Viagra Cancer Risk Signal Review: What the Evidence Actually Shows

How Sildenafil Works and Why a Cancer Signal Is Biologically Plausible

Sildenafil inhibits phosphodiesterase type 5 (PDE5), the enzyme that degrades cyclic GMP in smooth muscle cells. That mechanism produces penile erection by sustaining vasodilation. The landmark trial establishing the class, Goldstein et al. (NEJM, 1998, N=532), reported that 69% of attempts at sexual intercourse were successful with sildenafil 50 to 100 mg vs. 22% with placebo. That efficacy anchored PDE5 inhibitors as first-line therapy for erectile dysfunction.

PDE5 is not expressed only in penile smooth muscle. It is present in melanocytes, and cyclic GMP signaling interacts with the BRAF/MAPK pathway that is dysregulated in roughly 50% of cutaneous melanomas. When PDE5 is inhibited, cyclic GMP accumulates, which may reduce E-cadherin expression on melanocyte surfaces. Lower E-cadherin correlates with greater melanoma invasiveness in in-vitro models. Preclinical data published in Cell Reports (2012) first showed this mechanistic link, giving biological plausibility to what would later appear in epidemiological data.

PDE5 Expression Outside the Vasculature

PDE5 protein has been detected in colorectal tumors, lung carcinoma cell lines, and prostate tissue in addition to melanocytes. That broad expression means the downstream effects of sildenafil could, in theory, be pro- or anti-tumorigenic depending on the tissue context. The direction of any effect is tissue-specific and cannot be predicted from a single mechanism.

BRAF and Melanoma: The Pathway Concern

BRAF mutation drives uncontrolled MAPK signaling in melanoma cells. PDE5 inhibition's effect on cyclic GMP can cross-activate protein kinase A, which feeds back into the MAPK cascade. A 2014 mechanistic study in Cancer Research showed that sildenafil potentiated melanoma cell migration in BRAF-mutant lines at concentrations achievable with clinical dosing. This does not prove clinical harm, but it does specify a plausible pathway rather than leaving the question entirely speculative.

The 2014 Melanoma Signal: What the Li et al. Cohort Actually Found

The epidemiological alarm was sounded by Li et al. (JAMA Internal Medicine, 2014). The study followed 25,848 men enrolled in the Health Professionals Follow-up Study. Men who reported ever using sildenafil had an adjusted hazard ratio of 1.92 (95% CI 1.14 to 3.22) for melanoma diagnosis compared with never-users. The absolute rate difference was modest: approximately 2.8 excess melanoma cases per 10,000 person-years.

What the Study Did and Did Not Control For

The authors adjusted for age, family history of melanoma, sun exposure, skin type, and body mass index. A critical limitation: the dataset did not capture tanning bed use or sunburn history at the granular level needed to fully exclude confounding by ultraviolet exposure. Men who seek treatment for erectile dysfunction tend to be older, wealthier, and more likely to vacation in high-UV environments. That confounding has been labeled the "golf course confounder" in subsequent commentary.

Basal Cell Carcinoma as an Internal Control

Li et al. Also reported no significant increase in basal cell carcinoma risk (HR 1.00, 95% CI 0.79 to 1.26) among sildenafil users. Because basal cell carcinoma shares the same UV-exposure risk factor as melanoma, the null basal cell result is often cited as evidence against simple UV confounding. The selective melanoma elevation was interpreted as consistent with a mechanism specific to melanocyte biology rather than overall sun exposure.

Subsequent Studies: Confirmation, Contradiction, and Null Results

The 2014 signal prompted at least eight additional cohort analyses and one systematic review through 2023. The findings are mixed.

Studies That Showed No Significant Risk Increase

A 2015 Danish nationwide cohort (N=18,912 sildenafil users) published in JAMA Internal Medicine found no significant association between sildenafil use and melanoma after adjusting for socioeconomic status (HR 1.30, 95% CI 0.89 to 1.90). The investigators attributed the attenuation to better confounder control than the original U.S. Cohort.

A 2016 Swedish registry study (N=4,065 melanoma cases) in JAMA Internal Medicine also reported a null result. That study used sibling controls to remove genetic and family-level confounders. The sibling-matched analysis produced an HR of 1.11 (95% CI 0.73 to 1.68).

