Background This study assessed the prevalence and risk factors of unhealthy behaviors among survivors of early‐stage breast cancer. Methods Women (n = 9556) from the CANcer TOxicity cohort (NCT01993498) were included. Physical activity (PA), tobacco and alcohol consumption, and body mass index were assessed at diagnosis and at years 1 and 2 after diagnosis. A behavior was defined as unhealthy if patients failed to meet PA recommendations (≥10 metabolic equivalent task hours per week), reduce/quit tobacco, or decrease alcohol consumption to less than daily, or if they gained substantial weight over time. Multivariable‐adjusted generalized estimating equations explored associations with unhealthy behaviors. Results At diagnosis, 41.7% of patients were inactive, 18.2% currently used tobacco, 14.6% consumed alcohol daily, and 48.9% were overweight or obese. At years 1 and 2, unhealthy PA behavior was reported among 37.0% and 35.6% of patients, respectively, unhealthy tobacco use behavior was reported among 11.4% and 9.5%, respectively, and unhealthy alcohol behavior was reported among 13.1% and 12.6%, respectively. In comparison with the previous assessment, 9.4% and 5.9% of underweight and normal‐weight patients had transitioned to the overweight or obese category at years 1 and 2, respectively, and 15.4% and 16.2% of overweight and obese patients had gained ≥5% of their weight at years 1 and 2, respectively. One in 3 current tobacco smokers and 1 in 10 daily alcohol users reported improved behaviors after diagnosis. Older women (5‐year increment) were more likely to be inactive (adjusted odds ratio [aOR], 1.03; 95% confidence interval [CI], 1.01‐1.05) and report unhealthy alcohol behavior (aOR, 1.28; 95% CI, 1.23‐1.33) but were less likely to engage in unhealthy tobacco use (aOR, 0.81; 95% CI, 0.78‐0.85). Being at risk for depression (vs not being at risk for depression) was associated with reduced odds of unhealthy tobacco use (aOR, 0.67; 95% CI, 0.46‐0.97) and with a higher likelihood of unhealthy alcohol behavior (aOR, 1.58; 95% CI, 1.14‐2.19). Women with a college education (vs a primary school education) less frequently reported an unhealthy PA behavior (aOR, 0.61; 95% CI, 0.51‐0.73) and were more likely to report unhealthy alcohol behavior (aOR, 1.85; 95% CI, 1.37‐2.49). Receipt of chemotherapy (vs not receiving chemotherapy) was associated with higher odds of gaining weight (aOR, 1.51; 95% CI, 1.23‐1.87) among those who were overweight or obese at diagnosis. Conclusions The majority of women were adherent to healthy lifestyle behaviors at the time of their breast cancer diagnosis, but a significant subset was nonadherent. Unhealthy behaviors tended to persist after the breast cancer diagnosis, having varying clinical, psychological, sociodemographic, and treatment‐related determinants. This study will inform more targeted interventions to promote optimal health.

Expression of some LXR-dependent genes is related to breast tumor characteristics, but not time to recurrence. This may be due to a lack of study power or too short a follow-up time.

