Colorectal polyp prevention trials (PPTs) are randomized, placebo-controlled clinical trials that evaluate some chemo-preventive agent and include participants who will be followed for at least 3 years to compare the recurrence rates (counts) of adenomas. A large proportion of zero counts will likely be observed in both groups at the end of the observation period. Poisson general linear models (GLMs) are usually employed for estimation of recurrence in PPTs. Other models, including the negative binomial (NB2), zero-inflated Poisson (ZIP), and zero-inflated negative binomial (ZINB) may be better suited to handle zero-inflation or other forms of overdispersion that are common in count data. A model selection paradigm that determines a statistical approach for choosing the best fitting model for recurrence data is described. An example using a subset from a large Phase III clinical trial indicated that the ZINB model was the best fitting model for the data.

Yes / Background: The omega-3 polyunsaturated fatty acid eicosapentaenoic acid (EPA) and aspirin both have proof of concept for colorectal cancer chemoprevention, aligned with an excellent safety profile. Therefore, we aimed to test the efficacy of EPA and aspirin, alone and in combination and compared with a placebo, in individuals with sporadic colorectal neoplasia detected at colonoscopy. Methods: In a multicentre, randomised, double-blind, placebo-controlled, 2 × 2 factorial trial, patients aged 55–73 years who were identified during colonoscopy as being at high risk in the English Bowel Cancer Screening Programme (BCSP; ≥3 adenomas if at least one was ≥10 mm in diameter or ≥5 adenomas if these were <10 mm in diameter) were recruited from 53 BCSP endoscopy units in England, UK. Patients were randomly allocated (1:1:1:1) using a secure web-based server to receive 2 g EPA-free fatty acid (FFA) per day (either as the FFA or triglyceride), 300 mg aspirin per day, both treatments in combination, or placebo for 12 months using random permuted blocks of randomly varying size, and stratified by BCSP site. Research staff and participants were masked to group assignment. The primary endpoint was the adenoma detection rate (ADR; the proportion of participants with any adenoma) at 1 year surveillance colonoscopy analysed in all participants with observable follow-up data using a so-called at-the-margins approach, adjusted for BCSP site and repeat endoscopy at baseline. The safety population included all participants who received at least one dose of study drug. The trial is registered with the International Standard Randomised Controlled Trials Number registry, number ISRCTN05926847. Findings: Between Nov 11, 2011, and June 10, 2016, 709 participants were randomly assigned to four treatment groups (176 to placebo, 179 to EPA, 177 to aspirin, and 177 to EPA plus aspirin). Adenoma outcome data were available for 163 (93%) patients in the placebo group, 153 (85%) in the EPA group, 163 (92%) in the aspirin group, and 161 (91%) in the EPA plus aspirin group. The ADR was 61% (100 of 163) in the placebo group, 63% (97 of 153) in the EPA group, 61% (100 of 163) in the aspirin group, and 61% (98 of 161) in the EPA plus aspirin group, with no evidence of any effect for EPA (risk ratio [RR] 0·98, 95% CI 0·87 to 1·12; risk difference –0·9%, –8·8 to 6·9; p=0·81) or aspirin (RR 0·99 (0·87 to 1·12; risk difference –0·6%, –8·5 to 7·2; p=0·88). EPA and aspirin were well tolerated (78 [44%] of 176 had ≥1 adverse event in the placebo group compared with 82 [46%] in the EPA group, 68 [39%] in the aspirin group, and 76 [45%] in the EPA plus aspirin group), although the number of gastrointestinal adverse events was increased in the EPA alone group at 146 events (compared with 85 in the placebo group, 86 in the aspirin group, and 68 in the aspirin plus placebo group). Six upper-gastrointestinal bleeding events were reported across the treatment groups (two in the EPA group, three in the aspirin group, and one in the placebo group). Interpretation Neither EPA nor aspirin treatment were associated with a reduction in the proportion of patients with at least one colorectal adenoma. Further research is needed regarding the effect on colorectal adenoma number according to adenoma type and location. Optimal use of EPA and aspirin might need a precision medicine approach to adenoma recurrence. / Efficacy and Mechanism Evaluation Programme, a UK Medical Research Council and National Institute for Health Research partnership. / Research Development Fund Publication Prize Award winner, November 2018.

