The Effect of Diet on Breast Cancer Recurrence: The DIANA-5 Randomized Trial

Abstract

Purpose: The DIANA-5 randomized controlled trial assessed the effectiveness of a diet based on Mediterranean and macrobiotic traditions (macro-Mediterranean diet) in reducing breast cancer recurrence.

Patients and methods: The DIANA-5 study involved 1,542 patients with breast cancer at high risk of recurrence because of estrogen receptor-negative cancer, or metabolic syndrome, or high plasma levels of insulin or testosterone. Women were randomly assigned to an active dietary intervention (IG) or a control group (CG). Both groups received the 2007 American Institute for Cancer Research/World Cancer Research Fund recommendations for cancer prevention. The intervention consisted of meetings with kitchen classes, community meals, and dietary recommendations. Recommended foods included whole grain cereals, legumes, soy products, vegetables, fruit, nuts, olive oil, and fish. Foods to be avoided were refined products, potatoes, sugar and desserts, red and processed meat, dairy products, and alcoholic drinks. A compliance Dietary Index was defined by the difference between recommended and discouraged foods.

Results: Over the 5 years of follow-up, 95 patients of the IG and 98 of the CG developed breast cancer recurrence [HR = 0.99; 95% confidence interval (CI): 0.69-1.40]. The analysis by compliance to the dietary recommendations (IG and CG together) showed that the women in the upper tertile of Dietary Index change had an HR of recurrence of 0.59 (95% CI: 0.36-0.92) compared with women in the lower tertile.

Conclusions: The DIANA-5 dietary intervention trial failed to show a reduction in breast cancer recurrence, although self-reported diet at year 1 in IG and CG combined showed a protective association with the higher Dietary Index change. See related commentary by McTiernan, p. 931.

Figure 1.

CONSORT trial flow diagram.

Figure 2.

Kaplan–Meier plot describes the overall recurrence-free survival between the groups assigned at the time of randomization. The two lines represent overall recurrence-free survival over time (5 years) by randomization group. The blue line is for IG. The dashed green line is for the CG. For each year, the number of patients with breast cancer at risk and HRs are represented. The horizontal axis represents the time of observation. The vertical axis represents the HRs.

Figure 3.

Kaplan–Meier plot describes the recurrence-free survival by Dietary Index (DI) changes. Participants were classified on the basis of change in DI between baseline and the first year. The blue line describes participants in the first tertile of DI change. The red line describes participants in the second tertile of DI change. The green line represents participants in the third tertile of DI change. The horizontal axis represents the time of observation. The vertical axis represents the HRs.