Search

Table S1. Clinicopathological characteristics and sample distribution by site Table S2. Primers/probes design for ddPCR (ESR1, EIF2C, AP3P1) Table S3. Raw nanoString counts Table S4. Normalized copy number calls (ER+ samples) Table S5. Normalized copy number calls (ER- samples) Table S6. DdPCR analysis of cell lines Table S7. NanoString raw counts (cell lines) Table S8. NanoString copy number calls (cell lines) Table S9. Nanostring log2 normalized counts (cell lines) Table S10. ESR1 CN amplifications by site in ER- samples Table S11. Multiple Fisher exact comparisons Table S12. ESR1 5`-3` copy number imbalance Table S13. Correlation matrix of ESR1 copy number calls by the different exons and mRNA expression Table S14. Normalized copy number calls for other genes (ER+ samples) Table S15. Normalized copy number calls other genes (ER- samples)

Figure S1: Genomic organization of ESR1 nanoString probes Figure S2: Correlation of ESR1 amplification with mRNA expression Figure S3: ESR1 copy number status in ER+ primary-metastasis pairs Figure S4: Oncoprint visualization of copy number alterations in clinical samples Figure S5: eSNP-Karyotyping analysis of ERBB2 amplified samples Figure S6: Oncoprint visualization of copy number alterations in ER+ vs ER+ brain metastases Figure S7: Oncoprint visualization of copy number alterations in paired primary-met samples

Purpose: High-grade serous ovarian cancer (HGSOC) is an aggressive disease with few available targeted therapies. Despite high expression of estrogen receptor-alpha (ERα) in approximately 80% of HGSOC and some small but promising clinical trials of endocrine therapy, ERα has been understudied as a target in this disease. We sought to identify hormone-responsive, ERα-dependent HGSOC.Experimental Design: We characterized endocrine response in HGSOC cells across culture conditions [ two-dimensional (2D), three-dimensional (3D), forced suspension] and in patient-derived xenograft (PDX) explants, assessing proliferation and gene expression. Estrogen-regulated transcriptome data were overlapped with public datasets to develop a comprehensive panel of ERα target genes. Expression of this panel and ERα H-score were assessed in HGSOC samples from patients who received endocrine therapy. Time on endocrine therapy was used as a surrogate for clinical response.Results: Proliferation is ERα-regulated in HGSOC cells in vitro and in vivo, and is partly dependent on 3D context. Transcriptomic studies identified genes shared by cell lines and PDX explants as ERα targets. The selective ERα downregulator (SERD) fulvestrant is more effective than tamoxifen in blocking ERα action. ERα H-score is predictive of efficacy of endocrine therapy, and this prediction is further improved by inclusion of target gene expression, particularly IGFBP3.Conclusions: Laboratory models corroborate intertumor heterogeneity of endocrine response in HGSOC but identify features associated with functional ERα and endocrine responsiveness. Assessing ERα function (e.g., IGFBP3 expression) in conjunction with H-score may help select patients who would benefit from endocrine therapy. Preclinical data suggest that SERDs might be more effective than tamoxifen. Clin Cancer Res; 23(14); 3802–12. ©2017 AACR.

Supplementary File 3 contains the available pre- and post-treatment CA-125 data for our patient cohort.

Supplementary File 1 contains the genes included in the "EndoRx" NanoString code set.

This document contains the supplementary figures for the manuscript.

There are limited data in endometrial cancer for nodal control and appropriate treatment volume for non-surgically resected nodes treated with chemoradiotherapy (CRT) for patients who are not candidates for upfront extrafascial hysterectomy.Patients (n = 105) with clinical stage ≥ II endometrial cancer who were not candidates for upfront extrafascial hysterectomy treated with preoperative CRT were retrospectively reviewed. CRT included pelvic nodes to the common iliac for node-negative disease and para-aortic nodes to the renal vessel for any node-positive disease. Involved nodes most commonly received a boost of 55 Gy in 25 fractions ± additional 4-6 Gy sequential boost for nodes2 cm.Of the included 95 patients, 55 patients were node positive, with a total of 300 positive nodes. At a median follow-up of 25 months (interquartile range 9-46), the 3-year regional control was 91%. The 3-year involved nodal control rate was 96%. Involved nodal control was significantly higher in type I histology, nodes2 cm and by radiation dose (75% for55 Gy, 98% for 55 Gy in 25 fractions and 89% for55 Gy, P = 0.03). The 3-year para-aortic failure rate for node negative patients treated with pelvis-only CRT was significantly higher with positron emission tomography/computed tomography (PET/CT) versus computed tomography (CT)-based staging (0% versus 20%).This is the largest study examining regional control rates of involved lymph nodes with CRT for patients who were not candidates for upfront extrafascial hysterectomy. Nodal failure was low following CRT and dose ≥55 Gy in 25 fractions seems to be adequate for involved nodes.

