Objectives: Obesity affects more than one-third of US adults and promotes the development and progression of many cancers. One mechanism by which obesity may influence ovarian cancer is via adipocyte-tumor cell communication, including the production of chemotactic cytokines, or chemokines, in the tumor microenvironment. Our goal was to measure adipogenic chemokine expression and examine associations with body mass index (BMI) and ovarian cancer survival. Methods: We selected 25 chemokines from laboratory experiments that modeled adipogenesis, of which 22 had expression data available for 302 high-grade serous ovarian cancer cases from The Cancer Genome Atlas (TCGA). We measured mRNA expression using Taqman quantitative real-time polymerase chain reaction (QPCR) in 50 high-grade serous tumor samples and ascertained BMI from electronic medical records (EMR) from the Vanderbilt University Medical Center (VUMC). We also evaluated peri-diagnosis BMI (± 4 weeks) from EMR for 298 tumor registry-confirmed (165 serous and 133 nonserous) cases. Expression data were log transformed and dichotomized at median values; associations with disease-free survival (DFS) and overall survival (OS) were quantified using proportional hazards regression. Results: In TCGA analyses, two ligands and one receptor had significant associations with better (CCL8 and CXCL13) and worse (CX3CR1) OS, respectively. Higher CXCL13 expression was also associated with better DFS in VUMC tumors, but significance was attenuated after adjustment for clinical covariates. CX3CR1 was not expressed in all tumors but was positively associated with BMI (Pearson’s r=0.69, P=0.019). Finally, among all 298 tumor registry-confirmed VUMC cases, BMI was not associated with OS, although higher BMI was associated with a 4% significantly increased risk of death among 133 nonserous cases. Conclusions: Our findings suggest that obesity impacts chemokine expression in the tumor microenvironment and ovarian cancer survival. Further, differences between serous and nonserous subtypes may influence the relationship between obesity and ovarian cancer prognosis. Citation Format: Sara Duque, Nneka Anyanwu, George Bukenya, Shriya Karam, Deok-Soo Son, Andrew J. Wilson, Marta A. Crispens, Alicia Beeghly-Fadiel. Adipogenic chemokines, body mass index, and ovarian cancer survival [abstract]. In: Proceedings of the AACR Special Conference on Advances in Ovarian Cancer Research; 2019 Sep 13-16, 2019; Atlanta, GA. Philadelphia (PA): AACR; Clin Cancer Res 2020;26(13_Suppl):Abstract nr B54.

Background: Obesity has been implicated in the progression of ovarian cancer, but associations between body mass index (BMI) and survival outcomes are mixed. Studies evaluating pre- or peri-diagnosis BMI have generally found associations between higher BMI and worse overall survival (OS), but studies with post-diagnosis BMI measures are limited. Furthermore, no existing post-diagnosis BMI and ovarian cancer survival studies have conducted analyses stratified by race. Objective: We undertook this study to characterize changes over time in BMI among ovarian cancer cases from the Vanderbilt University Medical Center (VUMC) and to evaluate associations in relation to overall survival (OS); differences by race were examined in stratified analyses. Methods: We assembled a retrospective cohort of Tumor Registry confirmed ovarian cancer cases from VUMC EMR. BMI (kg/m2) at diagnosis, and 6, 12, 18, and 24 months after diagnosis (±8 weeks) were used to classify obesity according to the World Health Organization (WHO) guidelines as underweight (<18.5), normal weight (<25), overweight (<30), or obese (≥30.0). BMI changes over time were determined by the ratio of the last to first available measure, and categorized as increased (>1.05%), decreased (<0.95), or stable (reference). Associations with OS were quantified by Hazard Ratios (HRs) and 95% Confidence Intervals (CIs) from Cox proportional-hazards regression in multivariable adjusted race-stratified analyses. Results: Among 380 ovarian cancer cases with peri- or post-diagnosis BMI measures available from EMR, the relative prevalence of WHO normal weight cases decreased over time (diagnosis: 40.4%; 6 months: 38.2%; 12 months: 35.5%; 18 months: 27.7%; 24 months: 22.8%); there were more cases who had a decreased BMI (46.6%) than increased BMI (28.4%). Neither the prevalence nor percent of patients with changes differed by race. Among Caucasians (86.8%), a BMI increase was associated with a decreased risk of death (HR: 0.62, 95% CI: 0.40-0.97) while a BMI decrease was associated with an increased risk of death (HR: 2.18, 95% CI: 1.51-3.17). However, this seemed to differ among African Americans (6.1%), where both an increased and decreased BMI tended to have worse OS. Conclusions: Changes in BMI after a diagnosis of ovarian cancer may have different associations with mortality by race. To further expand upon these preliminary findings, our next steps include conducting analysis with time-varying covariates and identifying collaborators with ovarian cancer cases who have post-diagnosis BMI measures available for additional analysis. Citation Format: Alicia Beeghly-Fadiel, Nneka J Anyanwu, Shyria Karam, Demetra Hufnagel, George Bukenya, Sara Duque, Deok Son, Andrew J Wilson, Marta A Crispens. Body mass index, race, and ovarian cancer: Characterizing changes after diagnosis and associations with overall survival [abstract]. In: Proceedings of the Twelfth AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2019 Sep 20-23; San Francisco, CA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2020;29(6 Suppl_2):Abstract nr C069.

Background: Pseudoangiomatous stromal hyperplasia (PASH) is a proliferative mesenchymal lesion of the breast histologically defined by dense collagenous stroma with spindle-shaped myofibroblasts. Characterization of patients with PASH has largely been limited to histologic and radiographic studies with small sample sizes due to the rarity of this benign condition. Furthermore, no existing studies of PASH have yet to consider race. Objectives: We undertook this study to identify subjects with PASH from electronic medical records (EMR) at the Vanderbilt University Medical Center (VUMC) and evaluate patient characteristics, including differences by race. Approach: We used natural language processing (NLP)-assisted searches to identify 283 subjects with pathology reports containing ‘PASH’ or ‘pseudoangiomatous stromal hyperplasia’. Results: We confirmed PASH by pathology report for 229 subjects (80.9%). The majority were Caucasian women, although 24 African Americans (10.5%) and 4 males (1.7%) with PASH were also identified. A total of 27 (11.8%) had a diagnosis of cancer as some point in their EMR, most frequently breast (74.1%) or thyroid (14.8%) cancer. The median age at PASH diagnosis was 45.4 years (range: 12.9-84.6) but this seemed lower among Asians (31.4 years). PASH was found more frequently in the left breast (54.8%) and was ascertained most often by core needle biopsy (76.0%), stereotactic needle biopsy (11.8%), or excision (7.0%). Among initially biopsied patients, 20 (9.9%) and 39 (19.3%) had subsequent biopsies and excisions, respectively. No significant differences in any of these characteristics by race were identified. Conclusions: This data confirms that PASH is not exclusive to Caucasians and suggests that patient characteristics do not differ by race. Analysis of additional characteristics and patient outcomes, with inclusion of EMR time, is currently underway. Citation Format: George Bukenya, Demetra Hufnagel, Nneka Anyawu, Stephanie Kurita, Alicia Beeghly-Fadiel. Characteristics of patients with pseudoangiomatous stromal hyperplasia (PASH): A retrospective cohort study [abstract]. In: Proceedings of the Twelfth AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2019 Sep 20-23; San Francisco, CA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2020;29(6 Suppl_2):Abstract nr B118.