Antihypertensive Drugs and Cancer Risk: A Critical Reappraisal of Drug-Specific Signals, Hydrochlorothiazide-Associated Skin Cancer, Nitrosamine Impurities, and Regulatory Perspectives

Hypertension is a major modifiable determinant of cardiovascular, cerebrovascular and renal disease. The clinical benefit of antihypertensive therapy is firmly established, and treatment reduces major cardiovascular events and premature mortality. Nevertheless, antihypertensive exposure commonly extends over many years, making the evaluation of uncommon or delayed adverse outcomes scientifically and clinically relevant. Cancer concerns surrounding antihypertensive therapy have emerged from several evidence streams that answer different questions. Observational studies may identify statistical associations but remain vulnerable to confounding, treatment selection and reverse causation. Randomized trials provide stronger protection against confusion but may have limited duration for outcomes with long latency. Mechanistic studies can establish biological plausibility without proving clinical causality. Pharmaceutical-quality investigations address yet another question: whether a finished product contains a carcinogenic impurity generated during synthesis, formulation, storage, or other stages of manufacture. The controversy became especially prominent after the identification of NDMA in selected Alsatian products in 2018, followed by recognition of other nitrosamines in certain medicines. The pharmacological properties of Alsatian and the toxicological properties of NDMA are therefore separate issues. A second, independent signal concerns HCTZ, a photosensitizing diuretic for which epidemiological evidence has accumulated for selected skin cancers. Accordingly, this review uses a three-pathway framework: (1) intrinsic active-drug effects; (2) molecule-specific biological effects such as HCTZ-associated skin cancer; and (3) product-specific impurity-mediated risk. This distinction is central to rational pharmacovigilance and avoids the scientifically imprecise proposition that antihypertensive medicines as a therapeutic class are carcinogenic.

2. Three Distinct Cancer-Risk Scenarios

3. Evidence from Randomized and Contemporary Observational Studies

An individual-participant-data meta-analysis of 33 randomized trials involving 260,447 participants and 15,012 cancer events found no consistent increase in overall cancer associated with ACE inhibitors, ARBs, β-blockers, or thiazides. A small association for calcium-channel blockers emerged in active-comparator analyses but was not statistically significant when compared with placebo. This distinction illustrates why comparator selection is critical in drug-safety research.

More recent real-world evidence remains broadly consistent with the absence of a clinically meaningful class-wide cancer effect. A 2025 cohort study of 270,320 newly diagnosed patients with hypertension, with a median follow-up of 7.7 years and 14,264 cancer events, found no significant association of ACE inhibitors, ARBs, β-blockers, or thiazide diuretics with overall cancer. CCBs showed a modest association (HR 1.05, 95% CI 1.01–1.09), but the association disappeared in sensitivity analyses excluding short follow-up, supporting caution about residual confounding and reverse causation. These data do not prove that every antihypertensive molecule is completely free of carcinogenic potential. Rather, they argue strongly against treating antihypertensive therapy as a homogeneous carcinogenic exposure. Drug-specific analyses, exposure duration, cancer subtype, comparator selection, and latency remain essential.

4. Drug-Class-Specific Evidence

4.1 ACE Inhibitors

ACE inhibitors reduce angiotensin II generation and increase bradykinin availability. Because the renin–angiotensin system participates in angiogenesis, inflammation, vascular remodeling and cellular signalling, theoretical relationships with tumour biology have been proposed. Angiotensin II can influence oxidative stress, inflammatory signalling, vascular endothelial growth factor, proliferation and angiogenesis. However, mechanistic activity within a signalling pathway does not establish carcinogenicity of a drug. Available randomized evidence does not demonstrate an overall cancer excess attributable to ACE inhibitors.

4.2 Angiotensin Receptor Blockers

ARBs became central to the cancer debate for two different reasons: hypotheses concerning RAAS biology and the discovery of nitrosamines in selected products. Randomized evidence has not demonstrated convincing intrinsic class-wide carcinogenicity. The valsartan episode is best understood as a pharmaceutical-quality event in which manufacturing chemistry and product-specific contamination created an exposure to a potentially carcinogenic impurity.

4.3 β-Blockers

β-adrenergic signalling influences angiogenesis, inflammation, immune regulation and cellular migration, generating hypotheses about tumour biology. Nevertheless, randomized evidence does not show a consistent increase in overall cancer among β-blocker users. Continued surveillance is appropriate because of the very large population exposure, but current evidence does not justify classifying β-blockers as carcinogens.

4.4 Calcium-Channel Blockers

CCBs have generated periodic observational signals involving overall and site-specific cancer outcomes. The randomized individual-participant analysis identified a small active-comparator association, whereas placebo comparisons did not demonstrate a statistically significant excess. Contemporary cohort evidence similarly suggests that any association is modest and sensitive to follow-up and analytic choices. CCBs therefore represent an area for continued research rather than an established carcinogenic class.

