Background and aims Skin cancers are the most common cancers in Caucasian populations worldwide and their incidence is rising. Ultraviolet radiation (UV) exposure is currently the sole environmental factor on which skin cancer prevention can be based. Given the oxidant effects of UV leading to skin carcinogenesis, dietary consumption of antioxidants has been proposed to reduce skin cancer risk. However, previous findings were inconsistent. The Mediterranean diet (MD) has long been reported to reduce the risk of several diseases including cancers. While previous studies explored the relationships between major food groups or individual dietary components of the MD and skin cancer risk, no previous study has investigated adherence to a MD pattern in relation to skin cancer risk to date. Methods E3 N is an ongoing prospective cohort of 98,995 French women aged 40–65 years at inclusion in 1990. Participants completed self-administered questionnaires sent biennially. Skin cancer cases were confirmed through pathology reports. Dietary data were collected via a validated food frequency questionnaire in 1993. Adherence to a MD pattern was assessed using the 9-unit dietary score proposed by Trichopoulou et al. in 2005, which takes values from 0 to 9 points (minimum to maximum adherence) according to the combined intake of fruits, vegetables, legumes, cereals, lipids, fish, dairy products, meat products, and alcohol. In this revised version of the MD score, lipid intake was assessed by calculating the ratio of unsaturated (i.e. mono- and polyunsaturated) to saturated fatty acids. We used Cox proportional hazards regression models to compute hazard ratios (HR) and 95% confidence intervals (CI) adjusted for age, birth cohort, pigmentary traits, residential UV exposure at birth and at inclusion, family history of skin cancer, total energy intake, body mass index, physical activity, smoking status, education and coffee intake. Results Between 1993 and 2008, 404 melanoma, 232 squamous-cell carcinoma (SCC), and 1367 basal-cell carcinoma (BCC) cases were ascertained among the 67,332 included women. The MD score distribution was as follows: Tertile 1: 0–3 (29.4%), Tertile 2: 4–5 (45.2%) and Tertile 3: 6–9 (25.4%). Adherence to the MD was associated with an overall reduction in skin cancer risk (HR = 0.87, 95% CI = 0.77–0.97 for T3 vs.T1, P-trend = 0.01; HR = 0.96, 95% CI = 0.94–0.99 per unit increase in the MD score). When considering skin cancer type, there was a linear inverse association between MD score and risk of non-melanoma skin cancers (HR = 0.87, 95% CI = 0.76–0.99, P-trend = 0.04), particularly BCC (HR = 0.84, 95% CI = 0.72–0.97, P-trend = 0.01). However, while we found no evidence of a linear association with melanoma (P-trend = 0.08), we observed that high MD scores were associated with a borderline reduced risk of melanoma (HR = 0.76, 95% CI = 0.57–1.00) compared with the lowest scores. Moreover, there was no heterogeneity detected across skin cancer types (Pheterogenity = 0.23). Conclusion In this large prospective cohort study, adherence to the MD was linearly associated with a reduction in overall skin cancer risk, mostly driven by an inverse association with BCC; in addition, a high MD score was associated with a decreased melanoma risk. These results suggest that a high level of adherence to the MD pattern may confer protection against skin cancer in women. If confirmed in future research, these findings may ultimately provide additional knowledge for skin cancer prevention.