Cohort Of Breast Cancer Survivors Research Articles

Abstract A097: Racial and ethnic disparities in risk of persistent opioid use in a diverse cohort of breast cancer survivors and matched women without cancer

Abstract Background: Persistent opioid use (POU) in breast cancer survivors after they have completed active cancer treatment is an emerging area of concern. Disparities affecting the risk of POU remains understudied in breast cancer survivors. We evaluated disparities in examining the risk with POU in breast cancers survivors in the year after active cancer treatment along with a comparison group. Methods: We assembled a cohort of 14,347 female breast cancer (BC) survivors diagnosed in 2009-2019 treated for early-stage disease (in situ, local, regional), who had at least two opioid prescriptions after one year of survivorship and 51,369 age-matched women without cancer. Data were captured from the tumor registry, electronic health records, and pharmacy databases. The primary outcome was POU defined as continuously receiving at least 90 days supplies of opioids during follow-up. Patients were followed through the date once they met POU criteria, died, disenrolled from health plan, or reached study’s end (12/31/2021), whichever came first. We used the multivariable proportional hazard regression to estimate the association [hazard ratios (HR), 95% confidence interval (CI)] between race ethnicity and POU. Results: Overall, the cohort was diverse: 7% Asian/Pacific Islander (API); 14% Black; 28% Hispanic; and 50% White women. During a median follow-up of 6.4 years, 11134 (17%) women developed POU. Patients who developed POU had longer annualized median opioid duration, especially in Black patients (175 vs. 5 days) and were prescribed stronger opioids [median maximum daily morphine milligram equivalent (62.5 vs 42.9)], compared to those who did not develop POU. In the multivariable analyses adjusted for sociodemographic characteristics, opioid type, and other pain medications, Black, Hispanic and API women were 28%, 42% and 54% less likely to develop POU compared to White women [HR (95%CI): 0.72 (0.68-0.76), 0.58 (0.55-0.61), and 0.46 (0.42-0.52), respectively]. However, Black women were only 19% less likely [HR (95%CI): 0.81(0.72-0.92)] to develop POU compared to White women in BC survivors. Also, patients with older age, Elixhauser Comorbidity Index ≥3; smoking, type of opioid (Oxycodone and hydrocodone), and who used other psychiatric and pain medications had significantly greater POU risk. Conclusion: Findings suggest that women of color were less likely to develop POU. Further studies will include further examining clinical and pharmacological management which might have played a role in the race ethnicity groups. Citation Format: Lie Hong Chen, Rulin C. Hechter, Jiaxiao Shi, Zheng Gu, Moira Brady-Rogers, Rowan T. Chlebowski, Reina Hague. Racial and ethnic disparities in risk of persistent opioid use in a diverse cohort of breast cancer survivors and matched women without cancer [abstract]. In: Proceedings of the 17th AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2024 Sep 21-24; Los Angeles, CA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2024;33(9 Suppl):Abstract nr A097.

Abstract C062: Racial and ethnic disparities in risk of second primary non-breast cancer in breast cancer survivors

Abstract Purpose: Disparities in cancer outcomes are multifactorial and are attributed to differences in access to care, socioeconomic status, tumor characteristics, treatment, family history, and comorbidity burden. Studies evaluating disparities in breast cancer outcomes have mostly been restricted to Black and White patients. Additionally, whether racial and ethnic disparities affect the risk of second primary cancer in breast cancer survivors, remains under-studied. We evaluated disparities in risk of second primary non-breast cancer in a racially and ethnically diverse cohort of breast cancer survivors treated at an integrated healthcare system. Methods: This retrospective cohort study included 37,222 breast cancer patients (aged ≥18 years) diagnosed between 2008-2020 at Kaiser Permanente Southern California (KPSC). Data on tumor characteristics, cancer treatments and sociodemographic variables were extracted from KPSC’s cancer registry and electronic health records. Patients were followed electronically from breast cancer diagnosis until they developed second primary non-breast cancer, died, disenrolled from the health plan, or reached study’s end (12/31/2021), whichever occurred first. Cox proportional hazards regression was used to report the association [Hazard Ratio (95% Confidence Interval)] between race and ethnicity and second primary non-breast cancer. Results: More than half of the study cohort consisted of non-Hispanic White survivors (51.2%), followed by 20.9% Hispanic survivors, 13.9% Asian/Pacific Islander (API) survivors, 12.7% non-Hispanic Black survivors, and 1.3% survivors of other/mixed races. Overall, 55.7% and 23.9% were diagnosed with local and regional breast cancer, respectively. During a median follow-up of 5.2 years, 4.6% survivors developed second primary non-breast cancers. In the multivariable model adjusted for sociodemographic characteristics, breast tumor characteristics, and breast cancer treatments, Hispanic and API survivors were 25% and 28% less likely to develop second primary cancer compared to White survivors [adjusted HR (95%CI): 0.75 (0.65-0.86)] and 0.72 (0.61-0.86), respectively]. We did not find an increased risk in Black survivors compared to White survivors [adjusted HR (95%CI): 0.91 (0.78-1.04)]. However, factors related to higher risk of second primary non-breast cancers were older age at breast cancer diagnosis, smoking, and higher comorbidity burden. Conclusion: Interestingly, risk of second primary non-breast cancer was lower in Hispanic and API survivors compared to White survivors in this insured cohort. Next steps will include examining which lifestyle factors (e.g. less smoking), or if pharmacological management of modifiable comorbidities might have played a role in the decreased risks seen in these two racial and ethnic groups. Citation Format: Amrita Mukherjee, Zheng Gu, Lie H. Chen, Reina Haque. Racial and ethnic disparities in risk of second primary non-breast cancer in breast cancer survivors [abstract]. In: Proceedings of the 17th AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2024 Sep 21-24; Los Angeles, CA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2024;33(9 Suppl):Abstract nr C062.

Social relationships and their impact on health-related quality of life in a long-term breast cancer survivor cohort.

