Adjuvant nivolumab, capecitabine or the combination in patients with residual triple-negative breast cancer: the OXEL randomized phase II study

Chemotherapy and immune checkpoint inhibitors have a role in the post-neoadjuvant setting in patients with triple-negative breast cancer (TNBC). However, the effects of nivolumab, a checkpoint inhibitor, capecitabine, or the combination in changing peripheral immunoscore (PIS) remains unclear. This open-label randomized phase II OXEL study (NCT03487666) aimed to assess the immunologic effects of nivolumab, capecitabine, or the combination in terms of the change in PIS (primary endpoint). Secondary endpoints included the presence of ctDNA, toxicity, clinical outcomes at 2-years and association of ctDNA and PIS with clinical outcomes. Forty-five women with TNBC and residual invasive disease after standard neoadjuvant chemotherapy were randomized to nivolumab, capecitabine, or the combination. Here we show that a combination of nivolumab plus capecitabine leads to a greater increase in PIS from baseline to week 6 (91%) compared with nivolumab (47%) or capecitabine (53%) alone (log-rank p = 0.08), meeting the pre-specified primary endpoint. In addition, the presence of circulating tumor DNA (ctDNA) is associated with disease recurrence, with no new safety signals in the combination arm. Our results provide efficacy and safety data on this combination in TNBC and support further development of PIS and ctDNA analyses to identify patients at high risk of recurrence.

Adjuvant nivolumab, capecitabine or the combination in patients with residual triple-negative breast cancer: the OXEL randomized phase II study

Chemotherapy and immune checkpoint inhibitors have a role in the post-neoadjuvant setting in patients with triple-negative breast cancer (TNBC). However, the effects of nivolumab, a checkpoint inhibitor, capecitabine, or the combination in changing peripheral immunoscore (PIS) remains unclear. This open-label randomized phase II OXEL study (NCT03487666) aimed to assess the immunologic effects of nivolumab, capecitabine, or the combination in terms of the change in PIS (primary endpoint). Secondary endpoints include the presence of ctDNA, toxicity, clinical outcomes at 2-years and association of ctDNA and PIS with clinical outcomes. Forty-five women with TNBC and residual invasive disease after standard neoadjuvant chemotherapy were randomized to nivolumab, capecitabine, or the combination. Here we show that a combination of nivolumab plus capecitabine leads to a greater increase in PIS from baseline to week 6 (91%) compared with nivolumab (47%) or capecitabine (53%) alone (log-rank p = 0.08), meeting the pre-specified primary endpoint. In addition, the presence of circulating tumor DNA (ctDNA) was associated with disease recurrence, with no new safety signals in the combination arm. Our results provide efficacy and safety data on this combination in TNBC and support further development of PIS and ctDNA analyses to identify patients at high risk of recurrence.

Abstract P5-07-03: The Association Between Symptom Severity and Physical Function among Participants in I-SPY2

