No association between use of phosphodiesterase 5 inhibitors and colorectal cancer in men with erectile dysfunction

Abstract

PurposeThere is an increase interest on the potential chemoprotective effect of selective phosphodiesterase 5 (PDE5) inhibitors. Several authors have shown in vivo the immune‐mediated anti‐tumor effect of these inhibitors on tumors arising from the digestive tract.ObjectivesTo test the potential effect of selective PDE5 inhibitors against colorectal cancer (CRC) onset previously observed.MethodsWe used data from The Health Improvement Network database and identified an established cohort of 200 000 new users of low‐dose aspirin and a matched comparison cohort aged 40‐84 years between 1 January 2000 and 31 December 2011. A follow‐up to identify CRC cases was performed within an extensive validation exercise. Nested case‐control analyses compared PDE5 inhibitors vs non‐use on CRC risk were performed.ResultsRestricting to males (59.3% controls and 59.5% cases), no association was observed among current users of PDE5 inhibitors (1.05 [95% CI: 0.69‐1.60]) and neither among recent (1.36 [95% CI: 0.81‐2.28]) or past users (1.06 [95% CI: 0.72‐1.58]). No duration response effect was found.ConclusionsOur results do not support an increased risk of CRC associated with the use of PDE5 inhibitors among men with erectile dysfunction.

The protective role of phosphodiesterase inhibitors in preventing colorectal cancer and advanced colorectal polyps: a systematic review and meta‐analysis

“…However, omission of two relevant studies [22, 29], exclusion of subgroup analysis on primary and secondary prevention and lack of prospective PROSPERO registration of the previous meta‐analysis warranted the conduction of a more comprehensive review on this subject area. The protective effect of PDE5 inhibitors on CRC risk suggested in the present study contrasts with that reported by Soriano et al [36], who found that once the confounder of erectile dysfunction was accounted for no reduced risk of CRC was observed among users of PDE5 inhibitors. In contrast, Sutton et al studied a cohort of patients with erectile dysfunction and reported a lower hazard of CRC among those who had exposure to PDE5 inhibitors, suggesting that PDE5 inhibitor use is an independent protective variable in CRC incidence.…”

“…All studies were published between 2016 and 2021. Study populations originated from the USA [25, 29], the UK [35, 36] and Sweden [22, 34]. There was 100% agreement between the two reviewers on review of the extracted data.…”

“…A total of 675 publications were identified using the predefined search terms, of which six were eligible for inclusion [22,25,29,[34][35][36]. Four were retrospective cohort studies [22,25,29,34] and two were case-control studies [35,36]. All studies were published between 2016 and 2021.…”

“…Included studies adjusted for different variables. All studies adjusted for age [22,25,29,[34][35][36], two studies adjusted for race [25,29], four studies adjusted for comorbidities [25,29,34], three studies adjusted for family history [25,29,34] and five studies adjusted for lifestyle factors including body mass index and smoking [22,29,[34][35][36].…”

“…All studies had a total study population of over 1000 patients. Five studies included patients without a previous history of CRC [25,29,[34][35][36], while one study examined the chemopreventative role of PDE5 inhibitors in patients with a previous diagnosis of CRC [22]. Sutton et al [25], Zhang et al [29] and Soriano et al [36] included patients with erectile dysfunction and no history of PDE5 inhibitor use as a reference group to eliminate "confounding by indication" whereas Huang et al [22], Matthews et al [35] and Huang et al [34] used non PDE5 inhibitor users as a reference group.…”