Health-related quality of life (HRQOL) has become increasingly important for breast cancer survivors, but clinically relevant declines often persist for many years after treatment. This study aimed to investigate whether social relationships can mitigate or prevent this decline in HRQOL. Data were used from the German population-based Mamma Carcinoma Risk Factor Investigation (MARIE) cohort of 2022 breast cancer cases with follow-up information for more than 15 years after diagnosis. Correlations between social integration, social support, and global health status (GHS) as an overall measure of HRQOL were analyzed, and linear regression analysis was performed with structural equation modeling. The majority of participants reported high levels of social integration and social support and moderate levels of GHS. Social integration 5 years after diagnosis was associated with GHS 5 years after diagnosis (β=1.12; 95% CI,0.25-1.99), but no longitudinal effects were found. Social support 5 years after diagnosis was associated with better GHS 5 years (β=0.42; 95% CI,0.36-0.48) and 10 years after diagnosis (β=0.12; 95% CI,0.02-0.22), whereas social support 10 years after diagnosis was associated with GHS 10 years (β=0.29; 95% CI,0.20-0.39) and 15 years after diagnosis (β=0.10; 95% CI,0.01-0.21). These results confirm that social relationships positively influence HRQOL in long-term breast cancer survivors and that their association should receive more attention clinically and beyond routine care.

The Impact of Depression on Adherence to Diabetes Self-Management Behaviors in Breast Cancer Survivors.

Breast cancer survivors (BCS) have higher rates of depression which is associated with lower adherence to medications, diet, and physical activity. Managing diabetes (DM) requires adherence to several of these self-management behaviors (SMB), and BCS have an increased risk of DM. We investigated whether depressive symptoms were associated with adherence to DM SMB in a cohort of BCS. BCS with DM were surveyed semiannually for 2years. Depression was assessed with the Hospital Anxiety and Depression Scale (HADS). Adherence to DM medication, diet, and physical activity was self-reported using the Medication Adherence Report Scale (MARS), Summary of Diabetes Self-Care Activities Assessment (SDSCA), and International Physical Activity Questionnaire (IPAQ), respectively. Using generalized linear equation modeling, the association of depressive symptoms with nonadherence to SMB was assessed, adjusting for age, race, marital status, education level, and beliefs about cancer and DM risk. Among 244 BCS with DM, those who were nonadherent to medication, diet, and/or physical activity had higher depression scores (p<0.01). In adjusted analyses, higher depression scores were independently associated with dietary (OR=1.16, p<0.001) and physical activity nonadherence (OR=1.18, p<0.001) but not with medication nonadherence. Concerns about medications was independently associated with medication nonadherence (OR=1.17, p=0.024). Higher depression scores are associated with nonadherence to DM SMB in this cohort of BCS. These findings highlight the importance of addressing depressive symptoms in BCS to help improve adherence to DM medications, diet, and physical activity.

A Multiomics, Molecular Atlas of Breast Cancer Survivors.

Breast cancer imposes a significant burden globally. While the survival rate is steadily improving, much remains to be elucidated. This observational, single time point, multiomic study utilizing genomics, proteomics, targeted and untargeted metabolomics, and metagenomics in a breast cancer survivor (BCS) and age-matched healthy control cohort (N = 100) provides deep molecular phenotyping of breast cancer survivors. In this study, the BCS cohort had significantly higher polygenic risk scores for breast cancer than the control group. Carnitine and hexanoyl carnitine were significantly different. Several bile acid and fatty acid metabolites were significantly dissimilar, most notably the Omega-3 Index (O3I) (significantly lower in BCS). Proteomic and metagenomic analyses identified group and pathway differences, which warrant further investigation. The database built from this study contributes a wealth of data on breast cancer survivorship where there has been a paucity, affording the ability to identify patterns and novel insights that can drive new hypotheses and inform future research. Expansion of this database in the treatment-naïve, newly diagnosed, controlling for treatment confounders, and through the disease progression, can be leveraged to profile and contextualize breast cancer and breast cancer survivorship, potentially leading to the development of new strategies to combat this disease and improve the quality of life for its victims.

Defining musical toxicity after breast cancer treatment: A classification tree analysis.

12048 Background: Playing music has been an important part of human culture for millennia. Today more than half of American households have at least one person who plays music, and in the UK 43% of adults play an instrument. Despite the overlap between musicians and people diagnosed with cancer, the effect of cancer treatment on musicianship has not been well-studied. We previously reported that 26% of a musician breast cancer survivor cohort reported difficulty with musicianship after treatment, and here we conduct further analysis to identify treatment and disease factors associated with this difficulty, termed acute musical toxicity (AMT). Methods: The Musical Toxicity Questionnaire was distributed to participants who had previously enrolled in the Mayo Clinic Breast Cancer Registry. PROMIS (Patient-Reported Outcomes Measurement Information System) scores were available through the Registry and treatment details were collected retrospectively. Due to the inter-dependency and high correlation between treatment and disease-related variables, a classification tree analysis (CTA) was performed to identify the combination of variables that most accurately classified patients by AMT. Sixteen variables were analyzed by the algorithm. Logistic regression was utilized for examining associations between AMT and continuous variables. Results: Of the 4075 surveys distributed, 1871 were returned and 535 respondents identified as musicians. Median time from diagnosis was 5.2 years, respondents were mostly stage I or II (71%), and 32% were node positive (N+). Over a quarter (26% or 144 respondents) reported AMT. In the final CTA model, being N+ was strongly associated with AMT (42% of those N+ had AMT, 20% of those not N+ had AMT). For respondents who were N+, those who received endocrine therapy were more likely to have AMT than those who did not (46% vs 21% respectively). Subsequent leaves identified not undergoing axillary lymph node dissection (ALND) and receiving chemotherapy as associated with AMT. For respondents who did not receive chemotherapy, breast reconstruction was associated with AMT (60% vs 19%). Variables such as mastectomy, radiation volume, specific chemotherapies, and stage of disease were not selected by the algorithm. The odds of having AMT decreased by 9.7% and 17% with each additional point in the composite mental health (p=0.003) and physical health (p&lt;0.001) PROMIS scores, respectively. The odds of having AMT increased by 3.5% with each lymph node removed (p=0.001). Conclusions: The final model suggests that the group with the largest proportion of AMT was N+, received endocrine therapy, did not have ALND, and received chemotherapy. Both physical and mental health PROMIS scores were associated with AMT, although the directionality of this relationship requires more study. Care teams should counsel patients on potential musical toxicity and engage in shared decision making with musical patients.