Objectives A pooled analysis of PORTEC-1 & 2 identified substantial lymphovascular space invasion (LVSI) in 4.8% of patients, which predicted for pelvic recurrence, distant metastasis, and overall survival. Our institution implemented the PORTEC three-tier system of LVSI reporting (absent, focal, or substantial). We aimed to quantify the incidence of substantial LVSI in a North American population and to correlate extent of LVSI with lymph node (LN) involvement. Methods A retrospective review was conducted on patients with clinically uterine-confined, endometrioid type endometrial cancer who underwent surgical staging and were found to have pT1a-b disease. Binary logistic regression was used to assess predictors of LN involvement (defined as ITC, micrometastases, or macrometastases). Results In total, 438 patients with pT1a-b disease were identified. In the overall cohort and in the subset meeting PORTEC-1 inclusion criteria (n = 195), no LVSI was present in 67.4% and 50.8%; focal LVSI was present in 16.7% and 24.1%; and substantial LVSI was present in 16.0% and 25.1%, respectively. Among patients who underwent surgical LN assessment (79.2%, n = 347), LNs were involved in 3.3% without LVSI, 7.5% with focal LVSI (OR 2.4), and 15.2% with substantial LVSI (OR 5.3) (p = .005), with a similar trend in the PORTEC-1 cohort. Extent of LVSI correlated with disease burden in LN metastases. Conclusion Our incidence of substantial LVSI was three to five times higher than reported by PORTEC and correlated with LN involvement. This questions the reproducibility of the three-tier LVSI reporting system and emphasizes the need for multi-institutional data outside PORTEC for confirmation of our findings.

Purpose The role of neoadjuvant chemoradiation therapy in locally advanced type II endometrial cancer is controversial. We thus aimed to present our experience with the hypothesis that neoadjuvant chemoradiation therapy is associated with similarly high rates of downstaging and locoregional control for type II endometrial cancer and type I endometrial cancer. Methods and Materials Thirty-four patients with type II endometrial cancer with clinical evidence of cervical ± parametrium involvement treated with neoadjuvant external beam radiation therapy (45-50.4 Gy in 25-28 fractions) and high-dose-rate brachytherapy with a median total dose of 20 Gy (range, 15-27.5) in 4 fractions (range, 3-5) and concurrent platinum chemotherapy ± adjuvant chemotherapy from 2008 to 2018 were retrospectively reviewed. Patients with type I pathologic diagnoses and those treated with definitive (rather than preoperative) intent were excluded. Results Pathologic characteristics were as follows: 38% were carcinosarcoma, 18% serous, and 24% clear cell. Ninety-four percent of patients were downstaged to an extrafascial hysterectomy, and 94% had negative surgical margins. The 2-year local control, regional control, distant control, disease-free survival, and overall survival were 87.8%, 81.3%, 76.3%, 52.5%, and 63.7%, respectively. There was 1 subacute grade 3 and 1 late grade 3 small bowel obstruction, directly attributable to radiation therapy. Conclusions Neoadjuvant chemoradiation therapy effectively downstages the majority of locally advanced type II endometrial cancers, thereby increasing the likelihood of achieving complete resection with negative margins.

ObjectivesTo determine the impact of histological grade on overall survival in patients with clinical stage I endometrioid endometrial adenocarcinoma when radiation therapy is used as primary definitive treatment.MethodsPatients with stage I endometrioid endometrial adenocarcinomas who underwent definitive radiation therapy with brachytherapy ± external beam radiation therapy were identified from the National Cancer Database. Overall survival was estimated using the Kaplan-Meier method. Univariable and multivariable analyses were performed to determine factors affecting overall survival. Inverse probability of treatment weights were also used in multivariable analysis to estimate casual effects of external beam radiation therapy.ResultsA total of 947 patients were identified. Median overall survival for grade 1, grade 2, and grade 3 tumors was 62 months (95% CI 53.8 to 70.2), 48.5 months (95% CI 38.2 to 58.8), and 33.5 months (95% CI: 23.1 to 43.8), respectively. Grade, age, and insurance status were associated with overall survival in univariate analysis with only grade and age remaining significant in multivariate analysis. Brachytherapy with external beam radiation therapy was not associated with survival in comparison with brachytherapy alone. Compared with grade 1 tumors, patients with grade 3 (HR 1.4, 95% CI 1.15 to 1.89), but not grade 2 (HR 1.0, 95% CI 0.82 to 1.26), had an increased risk of death, which persisted in an inverse probability of treatment weights-adjusted model (HR 1.56, 95% CI 1.21 to 1.93).ConclusionsPatients with grade 3 stage I endometrioid endometrial adenocarcinoma treated with primary definitive radiation therapy have worse survival than those with lower grade tumors. Addition of external beam radiation therapy to brachytherapy did not affect survival.

Purpose The aim of this study was to assess the impact of air gaps at the cylinder surface on the rate of vaginal cuff failure (VCF) after image-guided adjuvant vaginal cuff brachytherapy (VCBT) in the treatment of high-intermediate risk (HIR) FIGO (Federation Internationale de Gynecologie et d'Obstetrique (International Federation of Gynecology and Obstetrics)) Stage I endometrial cancer. Methods and Materials A retrospective review of patients treated with image-guided VCBT from 2009 to 2016 for HIR FIGO Stage I endometrial cancer was performed. Air gaps present at the applicator surface on the first postinsertion CT were contoured. Vaginal cuff failure–free survival (VCFFS) was measured from the first fraction of VCBT to VCF. Results A total of 234 patients were identified. Air gaps were present on the first postinsertion CT scan in 82% of patients. The median number of air gaps was 2 (interquartile range [IQR] 1–3), median depth of the largest air gap was 2.7 mm (IQR 2.1–3.4 mm), and the median cumulative volume of air gaps was less than 0.1 cm3 (range Conclusions In this population, VCFFS remained high despite most patients having air gaps present on postinsertion CT scan.

