The common diuretic hydrochlorothiazide is linked to a dose-dependent increased risk of nonmelanoma skin cancer, in particular, squamous cell carcinoma, a case-controlled registry study showed.
This nationwide, case-matched control study examined patients’ cumulative hydrochlorothiazide use between 1995 and 2012 and found a clear dose-response patterns for both basal cell carcinoma (BCC) and squamous cell carcinoma (SCC), with a more than sevenfold increased risk of SCC for a cumulative use of greater than or equal to 200,000 mg of hydrochlorothiazide (HCTZ).
“Assuming causality, the present results suggest that 1 in 10 SCC cases diagnosed during the study period can be attributed to HCTZ use,” wrote the study authors, who were led by Sidsel Arnspang, MD, of the department of neurology at Odense (Denmark) University Hospital, which is affiliated with the University of Southern Denmark.
The authors noted that they previously had reported a sevenfold increased risk of lip squamous cell carcinoma with hydrochlorothiazide. Furthermore, the International Agency for Research on Cancer recently classified the diuretic and antihypertensive as “possibly carcinogenic to humans.”
“As HCTZ is among the most widely used drugs in the U.S. and Western Europe, a carcinogenic effect of HCTZ would have a considerable impact on public health,” they wrote in their paper, published in the Journal of American Academy of Dermatology.
According to the study authors, the few studies that have investigated a potential link between thiazide use and nonmelanoma skin cancer (NMSC) have reported inconsistent results.
They speculated that this could be because HCTZ often is prescribed in combination with other diuretics, and there may have been difficulties with disentangling its effect from those of the other drugs.
Using data from five nationwide data sources, the research team compared HCTZ use among people diagnosed with SCC or BCC of the skin to use among a matched control group without such cancers. People were excluded from the analysis if they had SCC of the lip because they had been evaluated in the research team’s previous study.
High use of HCTZ was defined as filled prescriptions totaling greater than or equal to 50,000 mg of HCTZ, which corresponds to greater than or equal to 2,000 defined daily doses (for example, approximately 6 years of cumulative use).
Overall, the study population involved 71,533 BCC and 8,629 SCC cases that were matched to 1,430,883 and 172,462 population controls, respectively.
Baseline characteristics of the cases and controls were similar, except BCC cases were slightly more educated than controls. Results showed that high use of hydrochlorothiazide was associated with odds ratios of 1.29 (95% confidence interval, 1.23-1.35) for BCC and 3.98 (95% CI, 3.68-4.31) for SCC.
A clear dose-response relationship was observed with HCTZ use for both BCC and SCC, with the highest ORs observed in the upper exposure category (greater than or equal to 200,000 mg): The OR for BCC in this category was 1.54 (95% CI, 1.38-1.71) and 7.38 for SCC (95% CI, 6.32-8.60).
The researchers observed no associations for BCC or SCC risk with use of other diuretics and other hypertensives, a finding that they said supported a potential causal association between HCTZ and NMSC risk.
Little variation was seen in the association between HCTZ use and BCC or SCC risk in the subgroup analyses, except for notably stronger associations among younger individuals and females.
In analyses stratified according to tumor localization, the authors saw stronger associations for cancers at sun-exposed skin sites, especially the skin of the lower limbs.
“Given the considerable use of HCTZ worldwide and the morbidity associated with NMSC, a causal association between HCTZ use and NMSC risk would have significant public health implications,” Dr. Arnspang and associates concluded. “The use of HCTZ should be carefully considered as several other antihypertensive agents with similar indications and efficiency are available but without known associations with skin cancer.”
The investigators cited several limitations. For example, information on ethnicity and skin type was not available. This information would have been useful in evaluating participants’ photosensitivity as a possible mechanism for a higher skin cancer risk with the use of HCTZ.
The study was funded by a grant from the Danish Cancer Society and the Danish Council of Independent Research. Several of the authors reported receiving grants and or honoraria from pharmaceutical companies.
SOURCE: Arnspang S et al. JAAD. 2017. doi: 10.1016/j.jaad.2017.11.042.