Background Chemotoxicity risk scores were developed to predict grade 3-5 chemotherapy toxicity in older women with early breast cancer. However, whether these toxicity risk scores are associated with clinically meaningful decline in patient health remains unknown. Methods In a prospective study of women ≥65 with stage I-III breast cancer treated with chemotherapy, we assessed chemotoxicity risk using the Cancer and Aging Research Group-Breast Cancer (CARG-BC) score (categorized as low, intermediate, high). We measured patient health status before (T1) and after (T2) chemotherapy using a clinical frailty index (Deficit Accumulation Index [DAI; categorized as robust, prefrail, frail]). The population of interest was robust women at T1. The primary outcome was decline in health status after chemotherapy, defined as a decline in DAI from robust at T1 to pre-frail or frail at T2. Multivariable logistic regression was used to examine the association between T1 CARG-BC score and decline in health status, adjusted for sociodemographic and clinical characteristics. Results Of the 348 robust women at T1, 83 (24%) experienced declining health status after chemotherapy, of whom 63% had intermediate/high CARG-BC scores. After adjusting for sociodemographic and clinical characteristics, women with intermediate (OR = 3.14, 95% CI 1.60-6.14, p < 0.001) or high (OR = 3.80, 95% CI 1.35-10.67, p = 0.01) CARG-BC scores had greater odds of decline in health status compared to women with low scores. Conclusions In this cohort of older women with early breast cancer, higher CARG-BC scores prior to chemotherapy were associated with decline in health status after chemotherapy independent of sociodemographic and clinical risk factors.

No abstract

PURPOSE: Family caregivers play an integral role in caring for older adults with cancer. Few studies have examined older adults with cancer and their family caregivers as a unit in a relationship or a dyad. Dyad congruence, or consistency in perspective, is relevant to numerous aspects of living with cancer, including the decision to enroll in a cancer clinical trial. METHODS: Semistructured interviews of 32 older women (age ≥ 70 years) with breast cancer and their family caregivers (16 dyads) were conducted at both academic and community settings from December 2019 to March 2021 to explore perceived facilitators and barriers to cancer trials. Dyad congruence was defined as aligned (matching) perspectives, and incongruence was defined as misaligned (nonmatching) perspectives. RESULTS: Five (31%) of 16 patients were age ≥80 years, 11 (69%) had nonmetastatic breast cancer, and 14 (88%) were treated in an academic setting. Six (38%) of 16 caregivers were in the 50-59 age group, 10 (63%) were female, and seven (44%) were daughters. Dyad congruence centered on the clinical benefit of trials and physician recommendation. However, compared with caregivers, patients were more motivated to contribute to science. Patients and caregivers also differed on the perceived extent to which the caregiver influenced enrollment. CONCLUSION: Older patients with cancer and their caregivers generally agree about the facilitators and barriers to cancer trial enrollment, but some perceptions are misaligned. Further research is needed to understand whether misaligned perspectives between patients and caregivers influence clinical trial participation of older adults with cancer.

12035 Background: Hospitalizations are frequent, burdensome, and potentially avoidable complications of cancer treatment in older adults. While geriatric assessments can identify older patients at risk for hospitalization, the use of these assessments in clinic is limited by time and resources. Screening for falls in the past 6 months is a quick and simple question that clinicians can ask in clinic and fall history has been linked to chemo-related toxicity. Here, we hypothesized that in older adults with cancer fall history prior to chemo is associated with an increased risk of hospitalization during chemo. Methods: In a prospective multicenter study of 497 women age ≥65 with stage I-III breast cancer treated with neo/adjuvant chemo, we assessed baseline self-reported fall history in the past 6 months as a dichotomous (yes/no) and continuous variable (no. of falls). Our primary endpoint was hospitalization during chemo (yes/no, yes defined as incident hospitalization assessed from start to end of chemo and attributed to toxicity from chemo). We used multivariable logistic regression to examine the association between fall history and hospitalization during chemo, adjusting for sociodemographic and clinical covariates. Results: The median age was 70 (65-86), 65% had stage II/III disease, 37% had anthracycline, and 72% received primary prophylaxis with G-CSF. Of the 497 participants, 60 (12%) reported falls (median 1.0 [1-6]) at baseline, and 114 (23%) were hospitalized during chemo (median 1.0 [1-4]). The most common toxicities that resulted in hospitalization were febrile neutropenia (25%), anemia (13%), dehydration (13%), and fatigue (13%). Women who fell in the 6 months prior to chemo had greater odds (adjusted OR = 2.69, 95% CI 1.42-5.11, p= 0.003) of being hospitalized during chemo compared to women who did not fall. Conclusions: In this cohort of older women with early breast cancer, fall history prior to chemo was associated with an increased risk of hospitalization during chemo, independent of sociodemographic and clinical factors. Oncologists do not routinely assess fall history when considering chemo in older adults. Screening for falls is a quick, simple, and important indicator of vulnerability that oncologists may use to identify patients at risk for hospitalization during chemo and inform personalized, patient-partnered treatment decisions. Clinical trial information: NCT01472094 . [Table: see text]