Abstract PO5-12-03: Cardiovascular Toxicities in Breast Cancer Survivors

Abstract Background: Chemotherapy, despite its effectiveness, presents drawbacks in breast cancer patients due to its side effects. Fatigue, hair loss, nausea, and anemia are among the many symptoms associated with chemotherapy. Moreover, chemotherapy can lead to cardiovascular diseases, including cardiac arrhythmias, hypertension, heart failure, heart attacks, and cardiomyopathy, particularly with the use of anthracyclines and HER-2-targeted agents. Risk factors such as weight, age, race, comorbidities (e.g., obesity, diabetes, hyperlipidemia), and radiation exposure further increase these risks. Despite these potential complications, guidelines are just emerging aiming to prevent these cardiovascular complications in breast cancer patients. Untreated cardiovascular abnormalities can have long-term effects on breast cancer survivors. Our aim is to analyze the relationships between breast cancer treatment and cardiovascular abnormalities, examining associations with various factors, including race, BMI, and age in a majority Hispanic cohort of breast cancer survivors. Methods: retrospective, single institution study data from a total of 53 patients treated with chemotherapy with either anthracyclines and/or HER 2 directed therapy with curative intent between 2018-2020. Our analysis included basic demographic information, menopausal status, breast cancer type, treatment types, use of cardio-protective medications, and cardiovascular effects (both pre-existing and developed during treatment). Results: Our sample of 50 women primarily comprised Hispanic individuals (73%), most of whom were postmenopausal (86%), with an average age of 63 years (range 35-86). Among the breast cancer types identified, 35 were hormone-positive, 8 were HER-2 positive, and 10 were triple negative. 72% of the patients received adjuvant hormonal therapy. 18% of the population received anthracycline-based treatment, 8% received Her 2 directed therapy and 2 % received Immune check point inhibitors. Radiation therapy was administered to 50% of the patients. The most prevalent cardiovascular disease observed was hypertension, affecting 62% of patients, followed by type 2 diabetes (38%), hyperlipidemia (34%), stroke (8%), and heart attack (6%). The predominant cardioprotective medications used were ACE inhibitors/ARBs (48%), followed by statins (46%), beta-blockers (20%), aspirin (16%), and GLP-1 agonists (14%). Conclusion: Our analysis revealed a high prevalence of cardiovascular conditions among breast cancer survivors treated with chemotherapy in a majority Hispanic cohort of patients. This population has been linked to higher risk of developing comorbidities such as diabetes, obesity and hypertension. Further analysis evaluating risk factors and implementation of treatment strategies to minimize adverse effects is needed in this population. Citation Format: Victoria Ayodele, Emily Sherry, Marcela Mazo-Canola. Cardiovascular Toxicities in Breast Cancer Survivors [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO5-12-03.

Abstract PO2-12-03: Home-based Resistance Exercise Improves Fatigue and Functional Strength in Female Breast Cancer Survivors Post Chemotherapy

Abstract Introduction: Chemotherapy is associated with an increase in fatigue and loss of strength in female breast cancer (BCa) survivors. Strength training has been found to reverse these effects, however, the COVID pandemic added a complexity to clinic-based exercise intervention. The purpose of this study is to determine if a home-based resistance exercise program can attenuate these debilitating effects in a cohort of breast cancer survivors that have recently completed chemotherapy. Methods: Ten BCa survivors who completed chemotherapy within the previous 6 months were recruited to participate in a 12-week home-based resistance exercise proof of concept study. Exercise was performed 3 times per week, supervised via Zoom. Participants completed 2-3 sets of upper body and lower body exercises to volitional fatigue. Self-reported fatigue was measured using the respective subscale from the EORTC QLQ-C30. Strength was measured using hand grip dynamometry and isometric leg strength. Physical function was measured using the 6-minute walk test. Assessments were completed at baseline and end of study. As a comparison, age matched controls (AMCs) were recruited to complete cross-sectional analysis of each of these variables. A one-way analysis variance was done with Tukey’s post hoc test. Significance was set at p&amp;lt; 0.05. Results: BCa survivors were 46.4 ± 12.5 years old with a BMI of 28.11 ± 5.25. AMCs were 50.3 ±13.3 years old with a BMI of 30.8 ± 5.8. There was no significant difference in patient characteristics. A significant difference in self-reported fatigue was found between groups (F=18.52; p&amp;lt; 0.0001). Post hoc analysis found significant difference in fatigue between BCa survivors at baseline (48 ± 23) and the age matched controls (p&amp;lt; 0.0001). Exercise was able to significantly reduce fatigue in BCa survivors (21 ± 15; p=0.003) to levels not significantly different to AMCs (3.3 ± 7.5). A significant difference was also observed between groups for right-hand grip strength (F= 4.923; p=0.015) and left-hand grip strength (F=8.685, p=0.001). Post hoc analysis found a significant increase in right hand grip strength in the BCa survivor group following the intervention (p=0.012). Left hand grip strength was found to be significantly lower at baseline than AMCs (p=0.048). Left hand grip strength was significantly increased in BCa survivors at end of study (p&amp;lt; 0.001). Finally, a significant difference was observed between groups for isometric leg strength (F= 3.699; p=0.038). Post hoc analysis found a significant difference in leg strength in the BCa survivor group following intervention (pre: 105.9 ± 50.6 N-M; post: 158.4 ± 13.2 N-M; p=0.030) which was 16.5% higher than AMCs (127.4 ± 36.5 N-M). No significant difference was found in 6-minute walk distance. However, there was a 21% increase in the distance walked for BCa survivors after interventions (pre: 441.2 ± 66.5m; post: 501.4 ± 123m), identical to distances walked by AMCs (501.4 ± 90.8m). Conclusion: The results of our study suggest that a virtually supervised resistance exercise program can attenuate the negative effects of chemotherapy in BCa survivors to levels similar to their peers that never had breast cancer. Further research using powered sample sizes is needed to validate these results. Citation Format: Taylor Moncrief, Crisann Moon, Angela Gonzalez, Allison Stepanenko, Tiffany Cortes, Darpan Patel. Home-based Resistance Exercise Improves Fatigue and Functional Strength in Female Breast Cancer Survivors Post Chemotherapy [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO2-12-03.

