Search

PURPOSE Platinum-based chemotherapy is the standard of care for platinum-sensitive ovarian cancer, but complications from repeated platinum therapy occur. We assessed the activity of two all-oral nonplatinum alternatives, olaparib or olaparib/cediranib, versus platinum-based chemotherapy. PATIENTS AND METHODS NRG-GY004 is an open-label, randomized, phase III trial conducted in the United States and Canada. Eligible patients had high-grade serous or endometrioid platinum-sensitive ovarian cancer. Patients were randomly assigned 1:1:1 to platinum-based chemotherapy, olaparib, or olaparib/cediranib. The primary end point was progression-free survival (PFS) in the intention-to-treat population. Secondary end points included activity within germline BRCA-mutated or wild-type subgroups and patient-reported outcomes (PROs). RESULTS Between February 04, 2016, and November 13, 2017, 565 eligible patients were randomly assigned. Median PFS was 10.3 (95% CI, 8.7 to 11.2), 8.2 (95% CI, 6.6 to 8.7), and 10.4 (95% CI, 8.5 to 12.5) months with chemotherapy, olaparib, and olaparib/cediranib, respectively. Olaparib/cediranib did not improve PFS versus chemotherapy (hazard ratio [HR] 0.86; 95% CI, 0.66 to 1.10; P = .077). In women with germline BRCA mutation, the PFS HR versus chemotherapy was 0.55 (95% CI, 0.32 to 0.94) for olaparib/cediranib and 0.63 (95% CI, 0.37 to 1.07) for olaparib. In women without a germline BRCA mutation, the PFS HR versus chemotherapy was 0.97 (95% CI, 0.73 to 1.30) for olaparib/cediranib and 1.41 (95% CI, 1.07 to 1.86) for olaparib. Hematologic adverse events occurred more commonly with chemotherapy; however, nonhematologic adverse events were higher with olaparib/cediranib. In 489 patients evaluable for PROs, patients receiving olaparib/cediranib scored on average 1.1 points worse on the NFOSI-DRS-P subscale (97.5% CI, –2.0 to –0.2, P = .0063) versus chemotherapy; no difference between olaparib and chemotherapy was observed. CONCLUSION Combination olaparib/cediranib did not improve PFS compared with chemotherapy and resulted in reduced PROs. Notably, in patients with a germline BRCA mutation, both olaparib and olaparib/cediranib had significant clinical activity.

Figure S6 shows the association of circulating cytokine levels with objective clinical response.

Purpose:A phase I clinical trial (GOG-9929) examined the safety and efficacy of adjuvant immune-modulation therapy with the checkpoint inhibitor ipilimumab [anti–CTL antigen-4 (anti–CTLA-4)] following chemoradiation therapy (CRT) for newly diagnosed node-positive human papillomavirus (HPV)-related cervical cancer. To better understand the mechanism of action and to identify predictive biomarkers, immunologic and viral correlates were assessed before, during, and after treatment.Patients and Methods:Twenty-one patients who received CRT and ≥2 doses of ipilimumab and 5 patients who received CRT only were evaluable for translational endpoints. Circulating T-cell subsets were evaluated by multiparameter flow cytometry. Cytokines were evaluated by multiplex ELISA. HPV-specific T cells were evaluated in a subset of patients by IFNγ ELISpot.Results:Expression of the activation markers ICOS and PD-1 significantly increased on T-cell subsets following CRT and were sustained or increased following ipilimumab treatment. Combined CRT/ipilimumab treatment resulted in a significant expansion of both central and effector memory T-cell populations. Genotype-specific E6/E7-specific T-cell responses increased post-CRT in 1 of 8 HPV16+ patients and in 2 of 3 HPV18+ patients. Elevation in levels of tumor-promoting circulating cytokines (TNFα, IL6, IL8) post-CRT was significantly associated with worse progression-free survival.Conclusions:Our data indicate that CRT alone and combined with ipilimumab immunotherapy show immune-modulating activity in women with locally advanced cervical cancer and may be a promising therapeutic option for the enhancement of antitumor immune cell function after primary CRT for this population at high risk for recurrence and metastasis. Several key immune biomarkers were identified that were associated with clinical response.

Background Successfully combining targeted agents with chemotherapy is an important future goal for cancer therapy. However, an improvement in patient outcomes requires an enhanced understanding of the tumor biomarkers that predict for drug sensitivity. NRG Oncology/Gynecologic Oncology Group (GOG) Study GOG-86P was one of the first attempts to combine targeted agents (bevacizumab or temsirolimus) with chemotherapy in patients with advanced endometrial cancer. Herein we performed exploratory analyses to examine the relationship between mutations in TP53, the most commonly mutated gene in cancer, with outcomes on GOG-86P. Methods TP53 mutational status was determined and correlated with progression-free survival (PFS) and overall survival (OS) on GOG-86P. Results Mutations in TP53 were associated with improved PFS and OS for patients that received bevacizumab as compared to temsirolimus (PFS: HR 0.48, 95% CI 0.31, 0.75; OS: HR: 0.61, 95% CI 0.38, 0.98). By contrast, there was no statistically significant difference in PFS or OS between arms for cases with WT TP53. Conclusions This exploratory study suggests that combining chemotherapy with bevacizumab, but not temsirolimus, may enhance PFS and OS for patients whose tumors harbor mutant p53. These data set the stage for larger clinical studies evaluating the potential of TP53 mutational status as a biomarker to guide choice of treatment for endometrial cancer patients. Clintrials.gov : NCT00977574 .

