目的:最近的证据表明植物性饮食可以降低乳腺癌(BC)的风险。然而,植物性饮食的大量营养素组成及其对BC风险的潜在影响尚未得到很好的研究。该分析使用多维方法调查了一系列基于植物的饮食指数中大量营养素组成与BC风险的关联。
方法:这项研究追踪了1993年至2014年全国教育(E3N)队列中64,655名参与者。在基线(1993年)和随访(2005年)时,使用经过验证的208项饮食史问卷对饮食进行评估,计算对整体植物性饮食(PDI)的依从性,健康的植物性饮食(hPDI),和不健康的植物性饮食(uPDI)。通过跨这些指标的不同水平的广义加性时间依赖性Cox模型评估了大量营养素组成与BC风险的关联。在25日生成响应面以可视化成分关联,50岁,和每个指数的第75百分位数(低,中度,和高)。
结果:在21年的随访中,共发现3,932例BC病例。大量营养素组成与hPDI的BC风险之间存在显着关联,uPDI,和PDI(均P<0.001)。Akaike信息标准支持hPDI模型来表征大量营养素与BC之间的关联。具有较低hPDI评分的个体的BC风险最高,这些个体也食用含有较低蛋白质(10%)的饮食,较低的碳水化合物(35%),和更高的脂肪(55%)。在hPDI评分较高、蛋白质摄入量最低(10%)的人群中观察到BC的风险最低。在较高的PDI和uPDI时,含有较高蛋白质(30%)和脂肪(45%)的饮食具有最高的BC风险.
结论:这些结果证明了大量营养素组成之间的复杂关系,植物性饮食质量,BC风险。需要进一步的研究来检查可能推动这些关联的特定食物。
■该方案在clinicaltrials.gov注册为NCT03285230。
OBJECTIVE: Recent evidence suggests that plant-based diets may reduce the risk of breast cancer (BC). However, the macronutrient composition of plant-based diets and its potential impact on BC risk has not been well explored. This analysis investigated the association of macronutrient composition with BC risk across a spectrum of plant-based diet indexes using a multidimensional approach.
METHODS: This study followed 64,655 participants from the Etude Epidémiologique auprès de femmes de la Mutuelle Générale de l\'Education Nationale (E3N) cohort from 1993 to 2014. Diets were evaluated using validated 208-item diet history questionnaires at baseline (1993) and follow-up (2005), to calculate adherence to the overall plant-based diet (PDI), healthful plant-based diet (hPDI), and unhealthful plant-based diet (uPDI). The association of macronutrient composition with BC risk was assessed via generalized additive time-dependent Cox models across different levels of these indexes. Response surfaces were generated to visualize compositional associations at the 25th, 50th, and 75th percentile of each index (low, moderate, and high).
RESULTS: A total of 3,932 incident BC cases were identified during the 21-year follow-up. There was a significant association between macronutrient composition and BC risk for hPDI, uPDI, and PDI (all P < 0.001). Akaike information criterion favored the hPDI model for characterizing the association between macronutrients and BC. BC risk was highest for individuals with a lower hPDI score who also consumed a diet containing lower protein (10%), lower carbohydrate (35%), and higher fat (55%). The lowest risk of BC was observed in those with higher hPDI scores with the lowest intake of protein (10%). At higher PDI and uPDI, diets containing higher protein (30%) and fat (45%) had the highest BC risk.
CONCLUSIONS: These results demonstrate a complex relationship between macronutrient composition, plant-based diet quality, and BC risk. Further research is needed to examine specific foods that may be driving these associations.
UNASSIGNED: The protocol is registered at clinicaltrials.gov as NCT03285230.