of Utrecht (the Netherlands) University.

Several other prediction models estimate the risk of a specific type of cancer, most commonly breast cancer or lung cancer. But the new Utrecht prediction tool is the first one to estimate total cancer risk. It’s also the first to apply specifically to patients with known CVD, thus filling an unmet need, because patients with established CVD are known to be on average at 19% increased risk of total cancer and 56% greater risk for lung cancer, compared with the general population. This is thought to be caused mainly by shared risk factors, including smoking, obesity, and low-grade systemic inflammation.

As the Utrecht/CANTOS analysis shows, however, that 19% increased relative risk for cancer in patients with CVD doesn’t tell the whole story. While the median lifetime and 10-year risks of total cancer in CANTOS were 26% and 10%, respectively, the individual patient risks for total cancer estimated using the Dutch prediction model ranged from 1% to 52% for lifetime and from 1% to 31% for 10-year risk. The same was true for lung cancer risk: median 5% lifetime and 2% 10-year risks, with individual patient risks ranging from 0% to 37% and from 0% to 24%. Likewise for colorectal cancer: a median 4% lifetime risk, ranging from 0% to 6%, and a median 2% risk over the next 10 years, with personalized risks ranging as high as 13% for lifetime risk and 6% for 10-year colorectal cancer risk.

The risk calculator performed “reasonably well,” according to Dr. van ’t Klooster. She pointed to a C-statistic of 0.74 for lung cancer, 0.63 for total cancer, and 0.64 for colorectal cancer. It’s possible the risk predictor’s performance could be further enhanced by incorporation of several potentially important factors that weren’t available in the UCC-SMART derivation cohort, including race, education level, and socioeconomic status, she added.

Potential applications for the risk calculator in clinical practice require further study, but include using the lifetime risk prediction for cancer as a motivational aid in conversations with patients about the importance of behavioral change in support of a healthier lifestyle. Also, a high predicted 10-year lung cancer risk could potentially be used to lower the threshold for a screening chest CT, resulting in earlier detection and treatment of lung cancer, Dr. van ’t Klooster noted.

In an interview, Bonnie Ky, MD, MSCE, praised the risk prediction study as rigorously executed, topical, and clinically significant.

“This paper signifies the overlap between our two disciplines of cancer and cardiovascular disease in terms of the risks that we face together when we care for this patient population,” said Dr. Ky, a cardiologist at the University of Pennsylvania, Philadelphia.

“Many of us in medicine believe in the importance of risk prediction: identifying who’s at high risk and doing everything we can to mitigate that risk. This paper speaks to that and moves us one step closer to accomplishing that aim,” added Dr. Ky, who is editor in chief of JACC: CardioOncology, which published the study simultaneously with Dr. van ’t Klooster’s presentation at ESC 2020. The paper provides direct access to the risk calculator.

Dr. van ’t Klooster reported having no financial conflicts regarding her study. UCC-SMART is funded by a Utrecht University grant, and CANTOS was funded by Novartis.

SOURCE: van ’t Klooster CC. ESC 2020 and JACC CardioOncol. 2020 Aug. doi: 10.1016/j.jaccao.2020.07.001.

Publications
Publications
Topics
Article Type
Sections
Article Source

FROM ESC CONGRESS 2020

Disallow All Ads
Content Gating
No Gating (article Unlocked/Free)
Alternative CME
Disqus Comments
Default
Use ProPublica
Hide sidebar & use full width
render the right sidebar.
Conference Recap Checkbox
Not Conference Recap
Clinical Edge
Display the Slideshow in this Article