PURPOSE: Sparing active bone marrow (ABM) can reduce acute hematologic toxicity in patients undergoing chemoradiotherapy for cervical cancer, but ABM segmentation based on positron emission tomography/computed tomography (PET/CT) is costly. We sought to develop an atlas-based ABM segmentation method for implementation in a prospective clinical trial. METHODS AND MATERIALS: A multiatlas was built on a training set of 144 patients and validated in 32 patients from the NRG-GY006 clinical trial. ABM for individual patients was defined as the subvolume of pelvic bone greater than the individual mean standardized uptake value on registered (18)F-fluorodeoxyglucose PET/CT images. Atlas-based and custom ABM segmentations were compared using the Dice similarity coefficient and mean distance to agreement and used to generate ABM-sparing intensity modulated radiation therapy plans. Dose-volume metrics and normal tissue complication probabilities of the two approaches were compared using linear regression. RESULTS: Atlas-based ABM volumes (mean [standard deviation], 548.4 [88.3] cm(3)) were slightly larger than custom ABM volumes (535.1 [93.2] cm(3)), with a Dice similarity coefficient of 0.73. Total pelvic bone marrow V(20) and D(mean) were systematically higher and custom ABM V(10) was systematically lower with custom-based plans (slope: 1.021 [95% confidence interval (CI), 1.005-1.037], 1.014 [95% CI, 1.006-1.022], and 0.98 [95% CI, 0.97-0.99], respectively). We found no significant differences between atlas-based and custom-based plans in bowel, rectum, bladder, femoral heads, or target dose-volume metrics. CONCLUSIONS: Atlas-based ABM segmentation can reduce pelvic bone marrow dose while achieving comparable target and other normal tissue dosimetry. This approach may allow ABM sparing in settings where PET/CT is unavailable.

Purpose: A phase I clinical trial (GOG-9929) examined the safety and efficacy of adjuvant immune-modulation therapy with the checkpoint inhibitor ipilimumab [anti–CTL antigen-4 (anti–CTLA-4)] following chemoradiation therapy (CRT) for newly diagnosed node-positive human papillomavirus (HPV)-related cervical cancer. To better understand the mechanism of action and to identify predictive biomarkers, immunologic and viral correlates were assessed before, during, and after treatment. Patients and Methods: Twenty-one patients who received CRT and ≥2 doses of ipilimumab and 5 patients who received CRT only were evaluable for translational endpoints. Circulating T-cell subsets were evaluated by multiparameter flow cytometry. Cytokines were evaluated by multiplex ELISA. HPV-specific T cells were evaluated in a subset of patients by IFNγ ELISpot. Results: Expression of the activation markers ICOS and PD-1 significantly increased on T-cell subsets following CRT and were sustained or increased following ipilimumab treatment. Combined CRT/ipilimumab treatment resulted in a significant expansion of both central and effector memory T-cell populations. Genotype-specific E6/E7-specific T-cell responses increased post-CRT in 1 of 8 HPV16+ patients and in 2 of 3 HPV18+ patients. Elevation in levels of tumor-promoting circulating cytokines (TNFα, IL6, IL8) post-CRT was significantly associated with worse progression-free survival. Conclusions: Our data indicate that CRT alone and combined with ipilimumab immunotherapy show immune-modulating activity in women with locally advanced cervical cancer and may be a promising therapeutic option for the enhancement of antitumor immune cell function after primary CRT for this population at high risk for recurrence and metastasis. Several key immune biomarkers were identified that were associated with clinical response.

Background Immune checkpoint inhibitors are being considered for locally advanced cervical cancer (LACC) together with standard-of-care pelvic chemoradiation (CRT). However, the safety of the combination and its optimal schedule are unknown. Defining the safety of the combination is a primary objective of a study examining concurrent and sequential schedules. This article presents a safety analysis that was fully accrued and met reporting requirements. Methods Pembrolizumab was given after CRT (arm 1) or during CRT (arm 2) according to a randomized phase 2 design. Patients who were 18 years old or older and had LACC (stages IB-IVA according to the 2009 International Federation of Gynecology and Obstetrics system) were randomized 1:1 to the treatment regimens. The CRT was identical in the 2 arms. Pembrolizumab was administered every 3 weeks for 3 doses; no maintenance was allowed. All patients receiving any treatment were evaluated for safety. Safety assessments included the incidence and severity of adverse events (AEs) and the occurrence of protocol-defined dose-limiting toxicity (DLT) through 30 days after the last pembrolizumab infusion. Results As of August 2019, 52 of the 88 planned patients had completed treatment and were evaluable for toxicity. Treatment-related grade 2 or higher toxicity was experienced by 88%; 11 had at least 1 grade 4 AE, and another 23 had at least 1 grade 3 AE. Grade 1 or higher diarrhea was reported in 34 patients (65%; 50% of these were grade 1), and there was no difference between arms (63% in arm 1 vs 68% in arm 2). Two patients experienced 3 DLTs. Most patients completed cisplatin (100% in arm 1 vs 82% in arm 2); 83% in both arms completed all pembrolizumab. Conclusions Preliminary results support the safety and feasibility of adding pembrolizumab to pelvic CRT concurrently or sequentially. Lay summary Pembrolizumab is a humanized antibody against programmed cell death protein 1 that is used in cancer immunotherapy. Preliminary data suggest that pembrolizumab can be safely combined with chemotherapy and pelvic radiation in the treatment of locally advanced cervical cancer. Future studies of the addition of immunotherapy to traditional chemoradiation are planned to determine the best way to deliver the treatment and whether any improvement is seen with the addition of immunotherapy to traditional therapy.

