A secondary analysis of the Aspirin in Reducing Events in the Elderly randomized clinical trial found that the effect of daily low-dose aspirin on cancer prevention varied by patient characteristics.
Specifically, aspirin was associated with reduced cancer risk in older patients who were nonsmokers or had lower body mass index, a family history of cancer, or clonal hematopoiesis of indeterminate potential (CHIP) at a variant allele frequency of 10% or greater. In contrast, current smokers and patients with higher body mass index showed no benefit.
The analysis included 9,350 Australian patients aged 70 years and older who were randomized to receive either 100 mg of aspirin daily (n = 4,667) or placebo (n = 4,683). Median follow-up was 4.5 years. Overall, the trial reported no statistically significant difference in cancer incidence between groups. However, when researchers applied an effect score model that integrated multiple patient factors, they identified treatment-favorable and treatment-unfavorable subgroups. The model showed that tailoring aspirin use by subgroup improved 5-year absolute cancer risk reduction by a median of 2%.
During the study, 1,009 patients developed cancer. The most common types were prostate (20%), colorectal (14%), breast (12%), melanoma (11%), and hematologic cancers (11%). CHIP emerged as the strongest predictor of aspirin benefit. A total of 529 patients (6%) carried CHIP at or above the defined threshold. Patients in the treatment-favorable subgroup were older, had lower body mass index, higher hemoglobin levels, more family history of cancer, and more CHIP. Patients in the treatment-unfavorable subgroup were more often smokers and had diabetes, higher body mass index, polypharmacy, and a personal history of cancer. The researchers noted that the mechanism for the association is currently unknown, but could be because of CHIP's association with elevated proinflammatory cytokines such as interleukin (IL)-6, IL-8, and tumor necrosis factor-α.
“Current smoking was the most important predictor for a detrimental effect of aspirin on cancer incidence, whereas CHIP was the most important predictor for a beneficial effect of aspirin, followed by age and a family history of cancer,” wrote lead author Le Thi Phuong Thao, PhD, of the School of Public Health and Preventive Medicine at Monash University in Melbourne, Australia, and colleagues.
They tested 12 candidate prediction models and selected one that maximized cancer risk reduction. The model was validated internally using bootstrapping. CHIP was measured across 18 genes using targeted sequencing of biospecimens collected before or within 1 year of randomization.
The study had limitations. The population was largely White, healthy, and Australian, which may restrict generalizability. Aspirin was started only in patients aged 70 years and older, so it is uncertain whether benefits apply at younger ages. Some biospecimens were collected following randomization, including after cancer diagnoses, which could introduce bias. The analysis was not prespecified and requires external validation prior to use in clinical practice.
"It is crucial to emphasize that identifying subgroups more likely to benefit from [low-dose aspirin (LDA)] for cancer prevention should be considered as one component of a comprehensive benefit-risk assessment that also incorporates careful consideration of potential benefits and risks on other disease processes," Dr. Thao and colleagues added. "In particular, the increased risk of bleeding associated with LDA use, as well as the potential benefit of LDA in cardiovascular disease prevention (eg, which might be seen in individuals with an elevated lipoprotein a genotype), should be considered."
Full disclosures can be found in the published study.
Source: JAMA Oncology