Yes / The seAFOod polyp prevention trial was a randomised, placebo-controlled, 2 × 2 factorial trial of aspirin 300 mg and eicosapentaenoic acid (EPA) 2000 mg daily in individuals who had a screening colonoscopy in the English Bowel Cancer Screening Programme (BCSP). Aspirin treatment was associated with a 20% reduction in colorectal polyp number at BCSP surveillance colonoscopy 12 months later. It is unclear what happens to colorectal polyp risk after short-term aspirin use. To investigate colorectal polyp risk according to the original trial treatment allocation, up to 6 years after trial participation. All seAFOod trial participants were scheduled for further BCSP surveillance and provided informed consent for the collection of colonoscopy outcomes. We linked BCSP colonoscopy data to trial outcomes data. In total, 507 individuals underwent one or more colonoscopies after trial participation. Individuals grouped by treatment allocation were well matched for clinical characteristics, follow-up duration and number of surveillance colonoscopies. The polyp detection rate (PDR; the number of individuals who had ≥1 colorectal polyp detected) after randomization to placebo aspirin was 71.1%. The PDR was 80.1% for individuals who had received aspirin (odds ratio [OR] 1.13 [95% confidence interval 1.02, 1.24]; p = 0.02). There was no difference in colorectal polyp outcomes between individuals who had been allocated to EPA compared with its placebo (OR for PDR 1.00 [0.91, 1.10]; p = 0.92). Individuals who received aspirin in the seAFOod trial demonstrated increased colorectal polyp risk during post-trial surveillance. Rebound elevated neoplastic risk after short-term aspirin use has important implications for aspirin cessation driven by age-related bleeding risk. ISRCTN05926847. / Efficacy and Mechanism Evaluation Programme. Grant Number: NIHR128210.

Yes / Aspirin and eicosapentaenoic acid (EPA) reduce colorectal adenomatous polyp risk and affect synthesis of oxylipins including prostaglandin E2. We investigated whether 35 single nucleotide polymorphisms (SNPs) in oxylipin metabolism genes such as cyclooxygenase [PTGS] and lipoxygenase [ALOX], as well as 7 SNPs already associated with colorectal cancer (CRC) risk reduction by aspirin (eg. TP53; rs104522), modified the effects of aspirin and EPA on colorectal polyp recurrence in the randomised 2x2 factorial seAFOod trial. Treatment effects were reported as the incidence rate ratio (IRR) and 95% confidence interval (CI) by stratifying negative binomial and Poisson regression analyses of colorectal polyp risk on SNP genotype. Statistical significance was reported with adjustment for the false discovery rate as the P and q value. Five hundred and forty-two (of 707) trial participants had both genotype and colonoscopy outcome data. Reduction in colorectal polyp risk in aspirin users compared with non-aspirin users was restricted to rs4837960 (PTGS1) common homozygotes (IRR 0.69 [95%CI 0.53,0.90]; q=0.06), rs2745557 (PTGS2) compound heterozygote-rare homozygotes (IRR 0.60 [0.41,0.88]; q=0.06), rs7090328 (ALOX5) rare homozygotes (IRR 0.27 [0.11,0.64]; q=0.05), rs2073438 (ALOX12) common homozygotes (IRR 0.57 [0.41,0.80]; q=0.05), and rs104522 (TP53) rare homozygotes (IRR 0.37 [0.17,0.79]; q=0.06). No modification of colorectal polyp risk in EPA users was observed. In conclusion, genetic variants relevant to the proposed mechanism of action on oxylipins are associated with differential colorectal polyp risk reduction by aspirin in individuals who develop multiple colorectal polyps. SNP genotypes should be considered during development of personalised, predictive models of CRC chemoprevention by aspirin. / Funder(s): Efficacy and Mechanism Evaluation Programme (EME) Award Id(s): NIHR128210. Funder(s): NIHR Senior Investigator grant. Funder(s): Cancer Research UK (CRUK) Award Id(s): C23434/A24939. Funder(s): European Union-BBSRC (UK) Award Id(s): BB/P028233/1.