Nonsteroidal Anti-inflammatory Drugs But Not Aspirin Are Associated With a Lower Risk of Post-colonoscopy Colorectal Cancer

Ka Shing Cheung; Lijia Chen; Esther W. Chan; Wai Kay Seto; Ian C.K. Wong; Wai K. Leung


Aliment Pharmacol Ther. 2020;51(9):899-908. 

In This Article

Abstract and Introduction


Background: Although nonsteroidal anti-inflammatory drugs (NSAIDs) reduce colorectal cancer (CRC) risk, their role in preventing post-colonoscopy CRC (PCCRC) remains undetermined.

Aims: To investigate whether NSAIDs reduce PCCRC risk after a negative baseline colonoscopy

Methods: This is a retrospective cohort study based on a territory-wide healthcare database of Hong Kong. All patients (aged 40 or above) who underwent colonoscopies between 2005 and 2013 were identified. Exclusion criteria included CRC detected within 6 months of index colonoscopy, prior CRC, inflammatory bowel disease and prior colectomy. The primary outcome was PCCRC-3y diagnosed between 6 and 36 months after index colonoscopy. Sites of CRC were categorised as proximal (proximal to splenic flexure) and distal. The adjusted hazards ratio (aHR) of PCCRC-3y with NSAID and aspirin use (defined as cumulative use for ≥90 days within 5 years before index colonoscopy) was derived by propensity score (PS) regression adjustment of 22 covariates (including patient factors, concurrent medication use and endoscopy centre's performance).

Results: Of 187 897 eligible patients, 21 757 (11.6%) were NSAID users. 854 (0.45%) developed PCCRC-3y (proximal cancer: 147 [17.2%]). NSAIDs were associated with a lower PCCRC-3y risk (aHR: 0.54, 95% CI: 0.41–0.70), but not CRC that developed >3 years (aHR: 0.78, 95% CI 0.56–1.09). The aHR was 0.48 (95% CI: 0.24–0.95) for proximal and 0.55 (95% CI: 0.40–0.74) for distal cancer. A duration- and frequency response relationship was observed (P trend < 0.001). For aspirin, the aHR was 1.01 (95% CI: 0.80–1.28).

Conclusions: Non-aspirin NSAIDs were associated with lower PCCRC risk after a negative baseline colonoscopy.


Colorectal cancer (CRC) is the third most common cancer and second leading cause of death worldwide.[1] Screening colonoscopy can reduce incidence[2–4] and mortality of CRC,[4–6] but CRC can still occur after initial colonoscopy in which no cancer was detected. These cancers are termed 'post-colonoscopy CRC' (PCCRC) by the recent World Endoscopy Organization (WEO) consensus.[7] In contrast to interval CRC, which refers to cancer that develops shortly after screening/surveillance colonoscopy, PCCRC encompasses cancers that develops after any diagnostic colonoscopy and could account for up to 9% of all diagnosed CRCs,[8,9] with a predilection for proximal colon.[10] The mechanisms for PCCRC development could be accounted by incomplete colonoscopy (due to technical difficulty or luminal obstruction), missed lesions at the index colonoscopy (around 50% of the cases),[8] incomplete resection of polyps, tumours arising from alternative pathway including the sessile serrated pathway with rapid growth[11–13] and tumour seeding by biopsy forceps or needle injectors.[14]

Nonsteroidal anti-inflammatory drugs (NSAIDs) have been shown to possess chemopreventive effect on CRC. A systematic review of clinical studies showed that NSAIDs reduced both adenoma and CRC development.[15] A recent population-based case-control study shows that non-aspirin NSAIDs were associated with a duration-dependent risk reduction of CRC with effect persisting up to 1 year after discontinuation.[16] Multiple mechanisms have been proposed. First, NSAIDs induce apoptosis in CRC cells by inhibiting prostaglandin (PG) synthesis and hence increase in the levels of precursor arachidonic acid, which is involved in mediating conversion of sphingomyelin to ceramide, a mediator of apoptosis.[17] Second, PGs are shown to be associated with tumour angiogenesis, proliferation of tumour cells and immune surveillance inhibition.[18] Third, inhibition of cyclooxygenase-2 (COX-2)-derived PG production inactivates epidermal growth factor receptor (EGFR) signalling.[19] Other non-COX-mediated mechanisms of NSAIDs in cancer prevention include inhibition of activating pathways of nuclear factor kappa B (NF-kappa B)[20] and insulin-related neoplastic pathways.[21] While NSAIDs inhibit both COX-1 and COX-2, aspirin is more selective for COX-1 inhibition,[22,23] which may explain why NSAIDs appear to be more efficacious than aspirin in preventing advanced metachronous neoplasia in patients with previous colorectal neoplasia.[24] In addition, the chemopreventive effect of aspirin requires prolonged use (at least 5 years) in comparison to NSAIDs.[15,25,26]

While there is ample evidence that NSAIDs and aspirin reduce colorectal adenomas and cancer, studies that specifically focus on their chemopreventive role in PCCRC are lacking. NSAIDs/aspirin may not be effective or minimally effective in individuals who have already undergone colonoscopy in which no cancer was found and all polyps were removed. Moreover, as NSAIDs are associated with side effects like gastrointestinal bleeding (GIB), nephrotoxicity and cardiovascular events, subgroups that will benefit from the chemopreventive effects of NSAIDs should be identified.

In this study, we aimed to determine the association between the use of NSAIDs/aspirin and PCCRC development in a large cohort of patients who had undergone colonoscopy with no baseline CRC.