Hydrochlorothiazide and the risk of skin cancer in adults with and without hypertension

Hydrochlorothiazide (HCTZ) is a diuretic drug that is usually prescribed for the treatment of hypertension or fluid retention as a result of heart failure [1]. Thiazides and thiazide-like diuretics were introduced in the 1960s and are still regarded as mainstay of antihypertensive treatment. As these drugs have been proven to reduce blood pressure and have been shown to be effective in preventing cardiovascular morbidities and mortality in multiple randomized controlled trials, HCTs have been recommended as one of the five main drug classes for the treatment of hypertension in the 2018 Guidelines of the European Society of Cardiology (ESC) and European Society of Hypertension (ESH) [2]. As a result, hydrochlorothiazide and other thiazide diuretics belong to one of the most prescribed antihypertensive drugs in the United States and Western European countries [3]. In the Netherlands, hydrochlorothiazide is usually prescribed with other antihypertensive drugs in combination preparations or in combination with potassium-sparing diuretics [4].

Besides common side-effects, such as hypokalemia, hyponatremia and elevated plasma glucose and cholesterol, hydrochlorothiazide has also been known to have photosensitizing properties [5,6,7]. Drug-induced photosensitivity, which mostly presents as skin eruptions, is caused by a combination of exposure to UV-radiation and exposure to a drug. The chemical structure of photosensitizing drugs allows for the absorption of UV-radiation [8,9], which results in the formation of reactive oxygen species (ROS) and cellular and DNA damage through either phototoxic or photoallergic reactions [10, 11]. As hydrochlorothiazide promotes increased exposure to UV-radiation, which is a major risk factor for all types of skin cancer, prolonged use might be linked to skin cancer [12].

Skin cancer is the most common form of cancer in Caucasian individuals [13,15,48]. Despite increasing knowledge about the risk of skin cancer with excessive UV-exposure, the incidence has been increasing since the last 10 years in the Netherlands [13,48]. Basal cell carcinoma (BCC) and squamous cell carcinoma (SCC) are the most common subtypes of non-melanoma skin cancer [15], with an incidence rate of 48,022 cases of BCC and 12,318 cases of SCC in 2017 [48]. Malignant melanoma are less common but are more lethal and associated with a higher mortality rate (4.6 in 100,000 individuals) [16,48]. Despite low mortality rates for BCC and SCC, morbidity can be high as these skin cancers are often associated with ingrowths in surrounding tissue. Moreover, new skin tumors are detected within 3 years after the first diagnosis in 25% cases of BCC and 10% cases of SCC, which requires additional surgical procedures. Predictive models have indicated that the annual incidence of BCC, SCC and melanoma is expected to increase between 2017 and 2027, with an increase of 35% in skin cancer patients that require hospital care [48]. As a result of this increased incidence but also the introduction of novel therapeutic drugs, such as monoclonal antibodies, health care costs are expected to increase as well the coming years [49].In 2017, skin tumors were the 4th most costly cancer in the Netherlands, with total hospital costs of €465 million. Based on a predicted annual increase of 5% in incidence rate, total costs for skin cancer are expected to be €1.35 billion in 2030 [49]. These data show that a potential association between skin cancer and hydrochlorothiazide, a commonly prescribed drug, can have a major impact on public health and changes in the current health care policy of hypertensive patients might be necessary, as patients using HCTZ are currently not monitored nor educated on potential risks of skin cancer.

Several studies have indicated that the use of hydrochlorothiazide may be associated with an increased risk of malignant melanoma (MM) and non-melanoma skin cancer (NMSC), particularly squamous cell carcinoma (SCC) [17,18,19,20].Pedersen et al. found an adjusted odds ratio (OR) of 1.29 for BCC with high users of hydrochlorothiazide and 3.98 for SCC [18]. Particularly strong dose-response associations were seen with the highest association of 1.54 for BCC in the highest dose category (≥200,000 mg, P <0.001) and 7.38 for SCC in the highest dosage group (P <0.001). Similar associations were seen in a study of Jensen et al. with an incidence rate ratio (IRR) of 1.92 for SCC with HCTZ prescriptions of more than 5 years and 1.10 for BCC [19]. Another Danish study also demonstrated an association with other types of skin cancers, with an odds ratio of 2.3 for Merkel cell carcinoma with high use of HCTZ (≥50,000 mg) and 3.6 for malignant adnexal skin tumors [20].

However, results are inconsistent as some studies have found weak or no associations between HCTZ use and risk of skin cancer [21,22].Ruiter et al. found no association with a hazard ratio (HR) of 1.00 for any use of thiazides [21] and similar results were seen in a study by Schmidt et al. with odds ratio of 1.03 for SCC with thiazide use and 1.05 for BCC [22].

These differences in association are likely due to different definitions of drug exposure and cancer outcomes, as several studies have investigated hydrochlorothiazide use separately while others studied thiazide diuretics as one group and the same applies to cancer outcomes, which are grouped differently by studies. Moreover, many studies have methodological limitations, including information, confounding and selection bias and/or are small sample sized to rule out a possible true effect. Additionally, multiple studies failed to collect information about potential risk factors of skin cancer such as UV-exposure and skin phenotype[17,18].

These inconsistent results and methodological limitations have inspired us to investigate the risk factors for skin cancer as well as the association between hydrochlorothiazide use and skin cancer in a cohort of unselected adults with and without hypertension.