Background. Patient-reported outcomes (PROs) are increasingly recognized as a valuable component to understand treatment tolerability and toxicity among patients on clinical trials. We have implemented a system for monitoring patient reported outcomes (PROs), symptoms, and quality of life (QOL) using electronic PRO (ePRO) instruments for patients enrolled in the I-SPY2 trial. I-SPY2 is a phase II multi-site clinical trial evaluating the effect of novel neoadjuvant therapies for locally advanced breast cancer. We correlated patient demographic factors with symptoms, investigated the trajectory of symptoms throughout treatment, and sought to characterize symptoms associated with decline in physical function (PF). Methods. Our study population included 259 I-SPY2 patients that completed surveys on one of 9 study arms (including novel oral taxane/immunotherapy combinations, IV paclitaxel, checkpoint inhibitor+/- LAG3 inhibitor, and control IV paclitaxel +/- trastuzumab/pertuzumab). After the 12 week period of investigational agents, most patients received standard adriamycin and cyclophosphamide (AC). Symptom severity, frequency, and interference was assessed weekly using 33 items from the PRO-CTCAE item bank. PF was assessed using the NIH PROMIS instrument and was evaluated at baseline, inter-regimen (after 12 weeks of treatment), pre-surgery, and 1 and 6 months at follow-up. An odds ratio was used to assess univariate associations between age and race, and symptoms. Regularized multi-variate regression was used to evaluate early symptoms (prior to week 6) predictive of a clinically significant (>5 point T-score) decline in PF from baseline to post-treatment follow-up among all races and age groups. Results. Of 259 patients (mean age (SD) = 46.8 (13.6)), 160 (58%) were White, 13 (5%) were Asian, 26 (10%) were African American (AA), 25 (9.3%) were Hispanic, and 35 (13.5%) self-reported “Other”. At baseline, AA patients had a higher severity of joint pain than White patients (OR = 14.9, P < 0.05). During study treatment with paclitaxel and/or novel agent within the first 12 weeks of treatment, AA patients and non-white (NW) patients were more likely to report severe vomiting than White patients (OR =13.22 and 12.72, P< 0.05 and P< 0.03 respectively). During treatment with AC, NW patients were more likely to report higher severity of neuropathy than White patients (OR = 5.43, P< 0.03). Among all patients, in analysis of early symptoms predictive of a clinically significant decline in PF between baseline and 1 month post treatment, predictors included high frequency of diarrhea, severity of itching, and severity of joint pain. Further analysis of symptom trajectories revealed that frequency of diarrhea reported rose sharply between baseline and Cycle 2 with 9 patients (7%) reporting occasional or frequent diarrhea to 39 patients (28%) reporting occasional to almost constant diarrhea and remained stable at that proportion for the remainder of treatment. Frequency of diarrhea declined slightly during AC (17%) and dropped to baseline levels by follow-up. In contrast, severity of joint pain persisted post-treatment, rising consistently from baseline through follow- up with 3 patients (2%) reporting moderate to severe joint pain at baseline to 18 patients (35%) reporting moderate to severe joint pain at follow-up. Conclusion. Among I-SPY2 participants, when higher grade of diarrhea is persistent (or uncontrolled), it impacts physical function even after end of therapy. In some cases, race was also a determinant in symptom trajectory, although a higher enrollment of AA and NW patients will enable more robust estimates to be computed. While some of these early symptom predictors are transient and resolve by the time of follow-up, others persist long-term and contribute more directly towards impaired physical function at follow-up.

Citation Format: Amrita Basu, Saumya Umashankar, Kaylee Blevins, Anna Northrop, Anika Christofferson, Ebunoluwa Olunuga, Jaeyoon Cha, Ananya Mittal, Julissa Molina-Vega, Laura Sit, Thelma Brown, Bev Parker, Diane Heditsian, Susie Brain, Carol Simmons, Alessandra Taboada, Tina J. Hieken, Kathryn Ruddy, Carolina Salvador, Candace Mainor, Anosheh Afghahi, Sarah Tevis, Anne Blaes, Irene M. Kang, Jane Perlmutter, Hope Rugo, Sai Kanaparthi, Garry Peterson, Lisa T. Weiss, Adam Asare, Laura J. Esserman, Michelle Melisko, Dawn Hershman. The Association Between Symptom Severity and Physical Function among Participants in I-SPY2 [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr P5-07-03.