Purpose GOG 205 safely increased clinical (cCR) and pathologic complete response (pCR) in locally-advanced vulvar cancer through dose escalation using three-dimensional radiotherapy (RT). The aim of this study is to assess the response of dose-escalated intensity modulated radiotherapy (IMRT) in locally-advanced vulvar cancer. Methods A retrospective review of patients treated with dose-escalated (≥ 55Gy) IMRT from 2012 to 2018 for locally-advanced vulvar cancer was performed. Patients treated with preoperative or definitive intent were included. Rates of cCR and pCR were assessed, and predictors of disease-free survival (DFS) were analyzed using the Kaplan Meier method with log rank test between groups and a parsimonious multivariate Cox model. Results Median dose to the vulva was 66.0 Gy (Interquartile Range [IQR]: 66.0–68.0) for definitive and 59.4 Gy (IQR: 58.0–59.4) for preoperative IMRT. The overall rates of cCR and pCR were 76% and 70%, respectively. DFS at two years was 65% (95% Confidence Interval [CI] 50–80%) for all patients, 81% (95% CI 63% - 98%) for definitive IMRT, and 55% (95% CI 35% - 76%) for preoperative IMRT. On multivariate analysis, cCR predicted for disease-free survival (HR 0.21; 95% CI 0.06–0.76; p = 0.02), and pCR predicted for OS (HR 0.12; 95% CI 0.02–0.60; p = 0.01). Grade 3 acute and late RT toxicity was seen in 14 (29%) and 3 (6%) of patients, respectively. Conclusion Dose-escalated IMRT for locally-advanced vulvar cancer is well tolerated, with rates of cCR and pCR that compare favorably with published data.

PURPOSE The Vienna and Venezia (Elekta) are hybrid intracavitary/interstitial brachytherapy (BT) applicators for cervical cancers unsuitable for intracavitary BT alone to improve target coverage or reduce critical organ dose. There is limited outcome data with the use of these applicators outside published experience of the EMBRACE group. We report feasibility and early outcomes with the use of these hybrid applicators at our institution. METHODS AND MATERIALS Hybrid applicators were used to treat 61 patients with cervical cancer from November 2011 to December 2019. Indications for hybrid applicator use were involvement of the vagina in 10 patients (16%), residual central or parametrial disease in 46 patients (75%), and a narrow introitus in 5 patients (9%). Toxicities were graded using the CTCAE v4.0. Outcomes were assessed with the Kaplan–Meier method. Results Median follow-up was 16 months (IQR 9–32 mos). Median HRCTV volume was 31.6 cm3 (IQR 25–48 cm3). Median HRCTV D90 was 86.1 Gy (IQR 84.3–88.0 Gy). In 54 patients with follow-up PET/CT at 3 months, complete initial imaging response locally was seen in 46 patients.Estimated 12-month Kaplan–Meier overall survival, locoregional control, distant control, and recurrence-free survival estimates were 86.9%, 80.6%, 73.8%, and 65.9%, respectively. The 12-month incidence of Grade 3+ GI/GU chronic toxicities was 5.7%, consisting of vesicovaginal fistula, rectovaginal fistula, and ureterovesical fistula. Conclusions Our single-institution data support the use of the hybrid applicators, as an alternative to traditional BT applicators when clinically warranted. Use of hybrid applicators is feasible with adequate coverage of disease in the vagina and parametrium.

To investigate changes in estrogen receptor alpha (ERα) signaling during progression of endometriosis to endometriosis-associated ovarian cancer (EAOC) as a driver of malignant transformation. We procured tissue samples of normal endometrium, endometriosis (benign, atypical, concurrent with EAOC), and EAOC. We evaluated expression of a 236-gene signature of estrogen signaling. ANOVA and unsupervised clustering were used to identify gene expression profiles across disease states. These profiles were compared to profiles of estrogen regulation in cancer models from the Gene Expression Omnibus (GEO). Gene Set Enrichment Analysis (GSEA) was performed to determine whether gene expression in EAOC was consistent with ERα activity. ANOVA revealed 158 differentially expressed genes (q