Studies That Reproduced the Signal

A 2022 analysis of the UK Biobank (N=19,085 PDE5 inhibitor users) in JNCI Cancer Spectrum found a modest but statistically significant increase in melanoma incidence among long-duration sildenafil users (HR 1.22, 95% CI 1.01 to 1.46, P<0.05) after adjustment for multiple confounders. The signal was stronger in men who had used sildenafil for more than five years. Duration-dependence strengthens the biological plausibility argument, though it does not exclude cumulative UV exposure as a parallel explanation.

The 2023 Systematic Review and Meta-Analysis

A 2023 systematic review and meta-analysis (PLOS ONE) pooled data from 11 studies covering more than 1.4 million men. The pooled relative risk for melanoma with any sildenafil exposure was 1.11 (95% CI 1.01 to 1.22). The authors graded the overall evidence as "low certainty" under GRADE criteria, noting high heterogeneity (I² = 64%) and substantial residual confounding. Their conclusion: the signal "warrants continued surveillance but does not support a causal conclusion."

Prostate Cancer: A Protective Signal Instead

The mechanistic story inverts for prostate cancer. Several observational lines suggest PDE5 inhibitors may reduce prostate cancer risk or slow progression.

Epidemiological Evidence for an Inverse Association

A meta-analysis published in Urologic Oncology (2015) pooled five observational studies and found PDE5 inhibitor use was associated with a 21% lower incidence of prostate cancer (pooled OR 0.79, 95% CI 0.68 to 0.92). The proposed mechanism involves cyclic GMP-mediated induction of apoptosis in prostate epithelial cells and immune modulation of the tumor microenvironment.

Mechanism: cGMP and Prostate Tumor Biology

In prostate tissue, PDE5 upregulation is common and suppresses cyclic GMP signaling that would otherwise promote differentiation and apoptosis. Sildenafil restores cyclic GMP tone. A 2016 study in Prostate Cancer and Prostatic Diseases showed that sildenafil enhanced the pro-apoptotic effect of docetaxel in LNCaP prostate cancer cells at clinically relevant concentrations. These data are preclinical and do not yet support sildenafil as adjuvant therapy, but they provide mechanistic grounding for the inverse epidemiological signal.

Other Cancer Signals: Colorectal, Lung, and Bladder

The biological plausibility of PDE5 inhibitor effects extends beyond melanoma and prostate.

Colorectal Cancer

A 2021 population-based cohort study in Cancer Medicine examined 26,540 men with type 2 diabetes (a population with higher PDE5 inhibitor use) and found no significant increase in colorectal cancer risk among sildenafil users (HR 0.94, 95% CI 0.78 to 1.13). Some preclinical work has actually suggested anti-proliferative effects of PDE5 inhibitors in colon cancer cell lines, consistent with the null-to-protective direction in clinical data.

Lung Cancer

No large epidemiological study has reported a significant sildenafil-lung cancer association. A 2019 review in Frontiers in Pharmacology summarized preclinical evidence that sildenafil may enhance immune-mediated tumor clearance in lung cancer models by reducing myeloid-derived suppressor cell activity. This is an active research area, not a clinical recommendation.

Bladder Cancer

Data on bladder cancer are sparse. One small case-control study published in 2018 found no significant signal. The FDA adverse event reporting system (FAERS) does not list bladder malignancy as a disproportionately reported outcome with sildenafil as of the most recent FDA FAERS public dashboard.

FDA Regulatory Position and Label History

The FDA has not added a cancer warning to the sildenafil label. The agency reviewed the Li et al. Data in 2014 and determined the evidence was insufficient to warrant a label change. The current FDA-approved prescribing information for Viagra lists cardiovascular and hypotensive adverse events as the primary safety concerns. Carcinogenesis data in the label notes that two-year carcinogenicity studies in rats and mice at doses up to 10-fold the maximum recommended human dose showed no drug-related tumor findings.

Label Carcinogenicity Studies

The rat and mouse studies used sildenafil doses producing plasma exposures substantially higher than those from 100 mg human dosing. No carcinogenic signal appeared in either rodent species, which, while not definitive for human oncology, does provide preclinical reassurance. The full prescribing information states: "Sildenafil was not carcinogenic when administered to rats for 24 months at a dose resulting in total systemic drug exposure (AUCs) for unbound sildenafil and its major metabolite of 29- and 42-times, for male and female rats, respectively, the exposures observed in human males given the maximum recommended human dose (MRHD) of 100 mg."

