Social factors and severity of phenotypic frailty in the Elevate longitudinal cohort of older adults with early-stage breast cancer.

e13847 Background: Phenotypic frailty is a clear risk factor for higher morbidity and mortality in older breast cancer patients. How social factors may be potentially protective against phenotypic frailty in this population remains unclear. We sought to define these associations at baseline in the Elevate cohort. Methods: The Elevate Longitudinal Cohort study prospectively enrolled patients >70 years with a new invasive non-metastatic breast cancer diagnosis between 4/19/19-9/28/23 from 7 US centers, to participate in 5 years of clinical data, biospecimen and serial survey assessments. Phenotypic frailty was calculated using a 10-item frailty index, categorizing participants as frail, prefrail, or robust. Life expectancy was calculated using the Schonberg Index. Elements of social frailty included education, income, living situation (living alone or with spouse), Medical Outcomes Study (MOS) Social Activity score, and MOS Social Support scores. Descriptive statistics were performed and multivariable modified Poisson analyses were run to estimate relative risk ratios. Results: Of 187 women with complete baseline survey data, the median age was 75.7, and 70% were robust, 22% were prefrail, and 8% were frail; 8% were non-White and 3% were Hispanic. Overall, 33% of frail patients, compared to 90% of robust patients, had a life expectancy > 10 years. Most patients were not married (53%), lived with someone else (63%), did not need help with their instrumental activities of daily living (82%), had some college education or higher (74%), and had a household income of > $40,000 (55%). On univariate analyses, living situation and marital status were not significantly associated with prefrailty or frailty. On multivariable analyses, high income, higher instrumental activities of daily living score and social activity score were associated with a lower likelihood of being prefrail or frail (Table). Conclusions: At diagnosis, specific social elements are significantly associated with a lower risk of phenotypic frailty in older women with early-stage breast cancer, suggesting a potentially protective effect. How social and phenotypic frailty may interact with respect to longitudinal outcomes in the patient population remains to be seen in this cohort. Baseline Characteristic Relative risk for prefrailty or frailty 95% CI Education At least some college vs High school or less 0.71 (0.44, 1.15) Income $40,001--$80,000 vs ≤ $40,000 0.45 (0.26, 0.76) >$80,000 vs ≤ $40,000 0.44 (0.24, 0.82) Prefer not to answer/Missing vs ≤ $40,000 0.36 (0.2, 0.64) Instrumental Activities of Daily Living (IADL) No help for any IADL items vs At least some help for at least one IADL items 0.34 (0.21, 0.54) MOS social support score* Score increase by 1 1.00 (0.99, 1.01) MOS Social Activity Score* Score increase by 1 0.99 (0.98, 1) Bold: p-value<0.05. *Range 0-100, higher score indicates better social support/activity.