Search

10552 Background: In the WHI DM randomized trial, randomization to the dietary intervention group was associated with a 21% lower breast cancer mortality (P = 0.02) (JCO 2020), and while not an intervention target, with higher physical activity as well. Therefore, we examined whether these lifestyle changes attenuate age-related physical functioning decline. Methods: From 1993-1998, 48,835 postmenopausal women, aged 50-79 years, were randomized to dietary intervention or usual diet comparison groups through 8 years intervention and 19 years cumulative follow-up. Breast cancer findings, as primary outcome, have been reported. Physical functioning was assessed using the RAND 36-Item Short Form Health Survey (SF-36), which assessed limitations of 10 hierarchical physical activities, scored from 0 to 100, with a higher score indicating less limited physical function. The trajectory of longitudinal physical functioning was the primary study outcome, assessed by comparing findings in the two randomization groups, overall, and by baseline physical activity and age decade. Additionally, findings were reported against a disability threshold (when assistance in daily activities is required). Results: Physical functioning was assessed nearly half a million times during the study (n = 495,317) with 11.0 (median) assessments per participant. Physical functioning score was significantly better in the intervention versus comparison groups during the 8-year intervention and extended follow-up through 12 years (median) (P = 0.001), representing a reduction in age-related functional decline. The intervention effect subsequently lost significance at 19 years and both randomization groups crossed the disability threshold at similar times. Differences between randomization groups in physical functioning emerged after stratification by physical activity and age decade (P-interaction = 0.007). Among all participants physically active at entry, the intervention initially had a statistically significant, favorable influence on physical functioning which attenuated post-intervention. In contrast, among younger, physically inactive women 50-59 years of age, the intervention had a persistent, statistically significant, favorable influence on physical functioning with associated delay in crossing the disability threshold. Conclusions: In the primary prevention setting of the WHI DM randomized trial, with long-term follow-up, a dietary intervention which has been shown to reduce breast cancer mortality also significantly reduced age-related functional decline through 12 years. Among all participants, the intervention effect was attenuated with longer follow-up. However, reduction in age-related functional decline was sustained in younger women in the intervention group who were inactive at entry, a potential target population for future behavior interventions. Clinical trial information: NCT00000611.

ImportanceAbout 25% of all triple-negative breast cancers (TNBCs) and 10% to 20% of high-grade serous ovarian cancers (HGSOCs) harbor BRCA1 promoter methylation. While constitutional BRCA1 promoter methylation has been observed in normal tissues of some individuals, the potential role of normal tissue methylation as a risk factor for incident TNBC or HGSOC is unknown.ObjectiveTo assess the potential association between white blood cell BRCA1 promoter methylation and subsequent risk of incident TNBC and HGSOC.Design, Setting, and ParticipantsThis case-control study included women who were participating in the Women’s Health Initiative study who had not received a diagnosis of either breast or ovarian cancer before study entrance. A total of 637 women developing incident TNBC and 511 women developing incident HGSOC were matched with cancer-free controls (1841 and 2982, respectively) in a nested case-control design. Cancers were confirmed after central medical record review. Blood samples, which were collected at entry, were analyzed for BRCA1 promoter methylation by massive parallel sequencing. The study was performed in the Mohn Cancer Research Laboratory (Bergen, Norway) between 2019 and 2022.Main Outcomes and MeasuresAssociations between BRCA1 methylation and incident TNBC and incident HGSOC were analyzed by Cox proportional hazards regression.ResultsOf 2478 cases and controls in the TNBC group and 3493 cases and controls in the HGSOC group, respectively, 7 (0.3%) and 3 (0.1%) were American Indian or Alaska Native, 46 (1.9%) and 30 (0.9%) were Asian, 1 (0.04%) and 1 (0.03%) was Native Hawaiian or Pacific Islander, 326 (13.2%) and 125 (3.6%) were Black or African, 56 (2.3%) and 116 (3.3%) were Hispanic, 2046 (82.6%) and 3257 (93.2%) were White, and 35 (1.4%) and 35 (1.0%) were multiracial. Median (range) age at entry was 62 (50-79) years, with a median interval to diagnosis of 9 (TNBC) and 10 (HGSOC) years. Methylated BRCA1 alleles were present in 194 controls (5.5%). Methylation was associated with risk of incident TNBC (12.4% methylated; HR, 2.35; 95% CI, 1.70-3.23; P P P P = .003). Across individuals, methylation was not haplotype-specific, arguing against an underlying cis-acting factor. Within individuals, BRCA1 methylation was observed on the same allele, indicating clonal expansion from a single methylation event. There was no association found between BRCA1 methylation and germline pathogenic variant status.Conclusions and RelevanceThe results of this case-control suggest that constitutional normal tissue BRCA1 promoter methylation is significantly associated with risk of incident TNBC and HGSOC, with potential implications for prediction of these cancers. These findings warrant further research to determine if constitutional methylation of tumor suppressor genes are pancancer risk factors.

