A treelet transform analysis to relate nutrient patterns to the risk of hormonal receptor-defined breast cancer in the European Prospective Investigation into Cancer and Nutrition (EPIC)

Abstract

Pattern analysis has emerged as a tool to depict the role of multiple nutrients/foods in relation to health outcomes. The present study aimed at extracting nutrient patterns with respect to breast cancer (BC) aetiology. Nutrient patterns were derived with treelet transform (TT) and related to BC risk. TT was applied to twenty-three log-transformed nutrient densities from dietary questionnaires. Hazard ratios (HR) and 95 % confidence intervals computed using Cox proportional hazards models quantified the association between quintiles of nutrient pattern scores and risk of overall BC, and by hormonal receptor and menopausal status. Principal component analysis was applied for comparison. The European Prospective Investigation into Cancer and Nutrition (EPIC). Women (n 334 850) from the EPIC study. The first TT component (TC1) highlighted a pattern rich in nutrients found in animal foods loading on cholesterol, protein, retinol, vitamins B12 and D, while the second TT component (TC2) reflected a diet rich in β-carotene, riboflavin, thiamin, vitamins C and B6, fibre, Fe, Ca, K, Mg, P and folate. While TC1 was not associated with BC risk, TC2 was inversely associated with BC risk overall (HRQ5 v. Q1=0·89, 95 % CI 0·83, 0·95, Ptrend<0·01) and showed a significantly lower risk in oestrogen receptor-positive (HRQ5 v. Q1=0·89, 95 % CI 0·81, 0·98, Ptrend=0·02) and progesterone receptor-positive tumours (HRQ5 v. Q1=0·87, 95 % CI 0·77, 0·98, Ptrend<0·01). TT produces readily interpretable sparse components explaining similar amounts of variation as principal component analysis. Our results suggest that participants with a nutrient pattern high in micronutrients found in vegetables, fruits and cereals had a lower risk of BC

Dietary Patterns and Breast Cancer Risk: A Systematic Review

“…The initial search resulted in 136 articles, of which 102 articles were excluded at the title and abstract level. Thirty-four full-length articles were selected for further evaluation and 18 of these were excluded for the following reasons: examined diet as a predictor of breast density not cancer (n=3) (24)(25)(26), the combined effect of diet and physical activity was studied (n=2) (27,28), statistical methods other than those specified in our inclusion criteria were used (n=2) (29,30), and evaluated various forms of vegetarian diet (derived without using any dietary index or factor analysis) (n=1) (31), hormone levels rather than diet were studied (n=3) (32)(33)(34), the effect of diet on breast cancer markers at the molecular level were studied (n=5) (35)(36)(37)(38)(39), assessed the dietary intake of women during their adolescent years (n=1) (40), used an average/reference dietary pattern to derive the risk estimates (n=1) (41), one paper was identified by hand searching the selected articles (42). This resulted in the selection of 17 original research studies published between January 2013 and May 2017 (42)(43)(44)(45)(46)(47)(48)(49)(50)(51)(52)(53)(54)(55)(56)(57)(58), where 13 of these used the posteriori approach while 2 of the 13 also included an a priori approach to identify dietary patterns.…”

Patterns in metabolite profile are associated with risk of more aggressive prostate cancer: A prospective study of 3,057 matched case–control sets from EPIC

“…To inform the choice of number of factors to retain and the corresponding cut‐level, we ran treelet transform for different numbers of retained components (1–10), identified the optimal cut‐level for each (ranging from 97 to 104) using cross‐validation, inspected scree plots (Supporting Information Fig. S1) and assessed pattern interpretability . The final treelet transform was then computed specifying the number of components to retain and the lowest optimal cut‐level.…”

“…S1) 17 and assessed pattern interpretability. 18 The final treelet transform was then computed specifying the number of components to retain and the lowest optimal cut-level. Finally, component scores were calculated for all participants.…”

Are women with breast cancer more likely to follow dietary recommendations and have sufficient vitamin A intake compared to healthy controls?

“…In addition, they compared total retinol intake, considering the highest value approximately 6000 mg/day and the lowest intake around 1000 mg/day. They analyzed 24 studies and found a 6% reduction in the risk of developing breast cancer (HR grouped)= 0.94, 95% CI= 0.89-0.99, p= 0.01) 26 .…”