No abstract

No abstract

Background: Persistence of unhealthy behaviors may adversely affect breast cancer (BC) outcomes and quality of life. Cancer diagnosis can represent a teachable moment and a powerful catalyst to correct such behaviors. It has been suggested that some cancer patients (pts) may be motivated towards a healthier lifestyle, particularly in case of a worse cancer-related prognosis. In this study we aimed at understanding the evolution and barriers in adoption of health behaviors after BC. Methods: We analyzed 8580 pts enrolled from 2012-2015 in a nationwide, multicenter study of stage I-III BC (CANTO, NCT01993498). Pts were longitudinally assessed at diagnosis (baseline), year(y)1 and y2 post-diagnosis. Recreational physical activity (PA) was reported using the GPAQ-16. An unhealthy smoking behavior was defined as active smoker status and an unhealthy alcohol behavior as consumption of ≥1 drink/day at y1 and y2. First, we described the longitudinal evolution of health behaviors. Then, we examined factors associated with health behaviors, to define pts that fail to adopt healthy lifestyle changes. Random effect mixed models evaluated changes in PA over time. Logistic regression models estimated odds of unhealthy smoke and alcohol behavior at y1 and y2. All models were adjusted for tumor, including BC stage, clinico-behavioral and treatment characteristics. Results: Mean age at diagnosis was 55 y (SD 11), 9% pts had stage III BC, 49% were overweight/obese (body mass index [BMI] ≥25 Kg/m2), 41% did not meet PA recommendations (<10 MET-h/week), 18% were active smokers, 14% reported ≥1 drink/day. Some small increases in PA were reported in the overall cohort (mean change from baseline [95%CI], MET-h/week: to y1 +0.8 [-0.2 to +1.9]; to y2 +1.2 [+0.1 to +2.3]). Particularly, there were significant increases in PA among pts that did not use to meet PA recommendations at baseline, regardless of BC stage (mean change from baseline [95%CI], MET-h/week: to y1 +8.7, to y2 +9.7 among stage I/II pts; to y1 +12.9, to y2 +10.4 among stage III pts [all p<0.0001]). However, pts who had stage I/II BC and who were overweight/obese failed to improve their PA uptake over time (p>0.05 at both time points). Among pts who were active smokers at baseline, 37% at y1 and 30% at y2 reported reduced tobacco, and among those who used to have ≥1 drink/day at baseline, 11% at y1 and 8% at y2 reported reduced alcohol consumption. 11% pts at y1 and 12% pts at y2 had unhealthy smoke behavior whereas 13% pts at y1 and 14% pts at y2 had unhealthy alcohol behavior (Cochran-Armitage trend test: p=0.034 for smoke and p=0.193 for alcohol). Older age (adjusted odds ratio [aOR] for a 5-year increase in age, 0.69 [95%CI 0.66-0.73]) and being in a relationship (aOR vs not, 0.66 [95% CI 0.54-0.82]) were associated with reduced odds of unhealthy smoke behavior, although older pts (aOR for a 5-year increase in age, 1.34 [95%CI 1.22-1.40]) and those in a relationship (aOR vs not, 1.35 [95% CI 1.10-1.68]) were also more prone to unhealthy alcohol behavior. Lower BMI was consistently associated with higher odds of unhealthy smoke and alcohol behavior (aOR for each decreasing BMI unit: for smoke behavior, 1.05 [95%CI 1.03-1.07]; for alcohol behavior, 1.02 [95% CI 1.01-1.04]). BC stage was not associated with smoke and alcohol behavior. Conclusions: This large clinical study suggests that a substantial proportion of pts pursue a healthy lifestyle after BC. However, there are still survivors that fail to adopt positive and durable lifestyle changes. Uptake of increased levels of PA was not homogenous in this cohort, and we found an upward trend in unhealthy smoke behavior and steady rates of unhealthy alcohol behavior over time. We defined diverse behavioral profiles of pts, according, among others, to age, social status, BMI and BC characteristics. These data will help better inform targeted lifestyle interventions to improve health behaviors after BC. Citation Format: Antonio Di Meglio, Arnauld S Gbenou, Elise Martin, Laurence Vanlemmens, Charles Guenancia, Olivier Rigal, Marion Fournier, Patrick Soulie, Marie-Ange Mouret-Reynier, Carole Tarpin, Florence Boiffard, Sophie Guillermet, Sibille Everhard, Anne-Laure Martin, Sylvie Giacchetti, Thierry Petit, Florence Dalenc, Philippe Rouanet, Antoine Arnaud, Fabrice Andre, Ines Vaz-Luis. Lifestyle changes after breast cancer: A prospective study among 8580 women [abstract]. In: Proceedings of the 2019 San Antonio Breast Cancer Symposium; 2019 Dec 10-14; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2020;80(4 Suppl):Abstract nr P2-13-03.