Objectives: Poly (ADP-ribose) polymerase inhibitors (PARPis) improve progression-free survival (PFS) in high-grade epithelial ovarian cancer (EOC) as frontline maintenance (FM), platinum sensitive maintenance (PSM) and active therapy (AT) of recurrence. Investigation of PARPi effectiveness within everyday clinical practice is necessary. Using real-world PARPi data we explored survival outcomes by PARPi indication (PI), toxicities and impact on next-line therapy. Methods: A retrospective review was conducted of EOC patients receiving PARPis from 3/2014-3/2020. Medical records were reviewed for clinicopathologic data, PI, treatment details, response, survival and toxicity. Statistics included descriptive, one-way ANOVA, Cox proportional hazard (PH) regression and Wilcoxon Gehan-Breslow survival analyses. Results: 175 patients were included. Mean age at diagnosis was 59.9 years (25.8-93). Most patients had high-grade serous histology (88%), advanced stage (III 62.9%; IV 23.4%) and platinum sensitive disease (84.6%). BRCA status included 54 (30.9%) germline, 5 (2.9%) somatic, 6 (3.4%) homologous recombination deficient and 88 (50.3%) negative. PI included FM (11.4%), PSM (41.7%) and AT (46.9%). PI predicted PFS on PARPi [median (months): FM=18.2, PSM=19.3, AT=6.3, p=0.0001]. On Cox PH regression, age (p=0.03), PI (p=0.001) and PARPi agent (p=0.003) predicted PFS. Of 123 patients with progression (FM=8, PSM=43, AT=72), 97 (78.9%) received next-line therapy including chemotherapy (73.2%), clinical trial (11.3%), VEGF inhibition (5.2%), immune therapy (5.2%), hormones (4.1%) and HIPEC (1.0%). Disease control rate on next-line was 43.3% (Figure 1). PI was not associated with PFS on next-line [median (months): FM=3.5, PSM=3.8, AT=4.4, p=0.48] or number of subsequent lines (mean, FM=1.8, PSM=2.3, AT=2.1, p=0.21). PI predicted overall survival (OS) [median (months): FM-not met, follow-up=27.6, PSM-not met, follow-up=65.7, AT=78.0, p Download : Download high-res image (70KB) Download : Download full-size image Conclusions: PARPi toxicities reflected clinical trial results. Across indications PARPi use did not limit administration or response to subsequent therapies, though patients receiving next-line cytotoxic had notable hematologic toxicity. PFS on PARPi and OS were associated with PI. Future research should aim to identify optimal sequencing of PARPi, and those most likely to benefit from retreatment.

Purpose Previous studies of stage II endometrial cancer have included cancers with cervical glandular involvement, a factor no longer associated with risk of recurrence. In order to better assess relapse patterns and the impact of adjuvant therapy, a retrospective analysis was conducted for patients with modern stage II endometrial cancer, defined as cervical stromal invasion. Materials and methods Patients diagnosed with surgically staged FIGO stage II endometrial cancer at the UPMC Hillman Cancer Center from 1990–2013 were reviewed. Factors associated with rates of locoregional control (LRC), distant metastasis (DM), disease-free survival (DFS), and overall survival (OS) were analyzed using the log rank test. Results 110 patients with FIGO stage II disease were identified. Most (84.5%) received EBRT±BT, with 13.6% receiving BT alone. With a median follow-up of 64.6months, the 5-year actuarial rates of LRC, DM, DFS, and OS were 94.9%, 85.1%, 67.9%, and 75.0%, respectively. With 5 locoregional failures, the only factor predictive of LRC was pelvic lymph node dissection. Characteristics associated with DM included age, LVSI, depth of myometrial invasion, and receipt of chemotherapy. Factors predictive of both DFS and OS were age, grade, adverse histology, LVSI, depth of myometrial invasion, and receipt of chemotherapy. Conclusions This represents the largest single-institution study for modern stage II endometrial cancer, confirming high rates of pelvic disease control after surgery and adjuvant therapy. With most patients receiving adjuvant radiotherapy, the predominant mode of failure, albeit low in absolute number, remains distant metastases.

PURPOSE: A needs assessment of family caregivers (CGs) in our gynecologic oncology clinic found that 50% of CGs report nine or more distressing unmet needs, but only 19% of patients had a documented CG. We conducted an ASCO Quality Training Program project with the following aims: (1) to identify and document primary CGs for 85% of patients within two clinic visits of a gynecologic cancer diagnosis, and (2) assess the needs of and provide interventions to 75% of identified family CGs. METHODS: Plan-Do-Study-Act (PDSA) methodology and tools endorsed by the ASCO Quality Training Program were used. An interprofessional team reviewed baseline data (ie, any mention of a family CG in the electronic health record visit note; CG distress survey), defined the problem and project aims, created process maps, and identified root causes of poor CG identification and documentation. Eight successive PDSA cycles were implemented between October 2018 and March 2019 to address identified root causes. RESULTS: For aim 1, CG identification increased from 19% at baseline to 57% postimplementation, whereas for aim 2, assessment improved from 28% at baseline to 60% postimplementation. Results fell somewhat short of initial goals, but they represent an important initial improvement in care. The core team has begun additional PDSA cycles to improve CG identification rates and extend the momentum of the project. CONCLUSION: This project demonstrated that a CG assessment protocol can be implemented in a large, academic, gynecologic oncology clinic. Additional efforts to integrate CG identification, assessment, and intervention more fully within the clinic and electronic health record are under way.

Invasive lobular carcinoma (ILC) is an understudied subtype of breast cancer that requires novel therapies in the advanced setting. To study acquired resistance to endocrine therapy in ILC, we have recently performed RNA-Sequencing on long-term estrogen deprived cell lines and identified FGFR4 overexpression as a top druggable target. Here, we show that FGFR4 expression also increases dramatically in endocrine-treated distant metastases, with an average fold change of 4.8 relative to the paired primary breast tumor for ILC, and 2.4-fold for invasive ductal carcinoma (IDC). In addition, we now report that FGFR4 hotspot mutations are enriched in metastatic breast cancer, with an additional enrichment for ILC, suggesting a multimodal selection of FGFR4 activation. These data collectively support the notion that FGFR4 is an important mediator of endocrine resistance in ILC, warranting future mechanistic studies on downstream signaling of overexpressed wild-type and mutant FGFR4.

