Abstract
BACKGROUND: We investigated the associations of postdiagnostic dietary glycemic index (GI), glycemic load (GL), insulin index (II), and insulin load (IL) with breast cancer-specific and all-cause mortality. METHODS: Among 8,932 women with stage I-III breast cancer identified in the Nurses' Health Study (NHS; 1980-2010) and NHSII (1991-2011), we prospectively evaluated the associations between postdiagnostic GI, GL, II, and IL, and breast cancer-specific and all-cause mortality. Participants completed a validated food frequency questionnaire every 4 years after diagnosis. RESULTS: During follow-up by 2014 in the NHS and 2015 in the NHSII, 2,523 deaths, including 1,071 from breast cancer, were documented. Higher postdiagnostic GL was associated with higher risk of both breast cancer-specific mortality [HR(Q5vsQ1) = 1.33; 95% confidence interval (CI) = 1.09-1.63; P (trend) = 0.008] and all-cause mortality (HR(Q5vsQ1) = 1.26; 95% CI = 1.10-1.45; P (trend) = 0.0006). Higher all-cause mortality was also observed with higher postdiagnostic GI (HR(Q5vsQ1) = 1.23; 95% CI = 1.08-1.40; P (trend) = 0.001), II (HR(Q5vsQ1) = 1.20; 95% CI = 1.04-1.38; P (trend) = 0.005), and IL (HR(Q5vsQ1) = 1.23; 95% CI = 1.07-1.42; P (trend) = 0.0003). The associations were not modified by insulin receptor or estrogen receptor status of the tumor, or body mass index. CONCLUSIONS: We found that higher dietary GL, reflecting postprandial glucose response, after a breast cancer diagnosis was associated with higher risk of breast cancer-specific mortality. Higher dietary GI, GL, II, and IL after a breast cancer diagnosis were associated with higher risk of death from any cause. IMPACT: These results suggest that carbohydrate quantity and quality may be important in breast cancer prognosis.See related commentary by McTiernan, p. 252.