INDIVIDUAL ANTIHYPERTENSIVE MEDICATION AND SKIN CANCER RISK: A SYSTEMATIC REVIEW AND META-ANALYSIS

Abstract

Objective: The use of hydrochlorothiazide has recently been linked to skin cancer in observational studies. This may be explained by its enninnsitising properties, but photosensitivity has also been reported for other antihypertensive drugs. We conducted a systematic review and meta-analysis to compare skin cancer risk among antihypertensive drug classes and between individual blood pressure lowering drugs. Design and method: We searched Medline, Embase and the Web of Science and included studies that investigated the association between antihypertensive medication exposure and non-melanoma skin cancer (NMSC) or cutaneous malignant melanoma (CMM). We combined the extracted the adjusted odds ratios (OR) using a random effects model. We performed sensitivity analyses comparing study type, population and studies with and without correction for important covariates. Results: We included 37 studies. Diuretics (n = 16,625,573), in particular hydrochlorothiazide (n = 16,068,501), and calcium channel blockers (CCB, n = 13,176,689) were examined more frequently than beta-blockers (n = 94,872) or ennin-angiotensin system inhibitors (n = 2,014,579). Exposure to diuretics (OR 1.33, 95% CI [1.10–1.61]) and CCB (OR 1.06, 95% CI [1.04–1.09]) was associated with an increased risk for NMSC. No antihypertensive drug class was associated with an increased risk for CMM. Within every drug class, at least 1 individual drug was associated with an increased risk for NMSC: nifedipine, hydrochlorothiazide, sotalol, verapamil, enalapril and bendroflumethiazide. Increased NMSC risk was only observed in case-control studies, specific populations (such as non-Hispanic whites, patients with coronary heart disease or veterans) and studies that did not correct for important covariates. No increased NMSC risk was observed in cohort studies, the general population, and studies that did correct for the covariates sun exposure, skin phototype or smoking (Figure). Only 2 out of 37 studies provided information about antihypertensive co-medication. Egger's test revealed a significant publication bias for the subgroup of diuretics and hydrochlorothiazide concerning NMSC (both p < 0.0001). Conclusions: Of all antihypertensive drugs, diuretics and calcium channel blockers may be associated with an increased risk of NMSC, but causal inferences remain difficult to make because of substantiation limitation of the available data.