Abstract ED03-03: Impact of exercise on breast cancer outcomes

Abstract Background: Exercise may improve breast cancer outcomes; however, current research is limited in its ability to inform clinical exercise trials and clinical oncology practice. The purpose of this presentation is to summarize the current research on exercise and breast cancer outcomes using a framework proposed to improve the clinical utility of this research. The Exercise as Cancer Treatment (EXACT) Framework organizes and characterizes the critical aspects of a clinical oncology setting to allow for a more systematic and clinically informative approach to the study of exercise as a cancer treatment. The EXACT Framework proposes that two key clinical oncology variables—tumor/disease status (i.e., primary tumor present, primary tumor removed, metastatic disease present) and treatment status (i.e., treatment naïve, actively being treated, previously treated)—are critical for understanding the effects of exercise on cancer outcomes, but few studies have addressed these variables. Application of the EXACT Framework to existing studies may improve their interpretation and identify directions for future research. Results: Presently, there are no adequately powered randomized controlled trials that have examined the effects of exercise on breast cancer outcomes in any clinical oncology scenario, although secondary analysis of smaller phase II trials have been reported. Preclinical research using rodent models has generally shown that exercise slows the growth and spread of treatment naïve breast tumors through various mechanisms including immune response, tumor metabolism, apoptosis, and DNA synthesis and repair. More recently, these preclinical studies have enhanced their clinical utility by including an existing cancer treatment in their design and have generally shown that exercise enhances the delivery and/or efficacy of concurrent cancer treatments through mechanisms related to improved tumor vasculature and perfusion. Most human evidence on exercise and breast cancer outcomes has been gleaned from secondary analysis of observational studies involving healthy cohorts or breast cancer survivor cohorts. These studies have generally included early-stage breast cancer patients who received heterogeneous cancer treatments. Physical activity data was typically collected either well before diagnosis and/or well after treatments. Overall, these studies have generally shown that higher prediagnosis and postdiagnosis physical activity are associated with a lower risk of breast cancer recurrence and mortality. Discussion: There is a disconnect between the clinical oncology scenarios addressed by animal and human studies of exercise and breast cancer outcomes. Moreover, most studies do not adequately address treatment combinations and sequencing. Future observational studies of exercise and breast cancer outcomes may improve their clinical utility by: (a) recruiting breast cancer patients with the same tumor/disease status receiving the same or similar first-line treatment protocols, (b) collecting detailed data on all planned and unplanned cancer treatments, (c) assessing exercise in relation to those cancer treatments (i.e., before, during, between, after) rather than in relation to the cancer diagnosis (i.e., various time periods before and after diagnosis), (d) collecting data on cancer-specific outcomes (e.g., disease response, progression, recurrence) in addition to mortality, and (e) conducting subgroup analyses based on cancer treatments received. Conclusions: Preclinical and observational studies may contribute important knowledge regarding the role of exercise as a cancer treatment; however, modifications to study design and analysis are necessary if they are to inform clinical research and practice. The ultimate goal of precision exercise oncology is to provide the right exercise prescription to the right breast cancer patient at the right time postdiagnosis. Citation Format: K. Courneya. Impact of exercise on breast cancer outcomes [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr ED03-03.

Strategies to increase survey participation: A randomized controlled study in a population of breast cancer survivors

BackgroundData collection by mailing questionnaires to the study population is one of the main research methods in epidemiologic studies. As participation rates are decreasing, easy-to-implement and cost-effective strategies to increase survey participation are needed. In this study, we tested the effect of a pragmatic combination of evidence-based interventions. MethodsWe conducted a two-armed randomized controlled trial, nested in a cohort of breast cancer survivors (n = 1000) in the setting of a health outcomes survey. The intervention arm received a postal pre-notification, a non-monetary incentive (ballpoint with the study logo) and an alternative invitation letter in which several lay-out and textual adjustments were implemented according to behavioural science techniques. The alternative invitation letter also contained a QR-code through which an information video about the study could be accessed. The control arm was invited according to standard practice. Participants had the option to fill-out a questionnaire either on paper or online. A questionnaire with more than 50% of the questions answered classified as participation. ResultsOverall participation rate was 62.9%. No significant difference in participation rate was observed between intervention and control arm (64.5% vs 61.3%, Risk Ratio (RR) 1.05, 95% CI [0.96 – 1.16]). Older age at study (>65 vs <51 years), and high socio-economic status (highest vs lowest quartile) were associated with higher participation rates (RR 1.30, 95% CI [1.07 – 1.57] and 1.24, 95% CI [1.09 – 1.42] respectively). In-situ carcinoma compared to invasive cancer and longer interval since treatment were associated with lower participation (RR 0.86, 95% CI [0.74 – 0.99] and RR 0.92, 95% CI [0.87 – 0.99] per 5 year increase, respectively). ConclusionOverall, the combination of four interventions tested in this study did not improve survey participation among breast cancer survivors. The overall participation rate was relatively high, possibly due to the study population of cancer survivors.