A phase I trial of palbociclib and bosutinib with fulvestrant in patients with metastatic hormone receptor positive and HER2 negative (HR+ HER2-) breast cancer refractory to an aromatase inhibitor and a CDK4/6 inhibitor

Standard treatment for metastatic hormone positive (HR+) breast cancer includes a combination of a CDK4/6 inhibitor and antiestrogen therapy. Despite durable responses, eventual endocrine resistance results in disease progression. The Src/Abl pathway has been shown to mediate endocrine resistance in breast cancer, thus providing a promising target for novel therapies. Bosutinib is a tyrosine kinase inhibitor that targets the Src/Abl pathway, which has been studied in hematologic malignancies. Preclinical data suggests that the addition of bosutinib to a CDK4/6 inhibitor and antiestrogen therapy has the potential to reverse endocrine resistance. This is a phase I, single arm, open-label clinical trial in which we evaluate the combination of palbociclib and fulvestrant with bosutinib in metastatic HR+ breast cancer. Patients with confirmed advanced HR+/HER2- breast cancer who have received no more than three lines of chemotherapy and have progressed on at least one aromatase inhibitor and one CDK4/6 inhibitor will be enrolled. Participants will be given a combination of palbociclib, fulvestrant and bosutinib over 28-day cycles. The primary objective of this study is to assess the safety and tolerability of bosutinib in combination with palbociclib and fulvestrant in the study population. Secondary objectives are to 1) determine the anti-tumor effect of this therapeutic combination by assessing overall response rate (ORR) and clinical benefit rate (CBR) after 6 months of treatment, 2) to determine the clinical pharmacology parameters of bosutinib in this regimen, and 3) to build a tissue repository at Georgetown Lombardi Comprehensive Cancer Center for further translational study.

HALT-D: a randomized open-label phase II study of crofelemer for the prevention of chemotherapy-induced diarrhea in patients with HER2-positive breast cancer receiving trastuzumab, pertuzumab, and a taxane

PURPOSE

To assess whether crofelemer would prevent chemotherapy-induced diarrhea (CID) diarrhea in patients with HER2-positive, any-stage breast cancer receiving trastuzumab (H), pertuzumab (P), and a taxane (T; docetaxel or paclitaxel), with/without carboplatin (C; always combined with docetaxel rather than paclitaxel).

METHODS

Patients scheduled to receive ≥ 3 consecutive TCHP/THP cycles were randomized to crofelemer 125 mg orally twice daily during chemotherapy cycles 1 and 2 or no scheduled prophylactic medication (control). All received standard breakthrough antidiarrheal medication (BTAD) as needed. The primary endpoint was incidence of any-grade CID for ≥ 2 consecutive days. Secondary endpoints were incidence of all-grade and grade 3/4 CID by cycle/stratum; time to onset and duration of CID; stool consistency; use of BTAD; and quality of life (Functional Assessment of Chronic Illness Therapy for Patients With Diarrhea [FACIT-D] score).

RESULTS

Fifty-one patients were randomized to crofelemer (n = 26) or control (n = 25). There was no statistically significant difference between arms for the primary endpoint; however, incidence of grade ≥ 2 CID was reduced with crofelemer vs control (19.2% vs 24.0% in cycle 1; 8.0% vs 39.1%, in cycle 2). Patients receiving crofelemer were 1.8 times more likely to see their diarrhea resolved and had less frequent watery diarrhea.

CONCLUSION

Despite the choice of primary endpoint being insensitive, crofelemer reduced the incidence and severity of CID in patients with HER2-positive breast cancer receiving P-based therapy. These data are supportive of further testing of crofelemer in CID.

TRIAL REGISTRATION

Clinicaltrials.gov, NCT02910219, prospectively registered September 21, 2016.

Abstract P4-12-02: Improving patient-reported outcome data capture for clinical research: ePRO in ISPY 2, a phase 2 breast cancer study

