|
Sub-category:
Adjuvant Therapy
Category:
Breast Cancer
Meeting:
2002 ASCO Annual Meeting
Session Type and Session Title:
Poster Discussion, Breast Cancer
Abstract No:
178
Citation:
Proc Am Soc Clin Oncol 21: 2002 (abstr 178)
Author(s):
Pamela J Goodwin, Marguerite Ennis, Hood Nicky, Kathleen I Pritchard, Maureen E Trudeau, Jarley Koo, Samuel Lunenfeld Research Institute at Mount Sinai Hospital, University of Toronto, Toronto, ON, Canada; Samuel Lunenfeld Research Institute at Mount Sinai Hospital, Toronto, ON, Canada; Toronto-Sunnybrook Regional Cancer Center, University of Toronto, Toronto, ON, Canada; St. Michael's Hospital, University of Toronto, Toronto, ON, Canada.
Abstract:
Diet, particularly fat intake, has been postulated to influence BC prognosis but existing evidence is weak and inconsistent. In a prospective cohort study we confirmed a previously reported curvilinear association of BMI with overall survival (OS) (ASCO 2000 Abstract 272), the best OS occurring when BMI was 20-25 kg/m2. We now seek evidence of similar curvilinear (quadratic) associations of major energy sources with OS. METHODS: 477 women with surgically resected T1-3, N0-1, M0 BC completed the Block Food Frequency Questionnaire 65.3¦32.5 days post-diagnosis, reporting intake over the preceding 12 months. Data on tumor-related factors, treatment and outcomes were obtained prospectively from medical records. Cox survival analyses were conducted after 52 deaths (median follow-up 6.1 years). All p values are for quadratic models. RESULTS: Mean age was 50.4¦9.8. 259 women had T1, 153 T2, and 22 T3 tumors; 146 were N1. 135 received adjuvant chemotherapy, 141 hormone therapy and 46 both. For all major energy sources Cox models identified significant curvilinear (quadratic) associations with OS, that is, OS was best for a mid-range quintile (Q) of daily intake and worse for higher and lower quintiles of intake. Significance persisted after multivariate adjustment for total energy intake, BMI, T and N stage and adjuvant therapy. For FAT (gm), the unadjusted hazard ratio (HR) was 3.6 in Q1 (lowest intake), 2.5 in Q2, 1.0 in Q3 (68.2-83.0 gm), 2.1 in Q4 and 2.5 in Q5, univariate p= 0.007, multivariate p=0.003. For PROTEIN (gm) the HR was lowest in Q4 (70.6-84.9 gm), univariate p=0.04, multivariate p=0.03. For ALCOHOL (drinks) the HR was lowest in Q4 (2.1-5.9 drinks), univariate p=0.004, multivariate p=0.008. For CARBOHYDRATES (% cal) the HR was lowest in Q3 (42.5-46.5 % cal), univariate p=0.001, multivariate p=0.001. The magnitude of HRs for protein, alcohol and carbohydrates was similar to those for fat. CONCLUSIONS: Our results suggest the association of major dietary sources of energy with OS is not linear: mid-range intake of each source is associated with the most favorable OS, extremes were associated with less favorable OS.
 Associated Presentation(s):
Other Abstracts in this Sub-Category:
Abstracts by Pamela J Goodwin:
Presentations by Pamela J Goodwin:
Educational Book Manuscripts by Pamela J Goodwin:
|