Abstract 4835: Demographic factors and risk of persistent opioid use in a diverse cohort of breast cancer survivors

Abstract Background: Persistent opioid use (POU), in breast cancer survivors after they have completed active cancer treatment is an emerging area of concern. We examined rates and identified risk with POU in breast cancers survivors in the year after active cancer treatment going forward, and if these risks differed by race/ethnicity. Methods: We assembled a cohort of female breast cancer survivors diagnosed in 2009-2019 treated for early-stage disease (in situ, local, regional), and had at least two opioid prescriptions after one year of survivorship in a large health plan in southern California. Data were captured from the tumor registry, electronic health records, and pharmacy databases. The primary outcome was POU defined as continuously receiving at least 90 days supplied of opioids during follow-up. Patients were followed through the date once they met POU criteria, died, disenrolled from health plan, or reached study’s end (12/31/2021), whichever came first. We calculated rates as the number of patients who experienced POU per 1,000 person-years (PY). We identified the predictors and estimated hazard ratios (HR) associated with POU using multivariable proportional hazard regression and stratified by race/ethnicity. Results: The cohort included 14,347 breast cancer survivors who used opioids followed for a maximum of 12 years (median: 6.6 years). Overall, the cohort was diverse: 9% Asian/Pacific Islander (API); 14% Black; 22% Hispanic; and 54% White women. Out of 14,347 patients who used opioids, 2,285 (16%) patients developed POU. POU rates was the highest in Whites (36.2 per 1,000 PY) compared other race/ethnicity groups including Blacks, Hispanics, APIs (30.7, 17.3, 10.6 per 1,000 PY, respectively). Patients who developed POU had longer annualized median opioid duration, especially in Black patients (175.0 vs. 5.3 days) and were prescribed stronger opioids (median maximum daily morphine milligram equivalent 83.7 vs 50.0), compared to those who did not develop POU. In adjusted analyses of POU risk, Black (HR, 0.81; 95% CI: 0.72-0.92), API (HR, 0.59; 95% CI, 0.52-0.66), and Hispanic women (HR, 0.59; 95% CI, 0.52-0.66) were less likely to develop POU compared to White women. Patients with older age, Elixhauser Comorbidity Index ≥3, smoking, breast cancer stage (regional); type of opioid (Oxycodone); and who used other psychiatric and pain medications had significantly greater POU risk. In stratified analysis, the POU risk was 2.6 times greater (HR, 2.61; 95% CI, 1.32-5.15) in Hispanic patients aged≥65 vs &amp;lt;40 years. The highest HR was in APIs who were current smokers (HR, 4.26; 95% CI, 1.77-10.29) compared with non-smokers. Conclusion: The findings suggest that patients who are older, current smokers, with more comorbidities, and diagnosed with higher cancer stage were most vulnerable to developing POU. Citation Format: Lie Hong Chen, Rulin C. Hechter, Jiaxiao Shi, Zheng Gu, Moira Brady-Rogers, Rowan T. Chlebowski, Reina Haque. Demographic factors and risk of persistent opioid use in a diverse cohort of breast cancer survivors [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 4835.

Abstract 2227: Physical activity at breast cancer diagnosis and incident cardiovascular disease in the Pathways Study

Abstract Background: Breast cancer survivors are at increased risk of cardiovascular disease (CVD) compared to women without a history of breast cancer. Although cardiotoxic cancer treatments influence these risks, lifestyles that are associated with both breast cancer and CVD also contribute. With the number of breast cancer survivors expected to increase, a growing need exists to identify modifiable risk factors that protect against the development of CVD after diagnosis. Physical activity at breast cancer diagnosis is associated with lower cancer recurrence and mortality risk. However, associations between physical activity at diagnosis and incident CVD have yet to be investigated in detail. Methods: The Pathways Study is a prospective cohort study of 4,504 women who were members of Kaiser Permanente Northern California (KPNC) when diagnosed with first primary invasive breast cancer between 2005 and 2013. At enrollment, women self-reported their frequency, duration, and intensity of physical activities during the previous six months using the Arizona Activity Frequency Questionnaire. This information was used to dichotomize the women’s physical activity levels by the CDC’s Physical Activity Guidelines for Americans (≥ 150 minutes of moderate-to-vigorous physical activity per week vs less). CVD events were obtained from electronic health records as part of the Pathways Heart Study and were defined as post-diagnostic incidence of ischemic heart disease, heart failure, cardiomyopathy, and stroke. Deaths were documented through several sources, including the National Death Index. Women were followed until the first of disenrollment from the KPNC health plan, death, or December 31, 2021. Competing risk Cox regression models were used to estimate associations between meeting CDC physical activity guidelines and CVD. Results: In this cohort of breast cancer survivors, women who self-reported meeting physical activity guidelines at diagnosis experienced lower rates of post-diagnostic CVD events (8.6% vs 13.1%). After adjustment for socio-demographic factors, meeting physical activity guidelines remained associated with lower risk of CVD (HR: 0.78, 95% CI: 0.61, 0.98); this association appeared to be driven by reduced risk of heart failure (HR: 0.68, 95% CI: 0.50, 0.93) and cardiomyopathy (HR: 0.51, 95% CI: 0.28, 0.92). Approximately 15% of the CVD risk reduction in Pathways Study participants can be attributed to meeting physical activity guidelines (population attributable risk percent: 14.5%, 95% CI: 8%, 21%). Conclusions: Compared to women who did not meet CDC physical activity guidelines, women who reported meeting guidelines near the time of their breast cancer diagnosis had decreased risk of future CVD, including heart failure and cardiomyopathy. Future studies investigating changes in post-diagnostic physical activity patterns and CVD risk are warranted. Citation Format: Jacob K. Kresovich, Alicia R. Richards, Isaac J. Ergas, Cecile A. Laurent, Rikki Cannioto, Jamal S. Rana, Janise M. Roh, Catherine Thomsen, Salma Shariff-Marco, Eileen Rillamas-Sun, Richard K. Cheng, Song Yao, Christine B. Ambrosone, Lawrence H. Kushi, Heather Greenlee, Marilyn L. Kwan. Physical activity at breast cancer diagnosis and incident cardiovascular disease in the Pathways Study [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 2227.