Introduction: Advances in technology and internet capability have provided an opportunity for efficient collection of Patient Reported Outcomes (PRO) during medical treatment. Here we describe the development and implementation of a system for monitoring patient reported adverse events (AEs) and quality of life (QoL) using electronic PRO (ePRO) instruments for patients enrolled on the Investigation of Serial studies to Predict Your Therapeutic Response with Imaging And moLecular analysis (I-SPY 2 TRIAL), a phase II adaptive platform clinical trial for locally advanced breast cancer. Methods: We designed an ePRO system to increase the accuracy of patient-reported QoL and AE data collection with the intent to act on symptoms in real time. Using the OpenClinica electronic data capture system, we developed rules-based logic to build automated ePRO surveys, customized to the I-SPY 2 treatment schedule. Weekly surveys contained a maximum of 126 validated, branching logic questions from the Patient Reported Outcomes Measurement Information System (PROMIS®) Health Measures and the National Cancer Institute’s Patient Reported Outcomes - Common Terminology Criteria for Adverse Events (PRO-CTCAE™) instruments. We piloted ePROs at the University of California, San Francisco (UCSF) to evaluate compatibility with a variety of I-SPY 2 patient scenarios (e.g., dose delays). We then staggered rollout of the ePRO system to 22 I-SPY 2 sites to ensure technological feasibility. In order to improve accuracy of data collection, we utilized real-time tracking and developed a Clinical Research Coordinator (CRC) training manual, which integrated workflow diagrams with technical solutions. CRCs were trained using remote video sessions. Results: The UCSF ePRO pilot began in September of 2020. Over 9-months, we accrued 43 I-SPY 2 patients (average age of 43.8 years), whose interactions with the ePRO system informed design improvements. Of the patients who received a baseline ePRO survey, the completion rate was 75.9% (average age of 44.2 years). This represents an increase from the 15-20% baseline completion rate for the 360 UCSF I-SPY 2 patients who received paper-based PRO surveys between May 2012 - January 2019. As of June 2021, the ePRO system was operational at all 22 I-SPY 2 sites. The UCSF pilot revealed that engagement with patients at critical timepoints improved survey completion. CRCs facilitated patient participation by sending instructional emails and communicating with patients weekly. We tracked data completeness using a Patient Tracking report, which displayed each patient’s survey completion history. This real-time tool enabled CRCs to identify patients who had not completed ePRO surveys prior to their visit, so they could be provided a tablet computer to complete the survey in the clinic. After introducing tablets into the workflow at UCSF, patient completion of the baseline survey increased from 75.9% to 80%. Conclusion: The transition from paper to electronic QOL and AE data collection improves the ability of patients to complete PRO surveys, but the process must also be optimized and integrated into clinical workflow and trial conduct. In the future, we will present additional results highlighting the feasibility of multilingual ePRO integration into I-SPY 2. ePRO also provides a new opportunity for data analysis, as well as the potential to reduce high grade toxicity through early intervention. It will allow us to assess QoL and AE data by drug regimen, site, provider, and study treatment. The creation of clinician-facing reports also enables access to patient responses in real-time. By implementing ePRO within I-SPY 2, we not only increase efficiency and accuracy of patient-reported data collection, but also improve quality of care and patient safety.

Citation Format: Anna Northrop, Anika Christofferson, Michelle Melisko, Laura Sit, Ebunoluwa Olunuga, Ananya Mittal, Adi Goldman, Thelma Brown, Diane Heditsian, Bev Parker, Susie Brain, Carol Simmons, Alessandra Taboada, Kathryn J Ruddy, Tina Hieken, Mara Piltin, Kiri Cook, Carolina Salvador, Candace Mainor, Anosheh Afghahi, Sarah Tevis, Anne Blaes, Irene Kang, Susan Melin, Laura Esserman, Adam Asare, Dawn L Hershman, Amrita Basu. Improving patient-reported outcome data capture for clinical research: ePRO in ISPY 2, a phase 2 breast cancer study [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr P4-12-02.