The sfRon kinase is an important therapeutic target in ovarian cancer that contributes to prominent tumor growth and disease progression. We reasoned that a multi-kinase inhibition of sfRon pathway might be an effective strategy to achieve a sustained anti-tumor response, while simultaneously preventing treatment resistance. We performed a detailed dissection of sfRon signaling in vitro and demonstrated that S6K1 is a key component of a multi-kinase targeting strategy in sfRon expressing ovarian tumors. We selected AD80 compound that targets several kinases within sfRon pathway including AKT and S6K1, and compared its efficacy with inhibitors that selectively target either sfRon or PI3 kinase. Using human ovarian xenografts and clinically relevant patient-derived xenografts (PDXs), we demonstrated that in vivo treatment with single agent AD80 shows superior efficacy to a standard-care chemotherapy (cisplatin/paclitaxel), or to the direct inhibition of sfRon kinase by BMS777607. Our findings indicate that ovarian tumors expressing sfRon are most effectively treated with multi-kinase inhibitors simultaneously targeting AKT and S6K1, such as AD80, which results in long-term anti-tumor response and prevents metastasis development.

INTRODUCTION: Chemotherapy plus radiation (Cis-RT+CP) did not demonstrate superiority in prolonging relapse-free survival compared to chemotherapy alone in patients with stage III or IVA endometrial carcinoma. The impact of treatment on quality of life (QOL), neurotoxicity (NTX) and psychometric properties of the gastrointestinal (GI) symptoms subscale during treatment and up to 1 year are described herein. METHODS: QOL assessments were scheduled at baseline, 6 weeks (post completion of RT (Cis-RT+CP) or prior to cycle 3 (CP)), then 18 weeks (end of treatment) and 70 weeks (1 year after the end of treatment) after starting treatment. QOL instruments included the FACT-En TOI, FACT/GOG-neurotoxicity (Ntx) subscale (short), and the gastrointestinal (GI) symptoms subscale. RESULTS: At the end of treatment, patients receiving Cis-RT+CP reported a statistically significant decreased QOL when compared to CP. The decline in QOL was reflected in physical well-being, functional well-being, and endometrial cancer specific concerns, but the minimally important differences (MID) were not considered clinically meaningful. Patients in both groups reported increased chemotherapy-induced Ntx symptoms with the CP group having worse scores and reaching peak symptoms at the time of chemotherapy completion. Patients on Cis-RT+CP reported statistically significantly worse GI symptoms after radiation therapy compared to patients on CP, this occurred across assessment intervals, though the MID was not meaningful. Psychometric evaluations indicated that the GI symptom scale is reliable, valid, and responsive to change. CONCLUSIONS: PROs indicate that the chemoradiotherapy group experienced worse HRQoL and GI toxicity compared to patients randomized to chemotherapy alone for locally advanced endometrial cancer though based on the MID, these were not clinically meaningful differences. The GI symptom subscale was a reliable and valid scale that has value for future trials.

BACKGROUND: Stage III or IVA endometrial cancer carries a significant risk of systemic and locoregional recurrence. METHODS: In this randomized phase 3 trial, we tested whether 6 months of platinum-based chemotherapy plus radiation therapy (chemoradiotherapy) is associated with longer relapse-free survival (primary end point) than six cycles of combination chemotherapy alone in patients with stage III or IVA endometrial carcinoma. Secondary end points included overall survival, acute and chronic toxic effects, and quality of life. RESULTS: Of the 813 patients enrolled, 736 were eligible and were included in the analysis of relapse-free survival; of those patients, 707 received the randomly assigned intervention (346 received chemoradiotherapy and 361 received chemotherapy only). The median follow-up period was 47 months. At 60 months, the Kaplan–Meier estimate of the percentage of patients alive and relapse-free was 59% (95% confidence interval [CI], 53 to 65) in the chemoradiotherapy group and 58% (95% CI, 53 to 64) in the chemotherapy-only group (hazard ratio, 0.90; 90% CI, 0.74 to 1.10). Chemoradiotherapy was associated with a lower 5-year incidence of vaginal recurrence (2% vs. 7%; hazard ratio, 0.36; 95% CI, 0.16 to 0.82) and pelvic and paraaortic lymph-node recurrence (11% vs. 20%; hazard ratio, 0.43; 95% CI, 0.28 to 0.66) than chemotherapy alone, but distant recurrence was more common in association with chemoradiotherapy (27% vs. 21%; hazard ratio, 1.36; 95% CI, 1.00 to 1.86). Grade 3, 4, or 5 adverse events were reported in 202 patients (58%) in the chemoradiotherapy group and 227 patients (63%) in the chemotherapy-only group. CONCLUSIONS: Chemotherapy plus radiation was not associated with longer relapse-free survival than chemotherapy alone in patients with stage III or IVA endometrial carcinoma. (Funded by the National Cancer Institute; ClinicalTrials.gov number, NCT00942357.)