Even the presence of minimal changes in regional wall motion abnormalities with exercise stress echocardiography predicts MACE and all-cause mortality in patients with previous chest radiotherapy

Abstract Background Exercise stress echocardiography (ESE) is a non-invasive, radiation-free, and widely available technique that has demonstrated a reliable accuracy for the detection of coronary artery disease (CAD). Chest radiotherapy (RT) is associated with accelerated age-related atherosclerosis and fibrosis in a dose-dependent manner. ESE plays a role as a screening tool of RT induced heart disease, however, the prognostic value of changes in regional wall motion abnormalities in this population during ESE has not been evaluated. Methods This is a retrospective cohort study of breast cancer survivors treated with adjuvant RT that underwent ESE from 2000-2020 after treatment completion. Outcomes evaluated were: major adverse cardiovascular events (MACE: cardiovascular death, myocardial infarction, revascularization and stroke); and all-cause mortality. A univariate analysis was performed to characterize the population. Continuous variables were presented as means ± standard deviation (SD) or median with quartile 1 and 3 (Q1, Q3), and categorical variables as numbers and percentages. A Cox regression model adjusted by age and cardiovascular risk factors was used to evaluate variables associated with outcomes, and a time to first event approach was used. A p-value of &amp;lt;0.05 was considered statistically significant. Results From 2000-2020, 507 patients, with a mean age 66 ± 10 years, underwent ESE at a median of 5.8 years after RT completion. The most prevalent comorbidities were dyslipidemia (59%), HTN (55%) and obesity (37%). 55% underwent RT for left-sided breast cancer and 45% for right-sided, and median RT dose was 6000 cGy. Baseline characteristics are presented in Table 1. Most patients underwent Bruce Protocol (92%). Patients exercised a mean time of 6.6 minutes achieving 7.3 METS and 109% of functional aerobic capacity (Table 2). At a median follow up of 7.8 years, 61 MACE events occurred. Predictors of MACE were double product, baseline and stress EF, baseline and peak regional wall motion abnormalities (RWMA), development of new ischemic RWMA, inability to reach 85% of maximum HR, and an abnormal EF response with stress after adjusting for age, HTN, DM, smoking history, dyslipidemia, obesity, prior MI and cancer staging (Fig 1). Predictors of all-cause mortality were METS, LV mass index, baseline and peak RWMA, development of new ischemic RWMA and an abnormal EF change with stress even after adjusting for variables described previously (Fig 2). Conclusions The development of new or worsening RWMA of just 2 segments with stress, was a significant predictor of MACE and all-cause mortality in a large cohort of breast cancer survivors undergoing multidisciplinary management including adjuvant RT. Special emphasis on management of cardiovascular comorbidities and closer surveillance are recommended for these patients.

Impact of workplace accommodations on long-term work outcomes in a diverse sample of breast cancer survivors.

501 Background: Low-income and racial/ethnic minority workers in the US have limited access to work accommodations, such as schedule flexibility, if they become ill. Access to such accommodations is a key predictor of work status immediately after treatment for breast cancer. We evaluated the impact of accommodations on long-term work outcomes in a demographically diverse cohort of breast cancer survivors. Methods: We surveyed pre-diagnosis employed women, aged 18-64, who spoke Chinese, English, Korean, or Spanish and were undergoing treatment for stage I-III breast cancer in New York City. Recruitment was enriched for racial/ethnic minority and immigrant status. Surveys were administered (telephone or online) during treatment (baseline) and 4 months after completion of primary therapy. Long-term work status was evaluated in brief surveys 1 and 2 years later. The primary outcome was post-treatment work status (working full- or part-time vs. any other work status). We used X2 and Fisher's Exact Tests to examine associations between participant characteristics and work status, and logistic regression to adjust for relevant characteristics in multivariable analyses. Results: From 06/2012 to 08/2018, 497 participants completed baseline surveys; 479 completed 4-month post-treatment surveys, and 401 and 387 reported on their work status 1 and 2 years later. Overall, 22% self-identified as Black, 22% Chinese, 7.1% Korean, 27% Latina, 19% non-Latina white, and 2.7% as other; 58% were born outside of the US. Post-treatment work status remained relatively stable over time: 29% were not working 4 months after treatment completion, and 26% and 25% were not working 1 and 2 years later. Non-Latina white women were most likely to be working after treatment completion (89%), followed by Korean (74%), Black (73%), Chinese (69%), and Latina women (61%; overall p&lt;0.001). Similar trends were observed 1 and 2 years later. Participants who reported that their employer had been accommodating in their 4-month post-treatment survey were more likely to be working 4 months after completion of treatment (83% vs. 57% of those with unaccommodating employers, p&lt;0.001) and 1 year (85% vs. 60%, p&lt;0.001) and 2 years later (84% vs. 65%, p&lt;0.001). The relationship between employer accommodations and post-treatment work status across all time points remained significant after controlling for age, race/ethnicity, language, household income as a % of the Federal Poverty Level, job type, and receipt of chemotherapy. Conclusions: In this diverse cohort of breast cancer survivors, relative to work status 4 months post-treatment, work status was relatively stable 1 and 2 years later. Women whose employers had been accommodating during treatment were more likely to be working 1 and 2 years later. Increasing access to accommodations could improve long-term work outcomes and abrogate disparities in patients who undergo cancer therapy.

Cardiometabolic Profile, Physical Activity, and Quality of Life in Breast Cancer Survivors after Different Physical Exercise Protocols: A 34-Month Follow-Up Study.

Breast cancer (BC) and cardiovascular (CV) disease share many risk factors associated with worse outcomes, in terms of cancer relapse, CV events, and quality of life (QoL), that could be counteracted by physical exercise (PE). We aimed to assess the impact of a 12-week differential PE protocol on cardiometabolic profile, QoL, CV- and BC-related long-term outcomes, and physical activity (PA) in a cohort of BC survivors (BCS) not treated with chemotherapy. 57 BCS participated in a 12-week PE protocol [aerobic exercise training (AET) or resistance exercise training (RET)]. Anthropometric and CV evaluation, health-related (HR)-QoL, daily PA, cortisol, and dehydroepiandrosterone sulfate (DHEA-S) levels were assessed before (T0) and after (T1) PE. We assessed BC and CV outcomes, HR-QoL, CV-QoL, and PA at the follow-up. RET improved waist circumference, DHEA-S, cortisol/DHEA-S, systolic and mean blood pressure, and ventricular/arterial coupling; AET ameliorated sagittal abdomen diameter and pulse wave velocity. Regarding HR-QoL, physical function improved only in AET group. At a mean 34 ± 3.6-month follow-up, we documented no significant differences in CV-QoL, HR-QoL, and PA or CV and BC outcomes. AET and RET determine specific, positive adaptations on many parameters strongly related to CV risk, CV and BC outcomes, and QoL, and should be included in any cardio-oncology rehabilitation program.

