Aspirin has been labeled as a promising agent for colorectal cancer (CRC) prevention, with analyses suggesting low doses can reduce CRC risk as well as CRC recurrence rates.

Could daily low-dose aspirin prevent other cancers as well, especially in older adults?

That’s the question researchers tackled in a new secondary analysis of the ASPREE study.

Overall, the study suggests that it could be effective for some, though not for others. There’s “significant heterogeneity” in the effectiveness of low-dose aspirin for cancer prevention in older adults, according to Le Thi Phuong Thao, PhD, Monash University, Melbourne, Australia, and colleagues.

More specifically, after dividing the population into treatment-favorable and unfavorable subgroups, low-dose aspirin was associated with a lower cancer risk in the favorable subgroup but a greater cancer risk in the unfavorable subgroup.

The new findings were, however, met with skepticism from two experts.

“While secondary analyses are interesting, we should think of them more as hypothesis-generating rather than confirmatory,” said Christopher Labos, MD, Department of Epidemiology, Biostatistics, and Occupational Health at McGill University, Montreal, Quebec, Canada, and a Medscape Medical News columnist.

“This is ‘personalized medicine’ run amok,” said Gilbert Welch, MD, general internist and senior investigator at the Center for Surgery and Public Health, Brigham and Women’s Hospital, Boston.

Inside the ASPREE Trial

An earlier ASPREE analysis, which randomized 19,114 adults aged 65 years or older to 100 mg/d of low-dose aspirin or placebo, found no overall difference in cancer incidence between the two groups over a median follow-up of 4.6 years.

In the latest analysis, researchers aimed to identify subgroups of individuals more likely to experience a benefit, or harm, from low-dose aspirin for cancer prevention, with the aim of guiding personalized prevention strategies.

The study randomized 9350 participants (median age, 73.7 years) to low-dose aspirin or placebo. Thao and colleagues also identified several factors associated with aspirin benefit, which included older age, nonsmoking status, family history of cancer, and lower BMI. The strongest predictor of benefit was the presence of clonal hematopoiesis of indeterminate potential (CHIP) with a variant allele frequency of 10% or greater. CHIP is an age-related phenomenon where blood-forming cells, known as hematopoietic stem cells, acquire genetic mutations, which have been linked to an increased risk of developing cardiovascular disease and blood cancers.

Using a validated predictive model, the researchers divided participants into treatment-favorable and unfavorable subgroups to predict the personalized treatment effect of low-dose aspirin on cancer incidence.

Over the follow-up period, 10.6% of participants in the low-dose aspirin group developed cancer, and 11% in the placebo group developed cancer.

Among all participants, 59.1% were in the treatment-favorable subgroup, while 40.9% were in the unfavorable-treatment subgroup, which aligned with the “null average treatment effect” seen in the overall trial population (hazard ratio [HR], 0.96; 95% CI, 0.85-1.09).

However, the treatment effect varied across the two subgroups. Low-dose aspirin was associated with a reduced cancer risk in the favorable subgroup (HR, 0.85; 95% CI, 0.71-1.00), but a greater cancer risk in the unfavorable subgroup (HR, 1.14; 95% CI, 0.95-1.38; P = .02 for heterogeneity).

Two experts, whom Medscape Medical News spoke to, questioned the study design and implications.

“I think it’s going to be hard to put too much weight on a secondary analysis of an RCT [randomized controlled trial], when the overall result was negative,” Labos said. If the research on CHIP continues, an RCT of CHIP and non-CHIP participants would be needed, he added.

While Welch supports personalized therapy based on tumor type, “personalizing prevention by screening for genomic information is a different kettle of fish,” he said. “It is an attempt to enumerate debatable risk factors, followed by unproven interventions that, at best, could help only a few.”

The risk for a cancer diagnosis is also the wrong outcome, Welch said, noting that“far more relevant is the risk of cancer death.”

Welch pointed to the 2018 ASPREE study, which explored the effects of low-dose aspirin on all-cause mortality in older adults. In the analysis, the risk for death from any cause was 12.7 events per 1000 person-years in the aspirin group and 11.1 per 1000 in the placebo group (HR, 1.14; 95% CI, 1.01-1.29), with cancer being “the major contributor to the higher mortality in the aspirin group,” accounting for 1.6 excess deaths per 1000 person-years.

The larger question, Welch noted, is where this research is headed. “Are we going to perform DNA-based liquid biopsies on the white blood cells of all older patients? And what are we going to tell the patient with CHIPs?”

Welch reported receiving royalties from three books including Should I Be Tested for Cancer? Labos and the study researchers reported having no disclosures.

The ASPREE trial was supported by the National Institute on Aging (NIA), the National Cancer Institute (NCI), the National Health and Medical Research Council (NHMRC), the Victorian Cancer Agency, and Monash University. The ASPREE-XT study was supported by NIA and NCI, while the ASPREE-CHIP sub-study was supported by NHMRC grants and the National Heart Foundation Future Leadership Fellowship.