Cancer-associated fibroblasts (CAFs) play a critical role in cancer progression, metastasis, and therapy resistance. Molecular events that confer CAF-phenotype to predecessor-cells are not fully understood. We demonstrate here that the ovarian cancer cell-conditioned medium (OCC-CM) induces CAF-phenotype in MRC5 lung-fibroblasts and it can be mimicked by LPA. While OCC-CM and LPA stimulated the expression of cellular CAF-markers by 3-days, they induced aerobic glycolysis, a metabolic marker for CAF, by 6 hrs. OCC-CM/LPA-induced glycolysis in lung (MRC5) as well as ovarian fibroblasts (NOF151) was inhibited by the LPA-receptor antagonist, Ki16425. Ovarian cancer patient-derived ascitic fluid-induced aerobic glycolysis in both NFs and Ovarian CAFs and it was inhibited by Ki16425. Further analysis indicated that LPA upregulated HIF1α-levels and the silencing of HIF1α attenuated LPA-induced glycolysis in both NOFs and CAFs. These results establish LPA-induced glycolytic-shift as the earliest, potentially priming event, in NF to CAF-transition. These findings also identify a role for LPA-LPAR-HIF1α signaling-hub in the maintenance of the glycolytic-phenotype in CAFs. Our results provide evidence that targeted inhibition of LPA-mediated metabolic reprogramming in CAFs may represent an adjuvant therapy in ovarian cancer.

Objectives: Although primary results from the IMagyn050 trial showed no statistically significant improvement in progression-free survival (PFS) with atezo added to carboplatin/paclitaxel/bevacizumab (CPB) [Moore, ESMO 2020], the impact of this regimen on selected patient (pt)-reported ovarian cancer symptoms, function, and health-related quality of life (HRQoL) is unknown and the focus of this study. Methods: IMagyn050 is a double-blind randomized phase 3 trial evaluating the efficacy and safety of adding atezo/placebo to CPB followed by maintenance bevacizumab + atezo/placebo. The trial includes two cohorts: neoadjuvant chemotherapy (NACT) and primary surgery (PS). Pts complete EORTC QLQ-C30, QLQ-OV28, and FACT-G single-item GP5 at baseline (BL) and at regular intervals during treatment and follow-up. In NACT pts, prespecified responder analyses (using a ≥10-point cutoff for clinically meaningful change) assessed improvement in abdominal pain (OV28 item 31) and bloating (OV28 item 32) at week 9, comparing treatment arms by Cochran-Mantel-Haenszel (CMH) testing. Additional secondary objectives in the NACT and PS cohorts were assessments of function (physical, role, emotional, social) and HRQoL, as measured by QLQ-C30 functional and global health status/quality of life scales. Exploratory endpoints included mean change from BL in symptoms (QLQ-C30 and OV28) and treatment side-effect bother (FACT-G GP5) in both cohorts. Results: Of 1301 randomized pts, completion rates for each of the 3 questionnaires were 83-100% at BL and >85% on treatment. In NACT pts, mean BL scores were similar in the atezo vs placebo arms for abdominal pain (40.2 vs 44.8) and bloating (50.4 vs 56.3). At week 9, there was no difference between treatments in the proportion of NACT pts with ≥10-point improvement in either symptom (abdominal pain: 69/136 [51%] with atezo vs 77/142 [54%] with placebo, CMH p=0.56; bloating: 74 [54%] vs 89 [63%], CMH p=0.14). Results were consistent when restricted to NACT pts with sufficient BL symptoms to show ≥10-point improvement. There was no difference between treatments in the proportion of NACT pts with improvement in function or HRQoL at week 9. In the PS cohort, similar proportions of pts in each arm showed on-treatment improvement, stabilization, or deterioration in function and HRQoL. In both cohorts, neither arm showed meaningful changes from BL in treatment-related symptoms and similar proportions of pts in both arms reported being ‘a little bit’ or ‘somewhat’ bothered by treatment side effects while on therapy. Conclusions: Consistent with PFS results, there were no differences between arms in the proportion of NACT pts with a clinically meaningful improvement in abdominal pain and bloating after 3 cycles of bevacizumab-containing therapy. Pt-reported outcome analyses in both cohorts showed that adding atezo to CPB did not increase treatment burden for pts, thereby demonstrating the tolerability of this 4-drug regimen and providing further insight on the benefit-risk assessment of atezo.