The impact of COVID-19 on anxiety among breast cancer survivors before, during and after the pandemic.

e24178 Background: The COVID-19 pandemic has caused unprecedented stressors on the mental health of patients with cancer as well as with cancer survivors. Little is known regarding anxiety among breast cancer survivors during and following the pandemic. Methods: We performed a cross-sectional study evaluating the mental health impacts of the COVID-19 pandemic on a cohort of breast cancer survivors (n = 1333). Between January and March 2022, participants enrolled in the Rays of Hope Breast Cancer Research Registry were invited to complete an online study questionnaire. We analyzed the data collected from a series of questions regarding outcomes and perspectives representing three distinct time periods: before (prior to March 2020), during (March 2020-June 2021) and after the COVID-19 pandemic (January-March 2022), based on the dates of the Massachusetts State of Emergency declaration. The COVID-19 Anxiety Scale (CAS) was used to measure anxiety specific to COVID-19. Generalized anxiety symptoms were measured using the Generalized Anxiety Disorder 7-item (GAD-7) scale. Multinomial logistic regression was used to identify factors related to increased or decreased anxiety symptoms. Results: A total of 272 independent participants (21.5%) completed the survey. Of these, 230 respondents with a confirmed history of breast cancer were included in the final analysis. While 26.2% met GAD-7 criteria for moderate to severe anxiety after the pandemic, only 13% self-reported a prior diagnosis of any mental health condition. Approximately 17% reported experiencing more anxiety symptoms after the pandemic compared to before. A history of smoking was associated with a higher likelihood of experiencing increased symptoms of generalized anxiety (OR = 1.97, 95% CI 0.93-4.14); whereas exercise frequency of at least three times per week was associated with experiencing a decrease in anxiety symptoms over this period (OR = 7.00, 95% CI 1.39-35.14). The CAS and reporting that the pandemic had a negative impact on cancer care were associated with a higher likelihood of increased generalized anxiety after the pandemic (OR = 2.21, 95% CI 1.19-4.09; OR = 3.05, 95% CI 1.06-8.83, respectively). Conclusions: Increased symptoms of anxiety were prevalent among breast cancer survivors during the pandemic and, in some, have not returned to pre-pandemic levels. Less than half of participants with symptoms meeting criteria for moderate to severe anxiety reported having a formal medical diagnosis, suggesting that clinically significant anxiety may be underrecognized in breast cancer survivors following the pandemic.

Fear of cancer recurrence inventory scores and their correlation with quality of life and stress levels in breast cancer survivors.

e24073 Background: For survivors of early stage breast cancer, the fear of cancer recurrence (FCR) can be distressing. A clinical challenge is recognizing which patients reporting FCR may need psychological intervention. Prior studies have suggested clinically significant FCR if the validated FCR Inventory (FCRI) scale score is &gt; = 22 while others suggest as low as &gt; = 13 may be an optimal threshold. Our goal was to understand the optimal thresholds for detection of clinically significant impairments in quality of life and high levels of stress. We thus examined FCRI thresholds that optimized detection with two standardized screening modalities of quality of life and perceived stress respectively: the Functional Assessment of Cancer Therapy-General (FACT-G) and the Perceived Stress Scale (PSS). Methods: A previously published cohort of breast cancer survivors at our institution consented to complete self-reported psychological measures and gave access to their medical records. The 2 measures being compared to the FCRI were PSS and FACT-G. FCRI, is a 9-item self-report measure, that evaluates worry about health status and illness returning, triggers that influence worry, uncertainty, and the concerns of significant others. FACT-G, a 28-item self-report measure, assesses quality of life in cancer patients, including physical well-being, social/family well-being, emotional well-being, and functional well-being; we examined the binary threshold score for reduced health related quality of life ( &lt; = 62) vs not reduced health related quality of life (&gt; 62). PSS, assesses the degree to which situations in one’s life are appraised as stressful; we examined the threshold score for low stress (0-13) vs moderate or high stress ( &gt; 13). Sensitivity (Sen) and specificity (Spe) were calculated for each cutoff value for the FCRI score between 16 and 22 (two proposed cutoffs) when compared to the FACT-G and the PSS binary cut points. Results: A total of 177 breast cancer survivors participated in the study. The mean age of participants was 57.6 (SD 13.1) years. Mean scores were 18.6 (SD 7.5) for the FCRI (n = 170), 13.92 (SD 7.0) for the PSS (n = 176) and 80.6 (SD 14.9) for the FACT-G (n = 155). FCRI &gt; = 20 appeared to optimize sensitivity and specificity both on the FACT-G (Sen: 82.4, Spe: 56.1) and PSS (Sen: 63.2 Spe: 69.1) in our patient population. Conclusions: Our results suggest that a FCRI cutoff of 20 may optimize inclusion of patients likely to have poorer quality of life and higher levels of stress. This may allow for more patients in this vulnerable population to be identified for possible intervention, without losing specificity to those with significant distress.[Table: see text]

Readability and understandability of instruments for collecting patient-generated data (PGD).