10511 Background: In the Women’s Health Initiative (WHI) Dietary Modification (DM) randomized trial, after 8.5 years dietary intervention and 19.5 years cumulative (median) follow-up, dietary intervention participation was associated with a statistically significantly 22% lower breast cancer mortality (P = 0.02). In observational studies, physical activity has been associated with lower breast cancer risk with emerging results now indicating, compared to inactivity, any increase in physical activity has health benefits. Currently, longitudinal data on whether an intervention targeting dietary change influences other health-related behaviors as a gateway effect is limited. To evaluate whether randomization to a dietary intervention was associated with self-directed change in physical activity. Methods: In the WHI DM trial, 48,835 postmenopausal women, ages 50-79 years, with no prior breast cancer and baseline normal mammogram were randomized at 40 US clinical centers to a dietary intervention (19,541) or a comparison group. Dietary goals were to reduce fat intake to 20% of energy and increase intake of vegetable, fruit, and grains addressed in 18 group sessions in year 1 then quarterly. Neither randomization group received specific or ongoing instructions to increase physical activity, but physical activity was referenced in written materials given to the intervention groups in 7 of the 56 sessions. Episodes per week of moderate or vigorous recreational physical activity (MVPA) were collected at baseline and serially through 15.9 years follow-up by self-report questionnaire. Marginal longitudinal logistic regression models were used to assess physically inactive (MVPA = 0) or physically active (MVPA > 0) participants by randomization group. Marginal Poisson regression models estimated mean weekly MVPA by randomization group. Results: 45.6% of participants reported 0 MVPA at baseline which largely persisted throughout follow-up. During cumulative follow-up, relative to the comparison group, dietary intervention group participation was associated with 7% lower physical inactivity rate (odds ratio [OR] 0.93 95% confidence interval [CI] 0.91, 0.95, P < 0.001) and a 4% higher mean MVPA (ratio of means [RM] 1.04 95% CI 1.02, 1.06, P < 0.001). The association between dietary intervention participation with higher physical activity level was stronger with increasing BMI (P-interaction 0.01) and for women with waist circumference ≥ 88 cm (P-interaction 0.02). Conclusions: In conclusion, in a randomized trial setting, a low-fat dietary pattern intervention was associated with a significantly lower physical inactivity rate and significantly higher moderate and vigorous physical activity level which could be associated with health benefits. Clinical trial information: NCT00000611 .

1569 Background: Associations between dietary protein intake and breast cancer are unclear, in part due to limitations of dietary self-report. Women’s Health Initiative (WHI) investigators compared the accuracy of food frequency questionnaire (FFQ) data on energy and protein intake with objective measures of dietary intake using biomarkers (doubly labeled water for energy and urinary nitrogen for protein [n=544]). Subsequently, regression equations incorporating participant characteristics were developed acknowledging differential reporting dietary data errors based on participant characteristics (Neuhouser Am J Epidemiol). FFQ findings were then used to determine biomarker- adjusted animal vs vegetable protein ratios. Methods: We examined associations of energy and protein intake with breast cancer incidence and mortality in Women’s Health Initiative (WHI) participants 50-79 years of age at entry between1993-1998, with breast cancers verified by medical record review and survival enhanced by serial National Death Index (NDI) searches through 2016. Associations between sources of protein intake (animal versus vegetable) quintiles and breast cancer incidence and mortality were estimated using multivariable Cox proportional hazards regression. Results: With 100,024 eligible participants, after 14 years follow-up, women with higher total protein intake had greater body mass index, were more likely White, menopausal hormone therapy users with higher total energy intake and fat intake. With 6,340 incident breast cancers, 764 deaths from breast cancer and 2,059 deaths after breast cancer, higher vegetable protein intake was associated with significantly lower breast cancer incidence (P for linear trend = 0.01) while higher animal protein intake was associated with significantly higher breast cancer incidence (P for linear trend = 0.03). Higher vegetable protein intake was also associated with significantly lower risk of death after breast cancer (P