e18838 Background: PORTEC-3, a phase III clinical trial comparing adjuvant radiation alone (RT) to combined chemoradiation therapy (CTRT) in patients with high-risk endometrial cancer (EC), demonstrated no benefit with CTRT compared to RT for patients with stage I and II EC. However, a subsequent tumor molecular analysis revealed that individuals with stage I and II high-risk EC with p53 mutations had statistically significantly improved progression-free survival (PFS) and clinically significantly improved overall survival (OS) with combined CTRT. As p53 immunohistochemical (IHC) testing and CTRT adds to cost and toxicity, we sought to evaluate the cost-effectiveness of p53 IHC tumor testing in patients with stage I and II high-risk EC. Methods: A Markov decision model was developed to compare two testing strategies in patients with stage I and II high-risk EC: p53 IHC testing vs. no IHC testing. IHC staining for p53 is an accurate and validated surrogate for TP53 mutational status. As a result, p53 IHC testing alone was used in the decision analysis. Data from PORTEC-3 and the subsequent molecular analysis were used for the base case model parameters, including treatment regimen, PFS, OS, and incidence of p53-mutated tumors. Cost and utility data were derived from the Federal Supply Schedule, the Centers for Medicare and Medicaid Services (CMS) Fee Schedules, the Tufts CEA registry, peer-reviewed literature, and expert input. Data analysis was performed using TreeAge Pro Software, 2021. Effectiveness outcomes included quality-adjusted life-years (QALYs) and incremental cost-effectiveness ratios (ICERs). Sensitivity analyses were performed in order to identify the influence of individual variables on the overall model. Results: When compared to no testing, p53 IHC testing resulted in a cost savings of $471 per patient while providing an additional 0.310 QALYs. Therefore, p53 IHC testing was less expensive and more effective than no IHC testing and was a dominant strategy in this model. In one-way sensitivity analyses, CTRT and RT costs, discount rate, and percentage of p53 mutations detected had the most impact on the model; however, p53 IHC testing remained a cost-saving strategy over all parameter ranges examined. Conclusions: For patients with high-risk stage I and II EC, this analysis suggests that p53 IHC testing is cost-effective compared to no testing in determining the costs and benefits of adjuvant chemotherapy. Although additional studies are warranted, this data provides further support for the role of molecular-based treatment decisions for high-risk stage I and II endometrial cancer.

Nausea is a common and potentially serious effect of cytotoxic chemotherapy for recurrent ovarian cancer and may function as a sentinel symptom reflecting adverse effects on the gut-brain axis (GBA) more generally, but research is scant. As a first exploratory test of this GBA hypothesis, we compared women reporting nausea to women not reporting nausea with regard to the severity of other commonly reported symptoms in this patient population.A secondary analysis of data systematically collected from women in active chemotherapy treatment for recurrent ovarian cancer (n = 158) was conducted. The Symptom Representation Questionnaire (SRQ) provided severity ratings for 22 common symptoms related to cancer and chemotherapy. Independent sample t tests and regression analyses were used to compare women with and without nausea with regard to their experience of other symptoms.Nausea was reported by 89 (56.2 %) women. Symptoms that were significantly associated with nausea in bivariate and regression analyses included abdominal bloating, bowel disturbances, dizziness, depression, drowsiness, fatigue, headache, lack of appetite, memory problems, mood swings, shortness of breath, pain, sleep disturbance, urinary problems, vomiting, and weight loss. Symptoms that were not associated with nausea included hair loss, numbness and tingling, sexuality concerns, and weight gain.Nausea experienced during chemotherapy for recurrent ovarian cancer may be an indicator of broader effects on the gut-brain axis. A better understanding of the mechanisms underlying these effects could lead to the development of novel supportive therapies to increase the tolerability and effectiveness of cancer treatment.

Endometrial and ovarian cancers are estrogen-dependent gynecologic malignancies. Although many are estrogen receptor (ER) positive, treatment with the selective estrogen receptor modulator (SERM) tamoxifen, a tissue selective partial-agonist, has demonstrated only modest clinical benefit. Selective estrogen receptor downregulators (SERDs) are pure ER antagonists showing a benefit for advanced ER positive breast cancer, which has bolstered their potential use for ER positive gynecologic malignancies. We summarize these preclinical and clinical data, suggesting that a subpopulation of patients with endometrial or ovarian cancer exists in which treatment with SERDs results in improved outcome. However, the full potential of SERDs for a gynecologic malignancies will be realized only when the appropriate predictive biomarkers are identified. Additionally, a further understanding ER signaling in the context of ovarian and endometrial tissues that appear to involve c-Src and other kinase pathways is needed to successfully address the emergence of resistance with rationally designed combination therapies.