Objectives To determine the progression free survival (PFS) of paclitaxel and carboplatin (PC) versus bleomycin, etoposide, and cisplatin (BEP) for treatment of newly diagnosed Stage IIA-IV or recurrent chemotherapy-naive ovarian sex cord-stromal tumors (SCST). Methods This study was a phase II, open-label, noninferiority trial. Eligible women with SCST were equally randomized to PC (6 cycles P 175 mg/m2 and C AUC=6 IV every 3 weeks), or BEP (4 cycles B 20 units/m2 IV push day 1, E 75 mg/m2 IV days 1–5, and cisplatin 20 mg/m2 IV days 1–5 every 3 weeks). The targeted 128 patient accrual and PFS hazard ratio (HR)=0.67 provided 85% power to exclude noninferiority margin HR=1.10. Results 63 patients were accrued at the interim futility analysis (31 PC and 32 BEP). Median age was 48 years. 87% had granulosa cell tumors. 37% had measurable disease. The DSMB closed the study early for futility of PC. The futility analysis was supported by 21/16 PFS events on the PC/BEP arms respectively, with an estimated HR=1.12 [95% CI: 0.58 to 2.16]. Median PFS was 27.7 months [7.4 to 41.0] for PC and 19.7 months for BEP [95% CI: 10.4–52.7]. PC patients had fewer grade 3 or higher adverse events (PC 77% vs BEP 90%). Differences included infections (0 vs 10%), low neutrophil count (65% vs 84%), and low WBC (22 vs 40%). One death NOS occurred on PC. Conclusions Compared to BEP, PC failed to improve PFS in ovarian SCSTs. PC showed a more favorable side effect profile.

Background Management of advanced endometrial cancer after failure with platinum therapy remains a challenge. Tumors characterized by DNA repair abnormalities are associated with high numbers of neoantigens; immunotherapy is promising in this setting as demonstrated in studies with checkpoint inhibitors (CPI). 1–6 Overcoming emerging resistance to CPI through novel combinations is a focus of research. Retifanlimab is an investigational humanized immunoglobulin G4 monoclonal antibody against programmed cell death 1 (PD 1). In POD1UM-101, retifanlimab monotherapy demonstrated acceptable tolerability and durable clinical benefit in multiple advanced tumor types, including pretreated endometrial cancer.7 POD1UM-204 is designed to further investigate efficacy and safety of retifanlimab alone or in combination with other immunotherapy or targeted agents in patients with advanced/metastatic endometrial cancer. Methods POD1UM-204 is a phase 2, multicenter, nonrandomized, open-label, umbrella study in women =18 years of age, with histologically confirmed diagnosis of advanced/metastatic endometrial cancer that has progressed on or after platinum-based chemotherapy. Patients must have an ECOG performance status =1, at least 1 measurable tumor lesion by Response Evaluation Criteria in Solid Tumors v1.1, and provide tumor tissue at baseline.Approximately 220 patients will be enrolled into 4 treatment groups: Group A–patients with MSI-H (microsatellite instability high) endometrial cancer and no prior CPI therapy (up to 100 patients) receiving retifanlimab monotherapy; Group B–patients with dMMR (deficient DNA mismatch repair) or POLE (DNA polymerase epsilon) endometrial cancer and no prior CPI therapy (up to 40 patients) receiving retifanlimab monotherapy; Group C–patients with unselected endometrial cancer and regardless of prior CPI treatment (up to 40 patients) receiving retifanlimab plus epacadostat (indoleamine 2,3-dioxygenase inhibitor); and Group D–patients with endometrial cancer and activating fibroblast growth factor receptor (FGFR1, 2 or 3) mutations or alterations outside of the kinase domain and regardless of prior CPI treatment (up to 40 patients) receiving retifanlimab plus pemigatinib (FGFR1, 2, 3 inhibitor) (figure 1). Patients can receive up to 26 treatment cycles if they continue to derive benefit and have not met criteria for withdrawal.The primary study objective is evaluating retifanlimab monotherapy antitumor activity (objective response rate [ORR] determined by independent central review [ICR]) in Group A. Secondary study objectives include assessing additional efficacy measures (duration of response, disease control rate and progression-free survival by ICR, and overall survival) in Group A; determining clinical activity (ORR by the investigator) in Groups B, C and D; and evaluating safety and tolerability of retifanlimab. Results N/A Conclusions N/A Acknowledgements This study is sponsored by Incyte Corporation (Wilmington, DE). Trial Registration ClinicalTrials. gov Identifier: NCT04463771; EudraCT 2020-000496-20 Ethics Approval The study was approved by institutional review boards or independent ethics committees of participating institutions. Consent N/A References Mittica G, Ghisoni E, Giannone G, et al. Checkpoint inhibitors in endometrial cancer: preclinical rationale and clinical activity. Oncotarget 2017;8:90532–544. Di Tucci C, Capone C, Galati G, et al. Immunotherapy in endometrial cancer: new scenarios on the horizon. J Gynecol Oncol 2019;30:e46. Brooks R, Fleming G, Lastra R, et al. Current recommendations and recent progress in endometrial cancer. CA Cancer J Clin 2019;69:258–79. Makker V, Rasco D, Vogelzang N, et al. Lenvatinib plus pembrolizumab in patients with advanced endometrial cancer: an interim analysis of a multicentre, open-label, single-arm, phase 2 trial. Lancet Oncol 2019;20:711–8. Marabelle A, Le D, Ascierto P, et al. Efficacy of Pembrolizumab in patients with noncolorectal high microsatellite instability/mismatch repair–deficient cancer: results from the phase II KEYNOTE-158 study. J Clin Oncol 2020;38:1–10. Oaknin, A, Duska L, Sullivan R, et al. Preliminary safety, efficacy, and pharmacokinetic/pharmacodynamic characterization from GARNET, a phase I/II clinical trial of the anti–PD-1 monoclonal antibody, TSR-042, in patients with recurrent or advanced MSI-h and MSS endometrial cancer. Gynecol Oncol 2019;154(suppl 1):17 [Abstract 33]. Mehnert JM, Joshua AM, Lakhani N, et al. First-in-human phase 1 study of INCMGA00012 in patients with advanced solid tumors: interim results of the cohort expansion phase. J Immunother Cancer 2018;6(suppl 1):115 [Abstract P669].