Abstract PD9-02: Peripheral immune subsets and circulating tumor DNA (ctDNA) in patients (pts) with residual triple negative breast cancer (TNBC) treated with adjuvant immunotherapy and/or chemotherapy (chemo): The OXEL study

Background: Poor clinical outcomes are noted in pts with TNBC who do not achieve a pathologic complete response (pCR). We characterized peripheral immune subsets and the role of minimal residual disease (MRD) detection via ctDNA in pts who participated in the OXEL study. Methods: OXEL (Opdivo® -XELoda ®) is a recently completed phase II open-label 3-arm randomized study of nivolumab (nivo), capecitabine (cape) or the combination as adjuvant therapy (tx) for pts with residual TNBC after appropriate neoadjuvant chemo. Residual disease was defined as ≥ 1.0 cm of primary tumor and/or nodal involvement. Eligible pts had completed definitive local tx. Pts were randomly assigned to nivo 360 mg iv q3wks x 6 (arm A); cape 1250mg/m2 po bid D1-D14 q3 wks x 6 (arm B); nivo 360mg iv q3wks + cape 1250mg/m2 po bid D1-D14 q3 wks x 6 (arm C). Peripheral blood mononuclear cells (PBMCs) and ctDNA were assessed at baseline (D1 of cycle 1), 6, and 12 wks and at time of recurrence, if applicable. PBMCs were stained with 30 markers and analyzed by flow cytometry to identify changes in 158 immune cell subsets at 6 wks, as a percent of total PBMCs. RaDaRTM, a deep sequencing based, tumor-informed personalized assay was utilized to detect the presence of ctDNA in plasma. Distant disease-free survival (DDFS) and overall survival (OS) were analyzed by the Kaplan-Meier method and Log-Rank test was used to compare DDFS and OS according to baseline MRD results. All pts will be followed for distant recurrence and survival for 3 yrs. Here we report the translational endpoints of the OXEL study. Clinical endpoints according to treatment received will be reported in a future analysis. Results: 45 pts were enrolled between 8/2018 and 6/2021. 29 (64%) were Caucasian and 14 (31%) were African American. Mean age at enrollment was 51 [+/- 12]. 93% of pts received a taxane-anthracycline containing neoadjuvant tx. 15 pts were randomized to each arm. DDFS probability at 1-yr and 2-yrs was 0.71 (+/- 0.07) and 0.66 (+/- 0.08) respectively. At 12 mos of median follow up, 13/45 pts (29%) experienced distant recurrence, none had local recurrence. 43 pts were evaluated for PBMC subsets. Changes in PBMC subsets at 6 wks were different amongst the arms; in arm A, reductions in NK subsets, including a 33% reduction in CD56dimCD16- cells, were observed, while in arm B, increases in naïve CD4+ T cells (+45%) and CD73+CD8+ T cells (+12%) and reductions in ki67+CD8+ T cells (-48%) were noted. In arm C, increases were observed in conventional dendritic cells (+36%), effector memory ki67+CD4+ T cells (+46%), and CD56dimCD16- NK cells (+29%). 33 pts underwent successful MRD analysis. 12/33 (36%) pts were MRD+ at baseline. 2/12 pts MRD+ at baseline subsequently cleared MRD, with undetectable ctDNA on future time points; neither patient has had recurrence to date. The remaining 10/12 MRD+ pts (83%) have experienced distance recurrence. 21/33 (64%) pts were ctDNA negative at baseline; 20/33 remained negative for all follow up timepoints. 10/11 pts experiencing distant recurrence were MRD+ at baseline, compared to 1/11 pt who became MRD+ at wk 6 post initiation of tx. At 12 mos of median follow-up, baseline MRD+ testing was significantly associated with an inferior DDFS ( p<0.0001 Log-rank test, median DDFS 4.0 mos vs. not reached) and OS (p=0.02 Log-rank test, median OS not reached for both groups). Results will be updated at the time of abstract presentation. Conclusions: Changes in PBMC subsets were associated with receipt of chemo and/or immunotherapy. Our results suggest that baseline MRD+ in pts without pCR is a poor prognostic factor. Future trials aiming to optimize adjuvant treatment with chemo and/or immunotherapy in residual TNBC should consider incorporating ctDNA as a selection marker of pts at higher risk of recurrence.