1513 Background: Next-generation sequencing enables rapid analysis of many inherited cancer susceptibility genes. Little is known about the prevalence and penetrance of pathogenic variants (PVs) in such genes among post-menopausal women with breast cancer, who comprise the majority of all breast cancer patients. Methods: The Women’s Health Initiative enrolled post-menopausal women from 1993-1998. We conducted a nested case-control study using banked DNA samples of 2,195 women who subsequently developed invasive breast cancer (cases) and 2,322 cancer-free controls. Sequenced genes were APC, ATM, BARD1, BMPR1A, BRCA1, BRCA2, BRIP1, CDH1, CDK4, CDKN2A (p16INK4a and p14ARF) , CHEK2, EPCAM, GREM1, MLH1, MSH2, MSH6, MUTYH, NBN, PALB2, PMS2, POLD1, POLE, PTEN, RAD51C, RAD51D, SMAD4, STK11, and TP53. PV were defined using American College of Medical Genetics criteria. PV prevalence is reported as proportions and penetrance as the odds ratio (OR) and 95% confidence interval (CI) of PV versus none among cases versus controls. Results: Among cases, the median age at diagnosis was 73 years; 66% were White, 18% Black, 6% Hispanic, 6% Asian and 4% other. The prevalence of PVs in any gene was significantly higher in cases (6.61%, 95% CI 5.57-7.65%) versus controls (4.09%, 95% CI 3.29-4.90%). The prevalence of BRCA1/2 PVs was 1.2% in cases and 0.22% in controls. Among cases, the prevalence of PVs in other breast cancer-risk genes was 2.3% ( ATM, CDH1, BARD1, BRIP1, CHEK2, NBN, and PALB2 collectively), two-fold higher than PVs in BRCA1/2. Prevalence of BRCA1/2 PVs decreased with age among cases, while prevalence of ATM, CHEK2 and PALB2 PVs did not. Statistically significant ORs for breast cancer penetrance were observed for BRCA1 (5.43, 95% CI 1.19-51.52), BRCA2 (4.71, 95% CI 1.84-15.08), BARD1 (9.78, 95% CI 1.04-1295.87) and PALB2 (6.30, 95% CI 1.93-31.94). Conclusions: Approximately 7% of women diagnosed with post-menopausal breast cancer carry a PV in a cancer susceptibility gene. In contrast to studies of younger breast cancer patients, PVs in other breast cancer-related genes were two times more common than in BRCA1/2. Results may guide genetic testing of women with post-menopausal breast cancer.

1539 Background: The WHI Diet Modification (DM) trial randomized 48,835 postmenopausal women with no prior breast cancer to a low-fat dietary intervention or comparison group. After 16.1 years follow-up, the intervention was associated with an 18% reduction in risk of death after breast cancer (P =0.01), with greater reduction (29%) in those with waist circumference≥88 cm (J Clin Oncol 2017). To extend these findings, we examined the influence of the dietary intervention on breast cancer mortality in subgroups defined by number of metabolic syndrome (MS) components with 19.6 years median cumulative follow-up. Methods: WHI DM has been previously described. Four MS components were determined at entry: 1) waist circumference≥ 88 cm, 2) high blood pressure or anti-hypertensive use, 3) high cholesterol history and 4) diabetes history, with women categorized as having 0 (n=10,639), 1-2 (n=30,948), or 3-4 (n=4,246) MS components. Forest plots of hazard ratios (HRs) were generated with P-values for interaction between randomized group assignment and number of MS components. Results: Women with 3-4 MS components were more likely to be Black, be obese (BMI ≥30), and have diabetes (all P < 0.001). Breast cancers in women with 3-4 MS components were less likely to be local stage (P = 0.005) or well differentiated (P = 0.03). The magnitude of reduction in deaths from breast cancer in the dietary intervention vs comparison group increased as the number of MS components increased (interaction P = 0.01). Hazard ratios (HR) and 95% confidence intervals (CI) for death from breast cancer for intervention vs comparison groups for women with 0 MS components was 1.09 95% CI, 0.63-1.87, with risk low in both randomization groups (0.028% and 0.026%, respectively); for women with 1-2 MS components, HR 0.80 95% CI 0.62-1.02; and for women with 3-4 MS components, HR 0.31 95% CI, 0.14-0.69, with risk in the intervention group reduced to 0.026%. Conclusions: Adoption of a low-fat dietary pattern had a greater effect on reducing deaths from breast cancer in women with more MS components, suggesting that this is a high risk group more likely to benefit from the dietary intervention. Clinical trial information: NCT00000611.