e18534 Background: PGD are crucial to inform and optimize cancer care. These may include patient-reported outcomes measures (PROMs), patient-reported experience measures (PREMS), or patient preferences information (PPI). The increasing use of PGD has led to the need for the adaptation of instruments to different languages and cultures. Recent data has brought attention to the additional need to consider patients’ ability to understand written information included in PROMs. Low health literacy is associated with higher risk for quality-of-life deterioration and generates disparities, including underrepresentation in cancer research. Aiming at improving equity and representation, we evaluated readability and understandability of nine French-language PROMs that are currently used in a contemporary cohort of breast cancer survivors. Methods: Readability of PROMs was assessed using the Flesh Reading Ease Score (FRES), the Gunning’s Fog Index (FOG), and the FRY graphics (yielding a mean score representing the n. of school years required to understand the respective text). Readability was considered ideal if mean score ≤6th grade level and acceptable if between 6-8th grade. Understandability was evaluated using the Patient Education Materials Assessment Tool (PEMAT, range 0-100%, higher scores corresponding to easier understandability) and defined as ideal if PEMAT ≥ 80%. Plain-language best practice was met if both readability and understandability were ideal. The Evaluative Linguistic Framework for Questionnaires (ELF-Q) provided additional qualitative elements to assess understandability. Two native French speakers, including a patient representative, independently evaluated each PROM. An additional independent assessment was requested if discordance. Results: None of the 9 PROMs evaluated had ideal readability scores, and only 1 had an acceptable score. Understandability ranged from 55-91%, and only 3 PROMs had ideal scores. None of the instruments met the definition of plain-language best practice. ELF-Q identified points for improvement in several dimensions of the PROMs (Table). Conclusions: The use of comprehensive linguistic and cultural frameworks during the development and translation of self-reported PGD instruments may improve experience and engagement of patients with lower literacy and foster equity and representation in cancer cohort studies. [Table: see text]

Cardiac dysfunction (CD) among breast cancer survivors (BCS): Role of cardiotoxic therapy and cardiovascular risk factors.

12078 Background: CD is the leading cause of mortality in BCS. Limited information regarding the long-term risk of CD after exposure to cardiotoxic agents (anthracyclines, trastuzumab/pertuzumab and left breast/chest wall radiation) and additional factors that place BCS at highest risk has precluded development of surveillance guidelines for early detection of CD in BCS. Methods: Patients, who survived ≥1y after BC and received potentially cardiotoxic therapy, underwent q2y echocardiographic (echo) screening at a single center. All echos from BC diagnosis to CD or last follow-up were processed in Python via automated text analysis. New-onset CD was defined as left ventricular ejection fraction (LVEF) &lt;50% or fractional shortening (FS) &lt;28%. Treatment exposures, demographics and cardiovascular risk factors (CVRFs: diabetes, hypertension, dyslipidemia, obesity, smoking) were abstracted from medical records. Cumulative incidence of CD described the magnitude of risk in the entire cohort and among those who received anthracyclines (±radiation), trastuzumab/pertuzumab (±radiation), and radiation alone (without anthracyclines/trastuzumab/pertuzumab). Cox proportional hazards models estimated hazard ratios (HRs) and 95%CIs of associations between CD and treatment, adjusting for CVRFs (time varying) and demographics. Longitudinal echocardiographic analyses (adjusted for covariates) estimated trends in cardiac function prior to CD. Results: We evaluated 2,540 echos in 847 BCS exposed to potentially cardiotoxic therapy (median age at BC: 54.4y; median follow-up: 7.6y). Overall, 39% of BCS received anthracyclines, 17% received trastuzumab/pertuzumab, 5% received anthracyclines + trastuzumab/pertuzumab, and 39% received radiation alone. The cumulative incidence of CD in the entire cohort was 12.3% at 2y, increasing to 29.7% by 10y after BC. Multivariable analysis revealed the following to be independently associated with CD: treatment (anthracyclines: HR=2.72, 95%CI=1.9-3.9; trastuzumab/pertuzumab: HR=1.92, 95%CI=1.1-3.2; ref: radiation alone), presence of CVRFs before CD diagnosis (hypertension: HR=1.77, 95%CI=1.3-2.4; obesity: HR=1.82, 95%CI=1.4-2.4; and dyslipidemia: HR=1.63, 95%CI=1.3-2.1). Adjusted longitudinal echo analyses revealed a statistically significant decline in SF over 20y from BC diagnosis (but prior to CD onset) for the entire cohort (p=0.0007), trastuzumab/pertuzumab (p=0.006), anthracyclines (p=0.037), and radiation alone (p=0.01). Conclusions: In this large cohort of BCS followed longitudinally for an extended period, exposure to potentially cardiotoxic therapy results in a substantial burden of CD; the risk increases over time, and with CVRFs. These results provide evidence for close surveillance of BC survivors at increased risk of CD for extended periods, and aggressive management of CVRFs before, during and after BC.

Dietary Fats and All-cause and Breast Cancer-specific Mortality among Women with Breast Cancer: The Western New York Exposures and Breast Cancer Study.

Study results of prediagnostic dietary fat intake and breast cancer mortality have been inconclusive. While dietary fat subtypes [saturated (SFA), polyunsaturated (PUFA), and monounsaturated (MUFA) fatty acids] may have different biological effects, there is little evidence regarding the association of dietary fat and fat subtype intake with mortality after breast cancer diagnosis. Women with incident, pathologically confirmed invasive breast cancer and complete dietary data (n = 793) were followed in a population-based study, the Western New York Exposures and Breast Cancer study. Usual intake before diagnosis of total fat and subtypes were estimated from a food frequency questionnaire completed at baseline. HRs and 95% confidence intervals (CI) for all-cause and breast cancer-specific mortality were estimated with Cox proportional hazards models. Interactions by menopausal status, estrogen receptor (ER) status, and tumor stage were examined. Median follow-up time was 18.75 years; 327 (41.2%) participants had died. Compared with lower intake, greater intake of total fat (HR, 1.05; 95% CI, 0.65-1.70), SFA (1.31; 0.82-2.10), MUFA (0.99; 0.61-1.60), and PUFA (0.99; 0.56-1.75) was not associated with breast cancer-specific mortality. There was also no association with all-cause mortality. Results did not vary by menopausal status, ER status, or tumor stage. Prediagnostic intake of dietary fat and fat subtypes was not associated with either all-cause or breast cancer mortality in a population-based cohort of breast cancer survivors. Understanding factors affecting survival among women diagnosed with breast cancer is critically important. Dietary fat intake prior to diagnosis may not impact that survival.