The CaPP3 trial found broadly similar cancer outcomes with lower-dose aspirin in Lynch syndrome, with fewer adverse events and less serious bleeding.
Written By: Creola Gonsalves, MS Biotech
Reviewed By: Pharmacally Editorial Team
The CaPP3 trial has provided important evidence on aspirin dose selection for cancer prevention in people with Lynch syndrome. Published in The Lancet Gastroenterology & Hepatology, the multicentre, randomised, double-blind, non-inferiority trial compared daily aspirin doses of 100 mg, 300 mg, and 600 mg.
Although formal non-inferiority could not be concluded for either lower dose, the 100 mg group showed cancer-risk characteristics broadly similar to 600 mg, while experiencing fewer adverse events and substantially fewer serious bleeding events.
Why Was CaPP3 Conducted?
The earlier CaPP2 trial (ISRCTN59521990) showed that 600 mg aspirin daily reduced colorectal cancer and Lynch syndrome-associated cancers among people with Lynch syndrome. However, long-term treatment with higher-dose aspirin raises concerns about adverse effects, particularly bleeding.
CaPP3 was designed to determine whether lower doses could retain the observed cancer-prevention benefit while reducing treatment-related harm.
Study Design
CaPP3 (NCT02497820) enrolled adults with Lynch syndrome carrying pathogenic variants in MLH1, MSH2, MSH6, or PMS2. Participants were recruited from clinical genetics centres in the UK, Australia, Finland, Israel, and Spain.
Between October 2014 and March 2019, 1,879 participants were randomly assigned in a 3:3:4 ratio to 100 mg, 300 mg, or 600 mg aspirin daily. The trial was double-blinded for two years, followed by three years of open-label treatment and planned longer-term cancer follow-up. In the UK, 75 mg replaced 100 mg during the open-label phase.
The primary outcome was new primary mismatch-repair-deficient cancers, referred to as Lynch syndrome cancers. Investigators assessed time to first cancer using hazard ratios (HRs) and overall cancer burden using incidence rate ratios (IRRs). The predefined non-inferiority margin was 1.5.
100 mg Showed Similar Cancer Characteristics, but Non-Inferiority Was Not Established
After a median follow-up of 66.4 months, 176 participants had developed 216 Lynch syndrome cancers. The numbers were 57 cancers in the 100 mg group, 75 in the 300 mg group, and 84 in the 600 mg group.
For the 100 mg dose, the intention-to-treat analysis produced an HR of 0.97 (95% CI 0.67–1.42) for time to first Lynch syndrome cancer and an IRR of 0.94 (95% CI 0.65–1.38) for cancer burden.
However, formal non-inferiority required consistency across the intention-to-treat and per-protocol populations and both statistical measures. In the per-protocol analysis, cancer burden was non-inferior with an IRR of 0.90 (95% CI 0.55–1.46), but time to first cancer produced an HR of 1.00 (95% CI 0.61–1.63). Because the upper confidence limit exceeded the 1.5 boundary, formal non-inferiority could not be concluded.
Thus, 100 mg should not be described as statistically proven non-inferior to 600 mg. Instead, the findings indicate broadly similar cancer-risk characteristics with a more favorable observed safety profile.
300 mg Did Not Demonstrate Non-Inferiority
Non-inferiority was not demonstrated for 300 mg in either the intention-to-treat or per-protocol analyses. The intention-to-treat HR was 1.28 (95% CI 0.91–1.80) and IRR was 1.12 (95% CI 0.79–1.61).
Colorectal Cancer Findings
Colorectal cancer accounted for an important proportion of the observed cancers. Among 1,846 participants included in the colorectal cancer analysis, 83 developed colorectal cancer: 21 in the 100 mg group, 30 in the 300 mg group, and 32 in the 600 mg group.
For colorectal cancer, 100 mg was non-inferior for cancer burden but did not formally meet non-inferiority for time to first colorectal cancer, while 300 mg did not demonstrate non-inferiority for either analysis.
Lower Dose Reduced Adverse Events and Serious Bleeding
The safety population included 1,875 participants. Adverse events occurred in 25.0% of participants receiving 100 mg, 26.8% receiving 300 mg, and 31.2% receiving 600 mg (p=0.03 for heterogeneity).
The difference was particularly notable for serious bleeding. Over five years, bleeding-related serious adverse events occurred in none of the 100 mg group, compared with 3 participants (0.5%) receiving 300 mg and 11 (1.5%) receiving 600 mg (p=0.004).
These findings are important because bleeding is one of the principal concerns associated with long-term aspirin therapy.
Possible Mechanisms Behind Aspirin’s Cancer-Preventive Effect
The precise mechanism through which aspirin may reduce cancer risk in Lynch syndrome remains uncertain, and CaPP3 was not designed to establish a mechanism. Several biological hypotheses may nevertheless explain the clinical observations.
Platelet inhibition and immune effects: Low-dose aspirin strongly inhibits platelet COX-1 and platelet activation. Experimental and translational evidence suggests that platelets can contribute to an immunosuppressive environment around emerging tumour cells. Aspirin’s antiplatelet activity may therefore influence immune surveillance, although this remains a proposed mechanism rather than a finding proven by CaPP3.
Cell death and mTOR signalling: Experimental studies have also suggested that aspirin can influence pathways involved in apoptosis, cellular proliferation and mTOR signalling. These effects could potentially contribute to suppression of abnormal cells, but their specific contribution to Lynch syndrome cancer prevention remains unproven.
Delayed onset of protection: The earlier CAPP2 trial showed that aspirin’s effect on colorectal cancer risk was not immediately apparent. Cancer incidence began to diverge between the aspirin and placebo groups approximately five years after treatment initiation, despite a median aspirin treatment duration of only 26.5 months. The persistence of the protective effect after treatment had ended suggests that aspirin may influence an early stage of carcinogenesis, although the biological basis for this delayed effect remains uncertain.
Clinical Perspective
The CaPP3 investigators have emphasised that the lower-dose findings are clinically relevant despite the strict statistical non-inferiority criteria. Lead investigator Sir John Burn has discussed 75–100 mg daily as a practical lower-dose approach, with higher doses potentially considered for people with greater body weight. Such BMI-based dose adjustment should be regarded as clinical guidance or expert interpretation rather than a dose requirement established by the CaPP3 randomised comparison.
Limitations and Conclusion
The current analysis remains limited by the duration of follow-up for a cancer-prevention study. Median follow-up was 66.4 months, and further follow-up is planned until participants reach 10 years. Long-term adherence was also incomplete, with 730 of 1,866 participants stopping aspirin during the two-year blinded phase.
CaPP3 did not formally establish non-inferiority of 100 mg or 300 mg aspirin compared with 600 mg. However, 100 mg showed broadly similar cancer-risk characteristics while producing fewer adverse events and substantially fewer serious bleeding events. The findings therefore raise an important dose-selection question: whether a lower aspirin dose can provide a sufficiently durable cancer-prevention benefit while reducing the harms associated with higher-dose treatment.
Longer follow-up will be important to determine the durability of these findings and establish the optimal aspirin dose for people with Lynch syndrome.
Reference
About the Writer
Creola Gonsalves (Linkedin) is an M.S. Biotechnology postgraduate with a strong interest in clinical research, evidence interpretation, and healthcare writing, with a focus on translating life-science knowledge into meaningful real-world insights.
She is trained in Good Clinical Practice (GCP), clinical research principles, and critical interpretation of randomized clinical trials, with certifications from NIH and Stanford University.
Her research background in biotechnological applications and microbial research strengthens her ability to understand scientific evidence and develop clear, accurate, and research-driven healthcare content.