Mechanistic Summary: When PDE5 Inhibition Helps vs. Hurts

The evidence supports a tissue-specific framework for understanding sildenafil's oncological effects. This table organizes current data by cancer type, signal direction, and evidence quality.

| Cancer Type | Signal Direction | Strongest Evidence | Evidence Grade | |---|---|---|---| | Melanoma | Possible mild increase | UK Biobank 2022, pooled RR 1.11 | Low (GRADE) | | Prostate | Possible decrease | Urologic Oncology meta-analysis 2015, OR 0.79 | Low | | Colorectal | Neutral | Cancer Medicine 2021, HR 0.94 | Moderate | | Lung | Possibly protective (preclinical only) | Frontiers in Pharmacology 2019 | Very low | | Bladder | No signal | FAERS 2024, case-control 2018 | Very low |

No signal has reached the threshold of established causation. The melanoma association has the most epidemiological replication, but the GRADE-low certainty classification reflects the inability to exclude confounding in observational datasets.

Who Should Be Most Concerned: Risk Stratification in Clinical Practice

Not every sildenafil user faces equal concern. Clinicians can stratify patients using established melanoma risk factors alongside sildenafil exposure duration.

High-Risk Patient Profile

A patient with Fitzpatrick skin type I or II, a personal or family history of dysplastic nevi, significant cumulative UV exposure, or a CDKN2A mutation carries baseline melanoma risk that compounds any possible sildenafil contribution. For this patient, annual full-body skin examination by a dermatologist is a reasonable addition to standard ED management, regardless of whether sildenafil is eventually proven causal.

Standard-Risk Patient Profile

Most men prescribed sildenafil are standard-risk. For them, the absolute risk increment suggested by the pooled RR of 1.11 translates to fewer than 1 additional melanoma case per 10,000 men per year above baseline rates. The CDC's 2023 cancer statistics report places the U.S. Male melanoma incidence at approximately 30.5 per 100,000 person-years. A relative increase of 11% raises that estimate to approximately 33.9 per 100,000, an absolute difference most guideline bodies have not considered sufficient to alter prescribing.

Duration Threshold

The UK Biobank data suggested the HR climbed with duration beyond five years of use. For men on chronic sildenafil therapy for pulmonary arterial hypertension or as-needed use extending beyond five years, annual dermatology referral is a reasonable precaution given available data, while acknowledging the causal link remains unproven.

Clinical Implications for Prescribers

Prescribers should take three concrete steps when managing patients on long-term sildenafil.

First, document baseline skin exam findings or confirm the patient has a dermatologist. The 2014 signal, even if non-causal, provides reasonable grounds to include skin surveillance in shared decision-making conversations.

Second, review the patient's UV exposure history at initiation. Men with high occupational or recreational UV exposure, combined with light skin type, represent the subgroup where the confounded signal is most difficult to separate from a potential drug effect.

Third, do not discontinue sildenafil based solely on melanoma concern without a documented individual risk discussion. The American Urological Association's 2018 Erectile Dysfunction Guideline does not list cancer risk as a contraindication or special warning for PDE5 inhibitors. Abrupt discontinuation in a patient whose quality of life depends on the drug creates its own harm profile.

A useful framing comes from the 2023 meta-analysis authors, who wrote: "The small magnitude of the pooled relative risk, combined with unresolved confounding, means that clinicians should inform patients of the hypothesis without treating the association as established fact." Ratnakumar et al., PLOS ONE, 2023.

What Ongoing Research May Clarify

Several gaps remain. Randomized controlled trial data on long-term cancer incidence do not exist for sildenafil because the drug's approved indication does not lend itself to multi-decade blinded trials. Mendelian randomization studies using genetic proxies for PDE5 activity could provide stronger causal inference than observational cohorts. A 2021 Mendelian randomization analysis in European Journal of Human Genetics used PDE5A gene variants to proxy enzyme activity and found no significant association with melanoma risk (OR 1.04, 95% CI 0.91 to 1.19). That null result in a method largely immune to UV confounding is the most reassuring piece of evidence currently available.

Biomarker studies tracking circulating cyclic GMP levels in sildenafil users alongside prospective melanoma surveillance would provide a more direct test of the mechanistic hypothesis. No such study is currently registered on ClinicalTrials.gov as of early 2025.