DNA sequencing has identified a limited number of driver mutations in metastatic breast cancer beyond single base-pair mutations in the estrogen receptor (ESR1). However, our previous studies and others have observed that structural variants, such as ESR1 fusions, may also play a role. Therefore, we expanded upon these observations by performing a comprehensive and highly sensitive characterization of copy-number (CN) alterations in a large clinical cohort of metastatic specimens. NanoString DNA hybridization was utilized to measure CN gains, amplifications, and deletions of 67 genes in 108 breast cancer metastases, and in 26 cases, the patient-matched primary tumor. For ESR1, a copyshift algorithm was applied to identify CN imbalances at exon-specific resolution and queried large data sets (>15,000 tumors) that had previously undergone next-generation sequencing (NGS). Interestingly, a subset of ER+ tumors showed increased ESR1 CN (11/82, 13%); three had CN amplifications (4%) and eight had gains (10%). Increased ESR1 CN was enriched in metastatic specimens versus primary tumors, and this was orthogonally confirmed in a large NGS data set. ESR1-amplified tumors showed a site-specific enrichment for bone metastases and worse outcomes than nonamplified tumors. No ESR1 CN amplifications and only one gain was identified in ER− tumors. ESR1 copyshift was present in 5 of the 11 ESR1-amplified tumors. Other frequent amplifications included ERBB2, GRB7, and cell-cycle pathway members CCND1 and CDK4/6, which showed mutually exclusivity with deletions of CDKN2A, CDKN2B, and CDKN1B. Implications: Copy-number alterations of ESR1 and key CDK pathway genes are frequent in metastatic breast cancers, and their clinical relevance should be tested further.

Purpose Many patients with endometrial cancer cannot undergo surgery and instead receive definitive radiation therapy (RT). We investigate the correlation between MRI response to RT and clinical outcomes. Methods and Materials Women with inoperable, clinical Stage I endometrial cancer were treated with definitive brachytherapy (BT) with/without pelvic RT (PRT). Patients underwent MRI with functional diffusion-weighted imaging before and after RT. A radiologist retrospectively classified cases as complete, partial, or indeterminate response (CR, PR, or IR, respectively) vs. disease progression. Local control was clinicopathologically defined. Results From 2007 to 2017, 50 women underwent definitive RT. Thirty-five (70%) received BT alone (median dose 37.5 Gy). For combined therapy, the median PRT and BT doses were 45 and 25 Gy, respectively. Median gross tumor volume and high-risk clinical target volume were 7.1 cc and 90.0 cc, respectively. Median followup among living patients was 20 months. All patients underwent post-RT MRI with T1/T2 sequencing at a median of 3.2 months after RT; 40 patients (80%) underwent functional diffusion-weighted imaging sequences. On initial post-RT MRI, CR was documented in 42 patients (84%), IR in 1 patient (2%), and PR in seven patients (14%). At median followup of 16.3 months, no CR patients had uterine failure. Among eight patients with initial PR/IR, all were found to be clinicopathologically no evidence of disease at the uterus on further evaluation. Conclusions Definitive RT with BT or BT + PRT is associated with high response rates on MRI. Overall, initial CR predicted for excellent outcome with no infield failure.

Objectives Management of endometrial cancer consists of surgical staging with adjuvant therapy guided by risk factors, though some women cannot undergo surgery due to comorbidities. We present a series of women treated with definitive high-dose rate image-guided tandem and cylinder brachytherapy (HDR-IGBT) alone. Methods Patients with grade 1–2, clinical stage I endometrial adenocarcinoma, Results From 2007 to 2016, 45 patients were treated to a median dose of 37.5Gy. The median gross tumor volume (GTV) and clinical target volume (CTV) were 5.9cm 3 (range, 0.7–18.7) and 80.9cm 3 (17.2–159.0), respectively. The median cumulative dose to 90% (D90) of the GTV was 132.8Gy (76.5–295.6) equivalent 2Gy dose, and the median CTV D90 was 49.7Gy (34.5–57.2). Median follow-up among living patients was 18.6months (3.0–64.3). Cessation of vaginal bleeding occurred in 98%. Among those with post-treatment MRI (64%), complete radiographic response was demonstrated in 90%. The 2-year LRC, cancer-specific survival, and overall survival rates were 90%, 86%, and 97%, respectively. No grade 3+ acute or late toxicity was observed. Conclusions HDR-IGBT alone for treatment of early-stage, medically inoperable endometrial cancer is feasible with excellent response rates and clinical results. This approach also allows sparing of critical organs and ensures target coverage, which contributed to the low toxicity rate and high LRC in comparison with 2D point-based series.

Purpose: High-grade serous ovarian cancer (HGSOC) is an aggressive disease with few available targeted therapies. Despite high expression of estrogen receptor-alpha (ERα) in approximately 80% of HGSOC and some small but promising clinical trials of endocrine therapy, ERα has been understudied as a target in this disease. We sought to identify hormone-responsive, ERα-dependent HGSOC. Experimental Design: We characterized endocrine response in HGSOC cells across culture conditions [ two-dimensional (2D), three-dimensional (3D), forced suspension] and in patient-derived xenograft (PDX) explants, assessing proliferation and gene expression. Estrogen-regulated transcriptome data were overlapped with public datasets to develop a comprehensive panel of ERα target genes. Expression of this panel and ERα H-score were assessed in HGSOC samples from patients who received endocrine therapy. Time on endocrine therapy was used as a surrogate for clinical response. Results: Proliferation is ERα-regulated in HGSOC cells in vitro and in vivo, and is partly dependent on 3D context. Transcriptomic studies identified genes shared by cell lines and PDX explants as ERα targets. The selective ERα downregulator (SERD) fulvestrant is more effective than tamoxifen in blocking ERα action. ERα H-score is predictive of efficacy of endocrine therapy, and this prediction is further improved by inclusion of target gene expression, particularly IGFBP3. Conclusions: Laboratory models corroborate intertumor heterogeneity of endocrine response in HGSOC but identify features associated with functional ERα and endocrine responsiveness. Assessing ERα function (e.g., IGFBP3 expression) in conjunction with H-score may help select patients who would benefit from endocrine therapy. Preclinical data suggest that SERDs might be more effective than tamoxifen. Clin Cancer Res; 23(14); 3802–12. ©2017 AACR.

