The lifetime and 10-year risk of total cancer, colorectal cancer and lung cancer can be estimated "reasonably well" with available clinical predictors in patients with cardiovascular disease, according to a study during ESC Congress 2020 and simultaneously published in JACC: CardioOncology.

Cilie C. van 't Klooster, MD, et al., sought to develop and validate prediction models for estimated lifetime and 10-year risk of total, colorectal and lung cancers in patients with established cardiovascular disease. The researchers developed the model using data from 7,280 patients from the Dutch UCC-SMART trial, which looked at manifestation of arterial disease. The model was then validated using a cohort of 9,322 patients from the CANTOS trial, which evaluated the efficacy of canakinumab in preventing adverse cardiovascular events.

Predictors to assess cancer risk included age, sex, smoking status, weight, height, alcohol use, use of antiplatelet medication, presence of diabetes and C-reactive protein levels. Total cancer was defined as the first diagnoses of any invasive neoplasm, excluding nonmelanoma skin cancer. For colorectal and lung cancers, the researchers counted the first diagnosis of that particular cancer.

For internal validation, there was "good agreement" between the predicted and observed 10-year risk for total cancer (0.61; 95% confidence interval [CI], 0.59-0.63); colorectal cancer (0.61; 95% CI, 0.57-0.66); and lung cancer (0.74; 95% CI, 0.70-0.77). External validation showed "reasonable agreement" for total cancers (0.63; 95% CI, 0.61-0.66); colorectal cancer (0.64; 95% CI, 0.58-0.70); and lung cancer (0.74; 95% CI, 0.70-0.78).

Overall, the model showed a median predicted absolute lifetime risk of 26% for total cancer (range, 1%-52%); 4% for colorectal cancer (range, 0%-13%); and 5% (range, 0%-37%) for lung cancer. The median predicted absolute 10-year risk was 13% for total cancer (range, 1%-31%); 2% for colorectal cancer (range, 0%-6%), and 2% for lung cancer (range, 0%-24%).

According to the researchers, the findings demonstrate that lifetime and 10-year risk of total, colorectal and lung cancer "can be estimated reasonably well" in patients with established cardiovascular disease. With further research, they note that the models could be used to motivate patients to make healthy lifestyle changes and could inform clinical practice in better targeting diagnostics and cancer screenings.

ACC.org Editor-in-Chief Kim A. Eagle, MD, MACC, notes that while there is some predictability with the model used in the present study, the predictive accuracy is not at a sufficiently high level, and cautions that a need for better models remains.

Clinical Topics: Prevention, Novel Agents, Smoking

Keywords: ESC20, ESC Congress, Early Detection of Cancer, Colorectal Neoplasms, Antibodies, Monoclonal, Lung Neoplasms, Diabetes Mellitus, Skin Neoplasms, Smoking, Primary Prevention

< Back to Listings