Adherence to the Mediterranean Diet and Obesity-Linked Cancer Risk in EPIC

Abstract

Importance: Adherence to the Mediterranean Diet (MedDiet) has been associated with a lower incidence of cancer and reduced weight gain. These associations suggest a potential role for the MedDiet in lowering the risk of obesity-related cancers (ORCs). Obesity is a known risk factor for various cancers and shows an inverse association with MedDiet adherence.

Objective: To examine the association between adherence to the MedDiet and the risk of ORCs, considering the possible mediating role of adiposity.

Design, setting, and participants: This prospective cohort study analyzed data from the European Prospective Investigation Into Cancer and Nutrition (EPIC) study, which enrolled participants aged 35 to 70 years from 1992 to 2000 across 23 centers in 10 countries. The data analysis was conducted from March 1 to May 31, 2023.

Exposures: Dietary intake before baseline was evaluated using country-specific, validated questionnaires administered at recruitment. Adherence to the MedDiet was scored on a 9-point scale and categorized as low (0-3 points), medium (4-6 points), or high (7-9 points).

Main outcomes and measures: The primary outcome was the incidence of ORCs, classified according to the 2015 International Agency for Research on Cancer criteria. Multivariable Cox proportional hazards regression models were used to assess the association between MedDiet adherence and ORC incidence. Mediation analyses were conducted to evaluate the role of waist to hip ratio and body mass index in this association.

Results: A total of 450 111 participants were included in the study (mean [SD] age, 51.1 [9.8] years; 70.8% women) and followed up during a median (IQR) time of 14.9 (4.1) years. Among participants, 4.9% experienced an ORC (rates, 0.053, 0.049, and 0.043 per person-year in the low, medium, and high MedDiet adherence groups, respectively). Participants with high adherence to the MedDiet (7-9 points) had a lower risk of ORC compared with those with low adherence (0-3 points) (hazard ratio [HR], 0.94; 95% CI, 0.90-0.98). A similar inverse association was observed for participants with medium adherence (4-6 points vs 0-3 points). However, mediation analyses did not show associations of waist to hip ratio or body mass index between MedDiet adherence and ORC risk.

Conclusions and relevance: These findings indicate that higher adherence to the MedDiet is associated with a modest reduction in the risk of ORCs, independent of adiposity measures. Further research is needed to clarify the mechanisms by which the MedDiet may contribute to cancer prevention.

Figure 1.. Flowchart of European Prospective Investigation Into Cancer and Nutrition (EPIC) Cohort

Figure 2.. Assessment of Linear Associations Between Mediterranean Diet Score and Incident Obesity-Related Cancer (ORC) Using Restricted Cubic Splines

The P values were obtained by testing for nonlinearity using a likelihood ratio test comparing 2 multivariable nested models: 1 model with only a linear term and 1 with a linear term and restricted cubic spline terms (P = .79). The reported values are in terms of the logarithmic hazard ratio (logHR), which can be converted to HR using the equation HR = exp(logHR). For example, the logHRs of 0.20, −0.20, and −0.30 correspond to HRs of 1.22, 0.82, and 0.74, respectively. In addition, logHRs of 0.10, −0.05, and −0.10 correspond to HRs of 1.10, 0.95, and 0.90, respectively. The fully adjusted model was stratified by country and sex and adjusted for age at recruitment (in 1-year categories), attained level of education (primary school, technical school, secondary school, university degree), physical activity (inactive, moderately inactive, moderately active, active), smoking status (never smoker, former smoker, current smoker) and intensity at recruitment, height, alcohol intake at recruitment (in grams per day), total energy intake at recruitment (kilocalories per day), and history of type 2 diabetes (yes, no).