Citation Format: Filipa Lynce, Candace Mainor, Xue Geng, Greg Jones, Ilana Schlam, Hongkun Wang, Ute Feger, Renee Donahue, Nicole Toney, Caroline Jochems, Jeffrey Schlom, Christopher Gallagher, Rita Nanda, Deena Graham, Erica M Stringer-Reasor, Neelima Denduluri, Julie Collins, Asma A Dilawari, Ami Chitalia, Shruti Tiwari, Raquel Nunes, Rebecca Kaltman, Katia Khoury, Margaret Gatti-Mays, Sandra M Swain, Heather A. Parsons, Paula Pohlmann, Claudine Isaacs. Peripheral immune subsets and circulating tumor DNA (ctDNA) in patients (pts) with residual triple negative breast cancer (TNBC) treated with adjuvant immunotherapy and/or chemotherapy (chemo): The OXEL study [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr PD9-02.

Abstract P5-12-13: Factors associated with aromatase inhibitor associated musculoskeletal symptoms in early stage breast cancer patients

Introduction: Aromatase Inhibitors (AI) are a mainstay of treatment for hormone receptor positive breast cancer (BC) in postmenopausal women. However, patients under AI treatment often suffer from Aromatase Inhibitor-Associated Musculoskeletal Symptoms (AIMSS), defined as myalgias, arthralgias, or joint stiffness in hands, fingers, knees, hips and feet, which may limit treatment and subsequently increase BC mortality. The depletion of estrogen following AI initiation has been postulated to cause AIMSS. However, it has not yet been fully established about the clinical significance of AIMSS and its risk factors and the underlying mechanisms. The aim of this study is to identify epidemiologic risk factors associated with the development of AIMSS in BC patients on adjuvant AI therapy.

Methods: We conducted a retrospective review of medical records among 111 stages 0-III hormone receptor positive BC patients who received adjuvant AI therapy, including anastrozole, letrozole or exemestane, between January 2009 and June 2017, at the University of Maryland Greenebaum Comprehensive Cancer Center. To be eligible, women had to be either postmenopausal or premenopausal, on a GnRH analog for at least 3 months, and report of no AIMSS before the AI initiation. We collected patients’ data on demographics, lifestyle factors, reproductive histories, tumor characteristics, types of AI and cancer treatment, co-morbidities, use of medications, length of AI therapy, and onset and severity of AIMSS. We defined severe AIMSS as AIMSS requiring the change of AI therapy or discontinuation of AI. The demographics and clinicopathological characteristics of patients on AI therapy and frequencies of AIMSS onset during AI treatment was characterized with means (SDs) or percentages. To estimate the association of severe AIMSS with epidemiologic risk factors, odds ratios (ORs) and 95% confidence intervals were calculated using multivariable-adjusted logistic regression.

Results: This study included 111 BC patients with a mean age of 64 years. Their mean BMI at time of BC diagnosis was 30.8 kg/m2, and 41% of patients were White, 39% were Black, 4% were Asian/Pacific Islander, and 16% were other. Majority of patients (77%) had no prior experience of tamoxifen use and used letrozole as an initial AI therapy (70.3%); 15.3% used anastrozole, and 14.4% used exemestane. We found that 52.6% of patients experienced any degree of AIMSS while on AI therapy. In particular, 19.8% of patients required change or holding AI therapy, and 3.6% permanently discontinued AI because of AIMSS. We observed that older age at AI therapy initiation and higher levels of circulating vitamin D-25OH were significantly associated with lower odds of severe AIMSS. Multivariate ORs (95% Cis) comparing the highest to the lowest tertile of age and vitamin D levels at AI initiation were 0.18 (0.05-0.71) and 0.10 (0.02-0.49), respectively (all P-trend: ≤ 0.04). There were no significant association between race, body mass index, smoking, alcohol use, reproductive factors, type of AI therapy, and tamoxifen use prior to AI therapy.