Purpose For locally-advanced uterine cancer clinically extending to the cervix, two treatment paradigms exist: surgical staging radical hysterectomy with tailored adjuvant therapy or neoadjuvant therapy followed by a less extensive simple hysterectomy. Currently, insufficient data exists to guide consensus guidelines and practical application of preoperative radiotherapy. Materials and methods Retrospective IRB approved cohort study from 1999 to 2014 of 36 endometrial cancer patients with clinical involvement of cervix±parametria treated with neoadjuvant external beam radiotherapy (45–50.4Gy in 25–28 fractions) and image-based HDR brachytherapy (5–5.5Gy times 3–4 fractions)±chemotherapy followed by extrafascial hysterectomy performed at a median of 6weeks after radiotherapy. Results All patients had clinical cervical extension, 50% also had parametria extension, and 31% had nodal involvement. At the time of surgery 91% had no clinical cervical involvement, 58% had no pathologic cervical involvement, and all had margin negative resection. The pathologic complete response rate was 24%. Median follow-up from the time of surgery was 20months (range: 0–153). The 3-year local control, regional control, distant control, disease free survival and overall survival rates were 96%, 89%, 84%, 73%, and 100%. The 3-year rate of grade 3 complications was 11%, with no grade 4+ toxicity. Conclusions Neoadjuvant radiation therapy±chemotherapy followed by extrafascial hysterectomy appears to be a viable option for patients with endometrial cancer clinically extending to the cervix and parametria. The HDR brachytherapy schema of 5–5.5Gy times 3–4 fractions, for a cumulative EQD 2 of 60–70Gy, is well tolerated with high rates of clinical and pathological response.

178 Background: Family caregiver (CG) demands in gynecologic oncology (gynonc) put CGs at risk for anxiety, depression, and inability to care for loved ones. Resources exist to support patients, but systematic support for CGs is lacking. Fifty percent of CGs in our clinic report 9 or more distressing unmet needs. Thus, an ASCO Quality Training Program (QTP) project aimed to assess and intervene on the needs of 75% of gynonc family CGs (identified in a concurrent QTP project). Methods: An interprofessional team reviewed baseline data, defined the problem and project aim, created process maps, and identified root causes of poor CG documentation. With stakeholder buy-in we implemented eight successive PDSA cycles over 6 months to address root causes. Biweekly team meetings were throughout the PDSA cycles. Outcomes were percent of CGs with 1) needs assessed, and 2) intervention received. Results: Root causes of poor CG assessment included 1) no protocol to assess/respond to CG needs; 2) no designated staff to ‘own’ CG assessment; and 3) lack of understanding of the value of CG support among clinicians. PDSA 1 (pre- launch) included: develop CG assessment/intervention protocol; create CG tracking database; pursue EHR capabilities. PDSA cycle 2-3 prioritized information provision about resources (52-73% of CGs) over individualized assessment (20-23% of CGs) due to the large number of CGs identified. Subsequent PDSAs sought to increase assessment effectiveness by focusing assessment/intervention on CGs of newly diagnosed and high risk patients. While there was variability in outcomes across cycles (Table), by PDSA cycle 8, staff had assessed and intervened on 58% and 50% of CGs, respectively. In addition, multiple potential serious CG and patient mental health crises were avoided from team interventions. Conclusions: CG assessment improved from 28% at baseline to >50% post implementation. Further efforts to more fully integrate assessment and tracking into EHR to reduce the burden of CG tracking over time is underway. [Table: see text]

ObjectiveOnly 3% of patients with epithelial ovarian cancer (EOC) have a longer treatment-free interval (TFI) after second-line intravenous (IV) platinum chemotherapy than with frontline IV therapy. We sought to examine what impact second-line combination IV/intraperitoneal (IV/IP) platinum therapy might have on the ratio of second-line to first-line TFI in patients treated with second-line IP platinum chemotherapy for first recurrence after front-line IV therapy.MethodsA retrospective analysis of women who received combination platinum-based IV/IP chemotherapy for recurrent EOC between January 2005 and March 2011 was conducted. Patients were identified from the tumor registry, and office records from a large gynecologic oncology practice and patient records were reviewed. The first and second TFIs were defined as the time from the end of previous platinum-based therapy to the start of next therapy.ResultsTwenty-five women received IV/IP chemotherapy for their first EOC recurrence after IV chemotherapy. In 10 patients (40%), we observed a longer TFI after IV/IP chemotherapy than after primary IV chemotherapy. For these 10 patients, the median TFI for primary response was 22 months (range, 15–28), whereas median TFI after IV/IP chemotherapy for recurrent disease was 37 months (range, 12–61).ConclusionsFor EOC patients with limited peritoneal recurrence, 40% of patients had a second-line IP-platinum TFI that exceeded their frontline IV-platinum TFI compared to published data. These data support the use of IV/IP chemotherapy as a treatment for recurrence.

