Importance: Type 2 diabetes (T2D) is associated with poor breast cancer prognosis. Adherence to a dietary pattern associated with lower T2D risk may improve breast cancer outcomes.

Objective: To evaluate the association between adherence to a diabetes risk reduction diet (DRRD) after diagnosis and survival outcomes following breast cancer.

Method: We followed 8,320 women diagnosed with stage I-III breast cancer from two large prospective cohort studies, the Nurses’ Health Study (NHS; 1980-2014) and the NHSII (1991-2015). Information on diet and other covariates was repeatedly measured in validated follow-up questionnaires every two to four years. The DRRD adherence score includes 9 dietary components: higher intakes of cereal fiber, coffee, nuts, whole fruits and polyunsaturated:saturated fat ratio; and lower glycemic index of diet, lower intakes of trans fat, sugar-sweetened beverages/fruit juices, and red meat. Cumulative average DRRD score was calculated using repeated measures of diet after breast cancer diagnosis. Deaths were reported by family members or via the search of National Death Index, and cause of death was assigned by reviewing death certificates/medical records. Multivariable-adjusted hazard ratios (HRs) and 95% confidence intervals (CIs) for the association between DRRD score and breast cancer-specific and all-cause mortality were estimated using Cox proportional hazards models.

Results: During a median of 16 years of follow-up after diagnosis, we ascertained 2,146 deaths, of which 948 were due to breast cancer. In the multivariable-adjusted models, women with higher post-diagnostic DRRD scores had a 17% lower risk of breast cancer-specific mortality (top vs. bottom quintile HR =0.83; 95%CI=0.67-1.02; p-trend=0.03) and 33% lower risk of all-cause mortality (HR =0.67; 95%CI=0.58-0.78; p-trend <0.0001). Associations were slightly attenuated after further adjusting for neighborhood socioeconomic status index (breast cancer mortality: HR =0.86; 95%CI=0.70-1.07; p-trend=0.06; overall mortality: HR =0.69; 95%CI=0.60-0.80; p-trend <0.0001). Compared with women who consistently had lower DRRD score (≤median) before and after diagnosis, those who improved their score from low to high had a significantly lower risk of breast cancer mortality (HR =0.81; 95%CI=0.65-1.00).

Conclusion: We found that greater adherence to the DRRD after breast cancer diagnosis was associated with better survival, suggesting dietary modification after diagnosis consistent with T2D prevention may be important to breast cancer survivors.

Table 1. Multivariable HRs and 95% CIs for the association between post-diagnostic diabetes risk reduction diet score and mortality among breast cancer survivors using pooled data from NHS abd NHSII (N=8,320)

Quintile 1Quintile 2Quintile 3Quintile 4Quintile 5p-trend
Median (IQR) 20 (18-21) 24 (23-25) 27 (26-28) 30 (29-31) 34 (33-37)  
Breast cancer-specific mortality       
No. of events (n=948), 589,953 py 207 205 184 176 176  
Model 1 1 (referent) 0.97 (0.80-1.18) 0.80 (0.65-0.97) 0.79 (0.64-0.96) 0.75 (0.61-0.92) 0.0008 
Model 2 1 (referent) 0.88 (0.72-1.07) 0.76 (0.62-0.93) 0.77 (0.63-0.95) 0.83 (0.67-1.02) 0.03 
Model 3 1 (referent) 0.99 (0.81-1.20) 0.82 (0.67-1.00) 0.85 (0.69-1.04) 0.86 (0.70-1.07) 0.06 
       
All-cause mortality       
No. of events (n=2,146), 589,953 py 506 447 476 383 334  
Model 1 1 (referent) 0.83 (0.73-0.94) 0.80 (0.71-0.91) 0.67 (0.59-0.76) 0.57 (0.50-0.65) <.0001 
Model 2 1 (referent) 0.81 (0.71-0.92) 0.87 (0.76-0.98) 0.74 (0.65-0.85) 0.67 (0.58-0.78) <.0001 
Model 3 1 (referent) 0.89 (0.78-1.01) 0.91 (0.80-1.04) 0.78 (0.68-0.90) 0.69 (0.60-0.80) <.0001 
Quintile 1Quintile 2Quintile 3Quintile 4Quintile 5p-trend
Median (IQR) 20 (18-21) 24 (23-25) 27 (26-28) 30 (29-31) 34 (33-37)  
Breast cancer-specific mortality       
No. of events (n=948), 589,953 py 207 205 184 176 176  
Model 1 1 (referent) 0.97 (0.80-1.18) 0.80 (0.65-0.97) 0.79 (0.64-0.96) 0.75 (0.61-0.92) 0.0008 
Model 2 1 (referent) 0.88 (0.72-1.07) 0.76 (0.62-0.93) 0.77 (0.63-0.95) 0.83 (0.67-1.02) 0.03 
Model 3 1 (referent) 0.99 (0.81-1.20) 0.82 (0.67-1.00) 0.85 (0.69-1.04) 0.86 (0.70-1.07) 0.06 
       
All-cause mortality       
No. of events (n=2,146), 589,953 py 506 447 476 383 334  
Model 1 1 (referent) 0.83 (0.73-0.94) 0.80 (0.71-0.91) 0.67 (0.59-0.76) 0.57 (0.50-0.65) <.0001 
Model 2 1 (referent) 0.81 (0.71-0.92) 0.87 (0.76-0.98) 0.74 (0.65-0.85) 0.67 (0.58-0.78) <.0001 
Model 3 1 (referent) 0.89 (0.78-1.01) 0.91 (0.80-1.04) 0.78 (0.68-0.90) 0.69 (0.60-0.80) <.0001 

Table 2. Multivariable hazard ratios (HRs) and 95% confidence intervals (CIs) for the association between changes of diabetes risk reduction diet score before or after diagnosis and mortality among breast cancer survivors using pooled data from NHS abd NHSII (N=8,320)

CharacteristicsHR (95% CI)
Breast cancer-specific mortality  
Cross classified changes  
No. of events (n=948), 589,953 py  
Low to Low 1 (referent) 
Low to High 0.81 (0.65-1.00) 
High to Low 0.97 (0.79-1.19) 
High to High 0.87 (0.74-1.02) 
All-cause mortality  
Cross classified changes  
No. of events (n=2,306), 589,953 py  
Low to Low 1 (referent) 
Low to High 0.88 (0.76-1.02) 
High to Low 0.99 (0.87-1.12) 
High to High 0.80 (0.72-0.89) 
CharacteristicsHR (95% CI)
Breast cancer-specific mortality  
Cross classified changes  
No. of events (n=948), 589,953 py  
Low to Low 1 (referent) 
Low to High 0.81 (0.65-1.00) 
High to Low 0.97 (0.79-1.19) 
High to High 0.87 (0.74-1.02) 
All-cause mortality  
Cross classified changes  
No. of events (n=2,306), 589,953 py  
Low to Low 1 (referent) 
Low to High 0.88 (0.76-1.02) 
High to Low 0.99 (0.87-1.12) 
High to High 0.80 (0.72-0.89) 

Citation Format: Tengteng Wang, Maryam Farvid, Jae Hee Kang, Michelle Holmes, Bernard Rosner, Rulla Tamimi, Walter Willett, A. Heather Eliassen. Diabetes risk reduction diet and survival following breast cancer [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr GS2-09.