The tumor-stroma ratio (TSR) reflected on hematoxylin and eosin (H&E)-stained histological images is a potential prognostic factor for survival. Automatic image processing techniques that allow for high-throughput and precise discrimination of tumor epithelium and stroma are required to elevate the prognostic significance of the TSR. As a variant of deep learning techniques, transfer learning leverages nature-images features learned by deep convolutional neural networks (CNNs) to relieve the requirement of deep CNNs for immense sample size when handling biomedical classification problems. Herein we studied different transfer learning strategies for accurately distinguishing epithelial and stromal regions of H&E-stained histological images acquired from either breast or ovarian cancer tissue. We compared the performance of important deep CNNs as either a feature extractor or as an architecture for fine-tuning with target images. Moreover, we addressed the current contradictory issue about whether the higher-level features would generalize worse than lower-level ones because they are more specific to the source-image domain. Under our experimental setting, the transfer learning approach achieved an accuracy of 90.2 (vs. 91.1 for fine tuning) with GoogLeNet, suggesting the feasibility of using it in assisting pathology-based binary classification problems. Our results also show that the superiority of the lower-level or the higher-level features over the other ones was determined by the architecture of deep CNNs.

Purpose We conducted a randomized phase III trial to determine whether adjuvant chemotherapy improves survival in women with uterine leiomyosarcoma. Methods Women with uterus-confined, high-grade leiomyosarcoma who were confirmed disease free by imaging were randomly assigned to four cycles of gemcitabine plus docetaxel, followed by four cycles of doxorubicin, or to observation. All were followed for evidence of recurrence. The primary end point was overall survival (OS). Results With international collaboration, 38 of the targeted accrual of 216 patients were enrolled, after which the study was closed by the National Cancer Institute for accrual futility. Twenty patients were assigned to chemotherapy, 18 to observation. Among the 17 patients treated with at least one cycle of chemotherapy, grade 3 or 4 toxicities were observed in 47%; among the 18 patients assigned to observation, one had grade 3 hypertension. There were six deaths (chemotherapy, n = 5; observation, n = 1), all due to disease. The restricted mean survival time for OS was estimated as 34.3 months (95% CI, 25.3 to 43.3 months) in the chemotherapy arm and as 46.4 months (95% CI, 43.6 to 49.1 months) in the observation arm. There were eight recurrences in each arm. The restricted mean survival time for recurrence-free survival was estimated as 18.1 (95% CI, 14.2 to 22.0) months in the chemotherapy arm and as 14.6 months (95% CI, 10.3 to 19.0 months) in the observation arm. Neither survival outcome comparison was considered statistically robust, due to the small sample size. Conclusion Despite international collaboration to test the role of adjuvant chemotherapy in uterine-confined leiomyosarcoma, this study was closed for accrual futility. Although the sample size precludes robust statistical comparison, observed OS and recurrence-free survival data do not show superior outcomes with adjuvant chemotherapy.

