(083) Incident Melanoma in HPV-Positive Adults: A Propensity-Matched Retrospective Cohort Analysis

Abstract Introduction High-risk human papillomavirus (HPV) genotypes are well-established in their contributions to epithelial cancers, but their role in melanoma is not well defined. Prior epidemiological analyses report increased melanoma risk among HPV-positive individuals and potential pathogenic mechanisms between HPV-induced immune evasion and ultraviolet-induced DNA damage. However, large-scale population-based analyses evaluating the role of HPV in melanoma development have not yet been conducted. In this retrospective cohort analysis, we evaluated the incidence of melanoma in adults with HPV using a global health database. Objective The objective of this study was to assess whether individuals with known HPV have a greater risk of developing incident melanoma as compared to those without HPV. Methods Data was extracted from TriNetX Global Collaborative Network, a large-scale electronic health record database. Adults (18+) with documented HPV (n = 788 196; identified by ICD-10 codes B97.7, R87.810, R87.811, A63.0) without history of HIV, sexually transmitted infections, or immunomodulator use were compared to age-, race-, and sex-matched “healthy” controls (n = 14 660 193). Following propensity-score matching for demographics, cohorts included 756 206 individuals per group. Outcomes were assessed beginning at 180 days after the index event (first HPV-related ICD-10-CM) to minimize detection bias. Incident melanoma was defined using general, site-specific, and in-situ ICD-10 codes. Any coded instances of melanoma prior to the desired time frame were excluded. Results Incident melanoma occurred in 9303 of 741 897 (1.254%) HPV patients and 8563 of 749 893 (1.142%) controls, corresponding to a risk difference of 0.112% (95% CI 0.077–0.147, p 0.001), a risk ratio of 1.098 (95% CI 1.067–1.131), and an odds ratio of 1.099 (95% CI 1.067–1.132). Independent t-test analysis showed that HPV-positive patients had a higher mean number of melanoma events (4.76 ± 11.02) than controls (4.11 ± 9.52). Conclusions In our large, propensity-matched network analysis, HPV was associated with a modest but statistically significant increase in incident melanoma cases. While the TriNetX platform limited the number of covariates included per model, clinically relevant variables were selected to optimize adjustment. Although residual confounding is possible and some differences in the other covariates may remain, the use of propensity matching across a large, multicenter network strengthens the generalizability of our findings. Further research is needed to clarify the underlying mechanisms and true prevalence of melanoma in HPV-positive individuals. Our findings suggest that HPV may exert oncogenic effects beyond traditional mucocutaneous disease. Future studies integrating comorbid risk factors, immunosuppressive states, and shared molecular pathways are warranted to better define the relationship between HPV and melanoma. Disclosure No.