Adherence to dietary guidelines and liver cancer risk: Results from two prospective cohort studies

Although dietary factors have been extensively investigated as potential risk factors for liver cancer, the evidence is inconclusive. Our study systematically assessed the associations between ten dietary guidelines compliance scores and liver cancer risk among adult people, and found out the dietary patterns for liver cancer prevention.

Participants of 59,844 men and 72,680 women, aged 40-74 years and living in urban Shanghai, were included in two prospective cohort studies conducted between 2002-2006 and 1996-2000, respectively. Dietary intakes were assessed during baseline in-person interviews using validated food-frequency questionnaires, and dietary guideline compliance scores were calculated by adjusting for total energy intake and adapting existing dietary recommendations. Hazards ratios (HRs) with 95 % confidence intervals (CIs) were evaluated by both tertile categories and per standard deviation (SD) increment using Cox proportional hazard regression models to assess the associations between ten dietary guideline compliance scores and liver cancer risk.

In the two cohorts, 431 male and 256 female incident liver cancer cases were identified during a mean follow-up of 11.90 and 17.44 years, respectively. There were no statistically significant associations between these ten dietary guidelines and male liver cancer risk (P > 0.05). In contrast, only the modified reversed Empirical Dietary Inflammation Pattern (rEDIP) tended to be associated with the low risk of female liver cancer (HR _{T3 vs. T1} = 0.67, 95 % CI: 0.48-0.92, P_trend = 0.016, HR per SD = 0.94, 95 % CI: 0.85-1.03). The inverse associations appeared stronger between rEDIP and liver cancer risk at younger ages (<55 years) in women (HR per SD = 0.91, 95 % CI: 0.84-0.99) compared to the older women (≥55 years). There were suggestive but non-significant inverse associations between the modified Diabetes Risk Reduction Diet (mDRRD) (men: HR per SD = 0.92, 95 % CI: 0.84-1.02; women: HR per SD = 0.97, 95 % CI: 0.84-1.02) and the modified World Cancer Research Fund/American Institute for Cancer Research (mWCRF/AICR) (men: HR per SD = 0.93, 95 % CI: 0.84-1.02; women: HR per SD = 0.91, 95 % CI: 0.80-1.03) and liver cancer incidence. The associations of mDRRD (HR per SD = 0.82, 95 % CI: 0.75-0.98) and mWCRF/AICR (HR per SD = 0.83, 95 % CI: 0.74-0.99) on liver cancer risk were significant in men who ever smoked.

Our findings confirm that greater adherence to some healthy dietary patterns (i.e. rEDIP, mDRRD and mWCRF/AICR) is inversely associated with liver cancer risk, especially in certain populations. Future studies are required to confirm these findings and elucidate potential mechanisms.