Purpose/Objective(s) Total neoadjuvant therapy (TNT) including chemotherapy (chemo) and concurrent chemo and radiotherapy (RT) is a promising strategy for rectal cancer. Characterizing tumor regression during TNT represents an important area of research. We report a prospective validation study of the Magnetic Resonance Tumor Regression Grade (MR-TRG) using MRIs from NRG-GI002, a phase II trial of TNT for rectal cancer. Materials/Methods This was the a priori designed imaging biomarker study of NRG-GI002 seeking to correlate the MR-TRG with the pathologic Neoadjuvant Rectal score (NAR) and pathologic complete response (pCR). 110 patients (pts) were needed with 2 MRI's, 1 pre-TNT and 1 post. Three diagnostic radiologists independently reviewed each MRI and completed a consensus read. Radiologists’ assessment of complete response (mriCR) was tested for its positive predictive value (PPV), negative predictive value (NPV), sensitivity (sen), and specificity (spec) with pCR. Chi-squared test was used for association of mriCR and pCR, and Wilcoxon-Mann-Whitney test was used for association of MR-TRG and pCR. Spearman Rho was used for association of MR-TRG/mriCR and NAR. P-values Results A total of 126 patients, from 76 institutions, had 2 MRI's (pre-TNT and post), 3 radiologists completed 756 MRI interpretations. Of 126 patients, 29% were female (N = 37), 85% white (N = 107) and median age was 55 (24-76). Individual and consensus MR-TRG scores were highly associated with pCR (P Conclusion MR-TRG is significantly correlated with pathologic NAR score. Binary assessment of CR, using MR-TRG and DWI, had near perfect agreement across readers. The addition of DWI to MR-TRG improved both the sen and spec, reflecting utility of DWI in this setting. Correlation of mriCR with pCR was statistically significant. Notably, PPVs here were limited, implicating the need for stronger biomarkers when planning a national rectal organ preservation study. NCT02921256.

Importance Total neoadjuvant therapy (TNT) is often used to downstage locally advanced rectal cancer (LARC) and decrease locoregional relapse; however, more than one-third of patients develop recurrent metastatic disease. As such, novel combinations are needed. Objective To assess whether the addition of pembrolizumab during and after neoadjuvant chemoradiotherapy can lead to an improvement in the neoadjuvant rectal (NAR) score compared with treatment with FOLFOX (5-fluorouracil, leucovorin, and oxaliplatin) and chemoradiotherapy alone. Design, setting, and participants In this open-label, phase 2, randomized clinical trial (NRG-GI002), patients in academic and private practice settings were enrolled. Patients with stage II/III LARC with distal location (cT3-4 ≤ 5 cm from anal verge, any N), with bulky disease (any cT4 or tumor within 3 mm of mesorectal fascia), at high risk for metastatic disease (cN2), and/or who were not candidates for sphincter-sparing surgery (SSS) were stratified based on clinical tumor and nodal stages. Trial accrual opened on August 1, 2018, and ended on May 31, 2019. This intent-to-treat analysis is based on data as of August 2020. Interventions Patients were randomized (1:1) to neoadjuvant FOLFOX for 4 months and then underwent chemoradiotherapy (capecitabine with 50.4 Gy) with or without intravenous pembrolizumab administered at a dosage of 200 mg every 3 weeks for up to 6 doses before surgery. Main outcomes and measures The primary end point was the NAR score. Secondary end points included pathologic complete response (pCR) rate, SSS, disease-free survival, and overall survival. This report focuses on end points available after definitive surgery (NAR score, pCR, SSS, clinical complete response rate, margin involvement, and safety). Results A total of 185 patients (126 [68.1%] male; mean [SD] age, 55.7 [11.1] years) were randomized to the control arm (CA) (n = 95) or the pembrolizumab arm (PA) (n = 90). Of these patients, 137 were evaluable for NAR score (68 CA patients and 69 PA patients). The mean (SD) NAR score was 11.53 (12.43) for the PA patients (95% CI, 8.54-14.51) vs 14.08 (13.82) for the CA patients (95% CI, 10.74-17.43) (P = .26). The pCR rate was 31.9% in the PA vs 29.4% in the CA (P = .75). The clinical complete response rate was 13.9% in the PA vs 13.6% in the CA (P = .95). The percentage of patients who underwent SSS was 59.4% in the PA vs 71.0% in the CA (P = .15). Grade 3 to 4 adverse events were slightly increased in the PA (48.2%) vs the CA (37.3%) during chemoradiotherapy. Two deaths occurred during FOLFOX: sepsis (CA) and pneumonia (PA). No differences in radiotherapy fractions, FOLFOX, or capecitabine doses were found. Conclusions and relevance Pembrolizumab added to chemoradiotherapy as part of total neoadjuvant therapy was suggested to be safe; however, the NAR score difference does not support further study. Trial registration ClinicalTrials.gov Identifier: NCT02921256.

