Tattooing and cervical cancer risk: an unexpected finding from a negative control analysis in the CRABAT cohort

Background: Tattooing is increasingly being investigated as a cancer risk factor, but is also strongly associated with smoking and other lifestyle factors that could confound such associations. To assess this potential confounding, we selected cervical cancer as a negative control, given that nearly all cases are attributable to HPV infection. Methods: We analyzed data from 50,769 women in the Cancer Risk Associated with the Body Art of Tattooing (CRABAT) study nested within the French Constances cohort. We applied latent class analysis (LCA) to identify latent lifestyle profiles from sociodemographic, lifestyle, and health-related variables. Logistic regression models estimated associations between tattoos and cervical cancer, adjusting for confounders and latent lifestyle profile. Additional models were stratified by latent lifestyle profile. Results: In minimally adjusted models, women with tattoos had a higher cervical cancer risk (odds ratio [OR]=1.54 [95% confidence interval [CI] 1.15-2.05]). Associations were stronger for larger tattooed surface area and longer time since tattooing. Associations were attenuated but persisted after adjusting for known confounders (1.33 [0.99-1.77]), and after additionally adjusting for latent lifestyle profile (1.29 [0.95-1.73]). Stratified models showed strong variations, with ORs ranging from 0.99 [0.37-2.61] in the younger profile to 2.56 [1.23-5.34] in the socially-advantaged profile. Conclusions: Despite being selected as a negative control, elevated cervical cancer risk among tattooed women persisted even after comprehensive confounding adjustment. Stronger associations were observed with heavier tattoo exposure. This raises the possibility of a true association involving tattoo-related immune effects on HPV clearance as a potential biological mechanism.