4.5 Thiazide and Thiazide-Like Diuretics

Thiazide diuretics should not be treated as pharmacologically or epidemiologically interchangeable. HCTZ has a substantially stronger skin-cancer signal than several other thiazide or thiazide-like agents. Evidence for indapamide and bendroflumethiazide is more limited or heterogeneous, and the available literature does not support assuming a class effect identical to HCTZ.

5. Hydrochlorothiazide and Skin Cancer

HCTZ is photosensitizing, providing a plausible pathway linking cumulative exposure to ultraviolet radiation and cutaneous carcinogenesis. The proposed sequence is: HCTZ exposure → cutaneous photosensitization → UV exposure → photochemical excitation → reactive oxygen species and DNA damage → mutagenesis → keratinocyte carcinogenesis.A 2022 systematic review and meta-analysis involving approximately 17.85 million participants reported associations between HCTZ and non-melanoma skin cancer, squamous-cell carcinoma and melanoma, with stronger signals for high cumulative exposure and SCC. Geographic and population heterogeneity was substantial, and the evidence for some outcomes was weaker than for SCC. A later systematic review likewise found associations across several antihypertensive categories but graded much of the evidence as low or very low quality. The most defensible interpretation is therefore molecule-specific: HCTZ has a clinically relevant epidemiological association with selected cutaneous malignancies, particularly SCC, supported by a plausible photosensitization mechanism. This should not be generalized automatically to every thiazide or thiazide-like diuretic.

6. Nitrosamine Impurities: A Pharmaceutical-Quality Problem

Nitrosamine contamination represents a fundamentally different pathway from intrinsic API carcinogenicity. Selected valsartan, losartan and other products were found to contain NDMA, NDEA or related impurities. Importantly, contamination depended on manufacturing route, process conditions, materials and batches rather than being a universal property of the ARB class.

Nitrosamines can arise through nitrosation of suitable amines in the presence of nitrosating species. Potential contributors include secondary or tertiary amines, nitrite-containing materials, contaminated raw materials, recovered solvents, recycled materials, cross-contamination, formulation conditions, storage and API-specific structural features. NDSRIs are especially challenging because they are structurally related to the API and may lack extensive compound-specific carcinogenicity data.

The toxicological pathway is well established in principle: nitrosamine exposure → absorption → metabolic activation, often involving cytochrome P450 enzymes → reactive alkylating intermediates → DNA adduct formation → replication errors and mutational accumulation → potential clonal expansion and carcinogenesis. The presence of a nitrosamine impurity therefore creates a chemically credible hazard, while actual human risk depends on concentration, exposure duration, metabolism and other determinants.

7. Regulatory and Analytical Science

The U.S. FDA's September 2024 revised guidance distinguishes small-molecule nitrosamines from NDSRIs and recommends risk assessment, testing and controls to prevent or reduce unacceptable nitrosamine levels. FDA's current acceptable-intake framework uses compound-specific data, read-across and the predicted carcinogenic potency categorization approach, with updates continuing as new scientific information becomes available.

Analytical control should combine sensitive targeted quantification with broader screening where appropriate. GC-MS and GC-MS/MS remain useful for volatile nitrosamines; LC-MS/MS supports targeted analysis of a broad range of compounds; LC-HRMS and QTOF-MS are useful for suspect and non-targeted screening, particularly for less-characterized NDSRIs. Stable-isotope-labelled internal standards are preferred where available.

A robust workflow can be expressed as: sample extraction → internal standardization → chromatographic separation → MS/MS or HRMS detection → calibration → recovery and precision assessment → limit-of-quantification determination → matrix-effect evaluation → retention-time and ion-ratio confirmation. Analytical methods should be appropriately validated for the intended matrix and concentration range.

8. Epidemiological Bias and Interpretation

Cancer outcomes should not be pooled indiscriminately. Overall cancer, melanoma, basal-cell carcinoma and squamous-cell carcinoma are biologically and epidemiologically distinct endpoints. Relative risks should also be interpreted alongside absolute risks, because a modest relative increase may correspond to a small absolute change in a low-baseline-risk population.

9. Integrated Cancer-Safety Decision Framework

A useful framework for a future signal is: epidemiological association → consistency assessment → biological plausibility → dose-response evaluation → pharmaceutical-impurity assessment → integration of randomized and observational evidence → long-term pharmacovigilance → integrated risk–benefit assessment. Where an impurity is identified, exposure reconstruction should be product- and batch-specific whenever feasible.

The most informative future databases would connect patient, prescribed product, batch, manufacturer, API supplier, manufacturing route and measured impurity profile. This would substantially improve exposure classification and allow epidemiological studies to distinguish API effects from impurity-mediated effects.

10. Future Research Directions

Future studies should combine pharmacoepidemiology with pharmaceutical chemistry and molecular toxicology. Priorities include batch-level exposure reconstruction; high-resolution mass-spectrometric screening; DNA-adduct analysis; cancer mutational signatures; biomarker-based exposure assessment; long-term cancer registries; and mechanistic studies of NDSRI formation. Quasi-experimental analyses of treatment switching may be informative for HCTZ, provided that clinical reasons for switching are carefully addressed.