Conclusions: In our study, nearly 52.6% of BC patients on adjuvant AI therapy experienced AIMSS. Of those, 23.4% changed or discontinued AI regimen due to severe AIMSS. Older age and higher circulating levels of vitamin D-25OH at the initiation of adjuvant AI therapy were associated with lower risk of developing severe AIMSS. The study results support the clinical significance of AIMSS in BC patients leading to decreased adherence to AI therapy. The replication of our association with age and vitamin D levels in a larger study is warranted as it may provide a greater insight into etiologies of AIMSS.

Citation Format: Yamin Sun, Simran Elder, Seungyoun Jung, Candace Mainor, Shruti Murali, Paula Rosenblatt, Katherine Tkaczuk. Factors associated with aromatase inhibitor associated musculoskeletal symptoms in early stage breast cancer patients [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 P5-12-13.

Abstract P5-13-02: Chemoprevention utilization in patients with a history of atypical hyperplasia, atypical lobular hyperplasia, or lobular carcinoma in-situ: A retrospective chart review of patients diagnosed at an urban hospital with a large minority patient population

Background: The Breast Cancer Prevention Trial (BCPT) and the Study of Tamoxifen and Raloxifene (STAR) trial showed that chemoprevention can reduce the risk of invasive breast cancer by nearly 50%. Despite these results, studies have shown that while an estimated 2 million women in the United States are eligible for chemoprevention, actual acceptance of these medications is low. Improving chemoprevention utilization rates hinges on better understanding current rates of utilization and factors affecting patient acceptance. Reported rates and barriers to chemoprevention use may not accurately reflect true utilization patterns in lower socioeconomic, minority patient populations. The aim of this IRB approved retrospective study was to characterize the rate and factors associated with chemoprevention use in patients with a diagnosis of atypical ductal hyperplasia (ADH), atypical lobular hyperplasia (ALH), or lobular carcinoma in-situ (LCIS) at an urban hospital with a high minority population.

Methods – A retrospective chart review was performed for all diagnoses of ADH, ALH, and LCIS made at the University of Maryland Medical Center between the years 2005-2015. Concurrent DCIS or invasive cancer were excluded. Demographic and clinical information including age, race, education, GAIL score, BMI, and use of chemoprevention was recorded. Univariable and multivariable logistic regression were performed to identify factors associated with chemoprevention discussion and use.

Results – 301 diagnoses of ADH/ALH/or LCIS were obtained and 127 women were eligible for analysis. The median age was 53 years old with 47% of patients being premenopausal. The majority were African-American (65%) and 51% had a high school degree or less. The median 5 year risk for developing breast cancer based on the GAIL model was 2.4%. The chemoprevention utilization rate for our patient population was 28% (n=34). Race, menopausal status, and breast density were not associated with chemoprevention discussion or use. We found that patients were more likely to have a chemoprevention discussion with their provider if they were older (p=0.03) or if they were referred to medical oncology (p<0.001, Fisher's exact test).

Conclusions - Our study evaluated chemoprevention use in an understudied predominantly African-American patient population. We found a higher rate of chemoprevention utilization (28%) compared to previously reported rates. Age and medical oncology referral had a significant impact on provider-patient chemoprevention discussion. Though limited due to small sample size, our study nonetheless provided thought provoking results. Older patients may be at higher risk for developing breast cancer, however, it is important to consider that younger patients with risk factors may have a more favorable endocrine therapy benefit-risk ratio. In addition, our results highlight the importance of encouraging all physicians who are involved in women's' breast health to have a chemoprevention discussion with eligible patients, or for these physicians to refer patients to a medical oncologist for further discussion.