Hyperthermic intraperitoneal chemotherapy (HIPEC) is often used to treat gastrointestinal malignancies and is of interest in epithelial ovarian cancer (EOC) given the propensity for intraperitoneal spread. The role of HIPEC in the treatment of gynecologic malignancies is not well defined. We sought to describe clinical characteristics and outcomes of our patient population treated with HIPEC.IRB approval was obtained. Patients diagnosed with EOC and treated with HIPEC from January 2007 until December 2013 were identified using a prospectively maintained HIPEC database. Patient charts were abstracted to identify patient demographic information, treatment characteristics, and outcome data. Statistical analysis was descriptive.Thirty-four patients were identified. Mean age at diagnosis was 56.5 years. The majority of cases (28, 82%) were of serous histology. The indications for HIPEC administration were as follows: 9% primary treatment, 41% first recurrence, 26% second recurrence, and 24% consolidative therapy in the setting of primary or recurrent disease. The majority of patients (21, 62%) received mitomycin C. The other drugs administered include cisplatin (10, 29%), oxaliplatin (2, 6%), and carboplatin (1, 3%). Mean length of hospital stay was 9 days (range, 3-39 days). The rates of postoperative bacteremia and hematologic toxicity were 6% and 54%, respectively. Seven (21%) patients developed transient renal dysfunction, and this was seen almost exclusively in the patients who received cisplatin. One (3%) additional patient had renal dysfunction that persisted longer than 30 days post-operative but did not go on to require dialysis. There were no perioperative deaths in this cohort. Eleven (32%) patients received additional chemotherapy following HIPEC administration. At a median follow-up of 20 months (range, 3-87 months), eight patients are alive with disease, seven have no evidence of disease, 14 have died of their disease, and five patients have been lost to follow-up.This data supports a reasonable side effect profile of treatment of EOC with HIPEC. Prospective studies are needed to elucidate the optimal drug and patient population that would derive the most benefit from treatment with HIPEC.

ObjectivesRecent data have shown high rates of clinical and pathologic responses to neoadjuvant radiation therapy for locally advanced endometrial cancer. There are limited data on the surgical outcomes of these patients in the era of modern radiation and surgical techniques. We sought to characterize surgical outcomes after extrafascial hysterectomy in this population.MethodsPatients with endometrial cancer of all histologies clinically involving the cervix or parametria treated with neoadjuvant brachytherapy followed by extrafascial hysterectomy from 1999 to 2014 were identified. Patient charts were reviewed for data regarding treatment characteristics and postoperative outcomes. Pearson χ2 and logistic regression analyses were used to assess correlations between surgical complications and treatment-related variables.ResultsTwenty-nine patients met inclusion criteria. Mean operating time for the cohort was 115 minutes. Mean estimated blood loss was 100 mL. No visceral injuries occurred. Mean length of hospital stay was 1 and 4 days for the minimally invasive and laparotomy groups, respectively. Rates of postoperative ileus, blood transfusion, wound infection, and readmission were 3%, 3%, 6%, and 3%, respectively. No case of prolonged urodynamic dysfunction was noted. The rate of vaginal complications was significantly higher in the group of patients who underwent minimally invasive surgery as compared with laparotomy (33% vs 5%, P < 0.041).ConclusionsThese data support adjuvant extrafascial hysterectomy after neoadjuvant radiotherapy for endometrial cancer with cervical or parametrial involvement as a safe and viable procedure, with low rates of postoperative complications. Extra care should be taken when closing the vaginal cuff to reduce the risk of vaginal cuff complications.

Objectives: To investigate changes in estrogen receptor–alpha (ERα) signaling during the progression of endometriosis to endometriosis-associated ovarian cancer (EAOC) as a putative driver of malignant transformation. Methods: We procured tissue samples of normal endometrium (n=23), endometriosis (benign, n=19; atypical, n=11; concurrent with EAOC, n=9), and EAOC (n=21). In this cohort, we evaluated expression of a 236-gene signature of estrogen signaling. ANOVA and unsupervised clustering were used to identify distinct gene expression profiles across disease states. These profiles were compared to public profiles of estrogen regulation in preclinical cancer models from the Gene Expression Omnibus (GEO). Gene set enrichment analysis (GSEA) was performed to determine whether gene expression in EAOC was consistent with ERα activity. Results: ANOVA revealed 158 differentially expressed genes (q Conclusions: Gene expression data suggest that classical ERα signaling becomes inactivated throughout the progression of endometriosis to EAOC. Rather, the gene expression pattern in EAOC is more consistent with profiles of endocrine resistance. De-repression of ERα target genes such as FGF18 may contribute to evolution of endometriosis into EAOC. Citation Format: Michelle Boisen, Courtney Andersen, Matt Sikora, Tianzhou Ma, George Tseng, Anda Vlad, Esther Elishaev, Uma Chandran, Robert Edwards, Steffi Oesterreich. The evolution of estrogen receptor signaling in the progression of endometriosis to endometriosis-associated ovarian cancer. [abstract]. In: Proceedings of the AACR Conference: Addressing Critical Questions in Ovarian Cancer Research and Treatment; Oct 1-4, 2017; Pittsburgh, PA. Philadelphia (PA): AACR; Clin Cancer Res 2018;24(15_Suppl):Abstract nr B23.