Photosensitizing Antihypertensive Drug Use and Risk of Cutaneous Squamous Cell Carcinoma

K.A. Su; L.A. Habel; N.S. Achacoso; G.D. Friedman; M.M. Asgari


The British Journal of Dermatology. 2018;179(5):1088-1094. 

In This Article

Patients and Methods

The source population was the membership of Kaiser Permanente Northern California (KPNC). KPNC is a prepaid, integrated healthcare delivery system that serves approximately 4 million members. All aspects of medical care provided through KPNC's facilities, including pharmacy and pathology records, are captured electronically. The study cohort consisted of participants in KPNC's Research Program in Genes and Environmental Health (RPGEH), details of which have previously been published.[25] Participants were initially enrolled in the RPGEH cohort through participation in mailed surveys of adult KPNC members conducted in 2002–2003 and 2007–2009. Eligible individuals were those who at cohort entry reported NHW race/ethnicity, were at least 18 years of age, had KPNC pharmacy benefits, had no diagnostic codes for rare genetic disorders associated with cSCC risk,[25] had no history of infection with HIV or history of a solid organ transplant and had two or more records for a diagnosis of hypertension [International Statistical Classification of Diseases and Related Health Problems, 9th revision (ICD–9) codes 401·0, 401·1 and 401·9] in the 5 years prior to study entry (n = 28 357). At least 5 years of KPNC health plan membership prior to cohort entry were required to provide sufficient health and medication history. The study was limited to NHW patients because cSCCs occur almost exclusively in fair–skinned individuals. We identified cohort members who developed a new, pathologically verified invasive or in situ cSCC from the survey date to the end of follow–up (31 December 2012) (n = 3010). Cases of cSCC were identified through a review of KPNC's pathology database. All cases of biopsy–proven cSCC in the RPGEH cohort were previously validated.[25]

Electronic pharmacy data were used to determine exposure to ADs. A computerized prescription record system has tracked all prescriptions dispensed within KPNC since August 1994. Based on a review of the literature, ADs were classified as photosensitizing [alpha–2 receptor agonists and diuretics (loop, potassium–sparing, thiazide and combination)], nonphotosensitizing [alpha blockers, beta blockers, central agonists and angiotensin receptor blockers (ARBs)] or unknown [angiotensin converting enzyme (ACE) inhibitors, calcium channel blockers (CCBs), vasodilators and other combinations].[1–7] Exposure was defined as two or more filled prescriptions for a given medication during the study period. While a single prescription for a photosensitizing AD could theoretically increase cSCC risk, two or more filled prescriptions were required to minimize the chance of including patients who filled a prescription but never took the medication. Drug exposure after baseline was treated as a time–varying covariate. Person–time from the start of follow–up until the first use of a given medication was classified as 'never use' for that medication. For the 'ever use' classification, once a patient met the exposure definition (two fills), he or she was considered exposed from that point forward, even if they later discontinued the medication. For exposure based on the number of prescriptions, once a patient met the criteria for a category of fills, he or she was considered exposed to that category until they met the criteria for the next higher category.

We used Cox modelling to estimate hazard ratios (HRs) with 95% confidence intervals (CIs), associating AD use with cSCC. Nonusers of any ADs after cohort entry served as the reference group. We adjusted for variables associated with increased cSCC risk, including age, sex, smoking (current, former, never, unknown), comorbidities [Charlson Comorbidity Index (CCI) score 0, 1–2 or ≥ 3], history of cSCC and actinic keratosis (AK) and prior history of photosensitizing AD use (all in the 5 years prior to baseline). We also adjusted for survey year, length of KPNC health plan membership, baseline healthcare utilization and surveillance measure. Smoking history was based on self–report at cohort entry. CCI is a weighted score of 17 conditions that predicts the risk of 10–year mortality for patients with a range of these diagnosis–based comorbid conditions.[26] History of cSCC was determined from pathology reviews; history of AK was captured using ICD–9 code 702·0; and prior photosensitizing AD use was determined based on pharmacy records. Healthcare utilization was defined as the average annual number of ambulatory visits in the 5 years prior to cohort entry, and the surveillance measure was defined as the average annual number of dermatology visits in the 5 years prior to cohort entry. All analyses were performed with SAS software version 9·3 (SAS Institute, Cary, NC, U.S.A.) and all statistical tests were two–sided. This study was approved by the Institutional Review Board of the Kaiser Foundation Research Institute.