70 Background: With demographic shifts leading to an aging cancer population, the American Society of Clinical Oncology (ASCO) has recommended that journal editors improve the reporting of clinical trial data specific to older adults. This study aimed to assess the current reporting status of clinical trial data regarding older adults with cancer. Methods: This was a review of oncologic therapeutic phase-3 clinical trial data published from 07/01/2016-06/30/2017. Based on a keyword search of EMBASE and PubMed, 929 manuscripts were identified. Removing duplicates (n = 116) and articles that did not meet this study’s cancer inclusion criteria (n = 589), a total of 224 were identified. In the pilot phase there was 85% agreement among the reviewers. Results: Thus far, 197 papers (88%), have been independently reviewed with 118 evaluated by more than one reviewer. Reviewed papers were published in 57 journals including Journal of Clinical Oncology (28, 14.2%), Lancet Oncology (25, 12.7%), and NEJM (20, 10.2%). Much of the literature focused upon the following cancer sites: Breast (33, 16.7%), Lung (27, 13.7%), Colorectal (18, 9.1%), and Prostate (12, 6.1%). 175 articles included inclusion/exclusion criteria, 32 (16.2%) had upper age exclusion criteria. Age was presented in 184 articles and data was stratified by age in 84 (42.6%). Age stratification of effectiveness and toxicity were presented in the results section in 65 (38.2%) and 21 (14.4%). Effectiveness by age was in the discussion section in 37 (18.8%) and toxicity by age in 16 (8.1%). Reviewers could not determine the proportion of study participants who were older adults in 111 articles (56.2%). In the remainder of the articles, the proportion ranged from 0% to 81%, except one in which all participants were older adults. Conclusions: A slim minority of phase-3 oncologic clinical trials included and discussed age-referenced results regarding the effectiveness and toxicity of treatments in older adults. Clinical investigators and journal editors should consider the ASCO recommendation and increase the reporting of oncologic clinical trial data specific to older adults.

e13567Background: Vasomotor symptoms (VMS) including hot flashes and night sweats are common during menopausal transition and may persist. While pathophysiology of VMS is complex, estrogen’s effici...

e22100Background: Cross-sectional studies suggest that falls are more prevalent among older cancer survivors than those without cancer. While nearly half of breast cancer survivors are > 70 years o...

45 Background: Only in the past decade has breast cancer survivorship earned formal recognition as a research discipline. Complicating survivorship research is the overlap between aging and treatment sequelae (Pan et al., Breast Cancer Res Treat 2016). The ACS/ASCO 2015 Breast Cancer Survivorship Care Guideline and the jointly sponsored 2016 Cancer Survivorship Symposium afforded an opportunity to review the “state of the science.” Methods: All 236 citations from the Guideline and all 250 abstracts from the Symposium were reviewed independently by two authors and prospectively categorized as follows: randomized trial; non-randomized study with controls; study without controls; review; and guideline. Additional categories were generated during the review process. Results: The Guideline most commonly cited reviews (n = 88, 37%), followed by 51 (22%) non-randomized, non-controlled studies and 37 (16%) randomized trials, which mostly addressed interventions for therapy sequelae such as lymphedema. Symposium abstracts most commonly described non-randomized, non-controlled studies (n = 113, 45%). Among 13 randomized trials (5%), 3 had not completed accrual; sample sizes were 60-467. 42 (17%) abstracts were identified as pilot studies and 17 (7%) as qualitative studies. 65 (26%) exclusively addressed breast cancer. Abstracts mostly covered health-related or psychosocial sequelae of therapy; however, some addressed overall survival or active cancer therapy. Conclusions: Much of the survivorship literature remains in the exploratory stage, consisting of non-controlled, pilot and qualitative studies. To optimally address survivorship issues, increasing incorporation of cancer-free, age-matched control populations is needed. [Table: see text]