Citation Format: DeRidder A, Mainor C, Goloubeva O, Ioffe O, Bellavance E, Tkaczuk K, Rosenblatt P. Chemoprevention utilization in patients with a history of atypical hyperplasia, atypical lobular hyperplasia, or lobular carcinoma in-situ: A retrospective chart review of patients diagnosed at an urban hospital with a large minority patient population [abstract]. In: Proceedings of the 2017 San Antonio Breast Cancer Symposium; 2017 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2018;78(4 Suppl):Abstract nr P5-13-02.

Abstract P1-08-05: An association between obesity and more aggressive breast cancer subtype

Background: Obesity has become epidemic in the United States with over one third of the adult population being obese. The incidence of obesity is even higher among African American population with slightly more than half of the population being obese. Elevated body mass index (BMI) has been associated with an increased risk of breast cancer in postmenopausal women. This is believed to be due to higher levels of estrogens, inflammation, and insulin. Given that adipose tissue is the major source of estrogen production via the aromatization of androgens into estrogens in postmenopausal women, there is a concern that obese women may have inadequate estrogen suppression with a fixed dose of aromatase inhibitors (AIs). Multiple recent retrospective analyses of large phase III adjuvant endocrine therapy trials demonstrated that obese patients had worse outcome when treated with AIs. However, it remains unclear whether there is also a difference in the tumor subtype among normal weight, overweight, and obese patients.

Method: A retrospective review of breast cancer cases diagnosed from July 2008 and August 2011 was performed. Patient characteristics, BMI, and pathologic findings were collected. An immunohistochemistry (IHC)-based molecular subtypes was assigned based on St. Gallen criteria: Luminal A (ER/PR+, HER2- and Ki67 ≤ 14%), Luminal B (ER/PR+, HER2+ or Ki67 > 14% or grade 3), HER2 enriched (ER/PR-, HER2+), and Triple negative breast cancer (TNBC; ER/PR-, HER2-). Using the WHO classification, normal weight was defined as BMI < 25, overweight BMI 25-30, and obese BMI ≥ 30.

Result: 143 patients were included in the analysis. The median age was 55 years. The majorities of our patients were African American (63%) and presented with early stage disease (stage I 29%, stage II 45%, stage III 22%, and stage IV 5%). IHC-based molecular subtypes were luminal A 20%, luminal B 48%, HER2 enriched 7%, and TNBC 25%. 21% were normal weight, 30% overweight and 49% obese. As shown in the table below, comparing between normal weight, overweight, and obese patients, there is a possible association between tumor subtype and BMI status (p = 0.03). The majority of TNBC patients (94.4%) were overweight or obese. Specifically, there is significantly more aggressive luminal B subtype compared to luminal A (52.86% vs. 15.71%) among obese patients (p = 0.017). We further evaluated the effect of obesity and tumor growth in xenograft model of MCF7Ca. Comparing between obese mice fed with high fat vs. lean mice fed with chow diet, there is a significant increase in tumor growth among obese mice (p = 0.04) which corresponds to the high proliferation index measured by Ki67 seen in patients with luminal B subtype.

IHC-based molecular subtypes and distribution by weight categoryWeight CategoryLuminal ALuminal BHER2 EnrichedTNBCNormal Weight8 (26.67%)15 (50%)5 (16.67%)2 (6.67%)Overweight9 (20.93%)17 (39.53%)2 (4.65%)15 (34.88%)Obese11 (15.71%)37 (52.86%)3 (4.29%)19 (27.14%)

Conclusion: Our study suggests that obesity may provide a microenvironment that support and accelerate tumor growth, particularly in luminal breast cancer subtypes. Furthermore, the poor outcome seen in obese patients may also be in part due to more aggressive tumor subtype.

Citation Format: Saranya Chumsri, Paula Rosenblatt, Candace Mainor, Gauri Sabnis, Olga Goloubeva, Angela H Brodie. An association between obesity and more aggressive breast cancer subtype [abstract]. In: Proceedings of the Thirty-Seventh Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2014 Dec 9-13; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2015;75(9 Suppl):Abstract nr P1-08-05.