INTRODUCTION: Endometrial carcinoma (EMC) is the most common gynecologic malignancy and is highly lethal in the metastatic setting with an overall survival of 1.5-2 years due to chemotherapy resistance. Uterine carcinosarcoma (UCS) is a rare but aggressive biphasic subtype that comprises only 3-4% of EMCs but accounts for greater than 16% of uterine cancer deaths. Sulfatase 2 (SULF2) is an extracellular enzyme that activates pro-tumorigenic growth factor signaling, including transforming growth factor β (TGFβ). SULF2 is overexpressed in many cancers and is associated with tumor progression and poor prognosis. TCGA data reveal that the SULF2 gene is commonly amplified or mutated in UCS (17.5%) and may be a viable therapeutic target. OKN-007, a disulfonyl derivative of phenyl-tert-butyl nitrone, interrupts the TGFβ pathway and decreases downstream markers of epithelial mesenchymal transition (EMT) through inhibition of SULF2 in the tumor extracellular matrix. As such, assessment of OKN-007 was undertaken in UCS and EMC.METHODS: The anticancer effect of OKN-007 was evaluated in a UCS cell line (CS99) and EMC cell lines (Ishikawa, AN3CA, Hec1A, and RL95-2A). Cell viability, migration, colony-formation, and protein expression of SULF2, TGFβ, and EMT markers were evaluated in 2D and 3D culture conditions in vitro. The effect of OKN-007 alone or in combination with chemotherapy (carboplatin and paclitaxel) on tumor growth was evaluated in an intraperitoneal UCS xenograft mouse model. In vivo tumor growth was monitored using magnetic resonance imaging. TGFβ, SULF2, and EMT markers were investigated by western blot. RESULTS: SULF2 was highly expressed in UCS and EMC cells in 2D and 3D conditions. Decreased cell viability, migration, and colony-forming ability were observed with OKN-007 treatment with differences observed in IC50 values in 2D vs 3D conditions. Mice treated with OKN-007 demonstrated no observable toxicity, had decreased mean tumor volume (p Citation Format: Anjalika Gandhi, Samrita Dogra, Rheal Towner, Bethany N. Hannafon, Katherine Moxley. OKN-007 suppresses uterine carcinosarcoma by inhibiting extracellular sulfatase 2 [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr LB137.

Objective To assess whether the addition of oncolytic reovirus (Reolysin®) to weekly paclitaxel prolonged progression-free survival (PFS) in the treatment of women with recurrent or persistent ovarian, tubal or primary peritoneal cancer. Patients and methods Patients with recurrent or persistent epithelial ovarian, tubal, or peritoneal carcinoma, measurable or detectable disease, and three or fewer prior regimens were randomly assigned to paclitaxel (80mg/m 2 intravenously days 1, 8, and 15 every 4weeks) or the combination of paclitaxel (80mg/m 2 intravenously days 1, 8, and 15) plus reovirus 3×10 10 TCID 50 /day intravenously on days 1–5, both every 4weeks until disease progression or toxicity. The primary end point was PFS. The study was designed with 80% power for a one-sided alternative at a 10% level of significance to detect a reduction in the hazard by 37.5%. Results The study accrued 108 patients, 100 of whom were evaluable for toxicity. Median PFS was 4.3months for paclitaxel and 4.4months for paclitaxel plus reovirus (hazard ratio, 1.11; 90% two-sided CI, 0.78 to 1.59; one-sided P=0.687). The proportion responding (overall response rate) to paclitaxel was 20% among 45 patients with measurable disease receiving paclitaxel alone, and 17.4% among the 46 patients treated with the combination. The asymptotic relative probability of responding was 0.87 (90% CI, 0.42 to 1.79). Severe adverse events were more common in the combination regimen than in paclitaxel arm for severe neutropenia (grade≥4, 12% versus 0%), and severe respiratory adverse events (grade≥3, 25% versus 2%). No deaths were considered treatment related. Conclusion The addition of reovirus to weekly paclitaxel in the treatment of women with recurrent or persistent ovarian, tubal or peritoneal cancer did not sufficiently reduce the hazard of progression or death to warrant further investigation.

With the goal of identifying diagnostic and prognostic biomarkers in endometrial cancer, miRNA-profiling was carried out with formalin-fixed paraffin embedded (FFPE) tissue samples from 49 endometrial cancer patients. Results using an 84-cancer specific miRNA panel identified the upregulation of miR-141-3p and miR-96-5p along with a downregulation of miR-26, miR-126-3p, miR-23b, miR-195-5p, miR-374a and let-7 family of miRNAs in endometrial cancer. We validated the dysregulated expression of the identified miRNAs in a panel of endometrial cancer cell-lines. Immunohistochemical analysis of the tissue micro array derived from these patients established the functional correlation between the decreased expression of tumor suppressive miRNAs and their target oncogenes: ERBB2, EGFR, EPHA2, BAX, GNA12, GNA13, and JUN. Comparative analysis of the samples from the patients with extended progression-free survival (PFS) ( > 21 months) versus the patients with the PFS of < 21 months indicated increased expression of tumor suppressive miR-142-3p, miR-142-5p, and miR-15a-5p in samples from extended PFS patients. In addition to defining a specific set of miRNAs and their target genes as potential diagnostic biomarkers, our studies have identified tumor suppressive miR-142 cluster and miR-15a as predictors of favorable prognosis for therapy response in endometrial cancer.

Purpose To assess the diagnostic accuracy of fluorine 18 fluorodeoxyglucose (FDG) positron emission tomography (PET) combined with diagnostic contrast material-enhanced computed tomography (CT) in detecting lymph node (LN) metastasis in high-risk endometrial cancer. Materials and Methods This prospective multicenter HIPAA-compliant study had institutional review board approval, and all participants gave written informed consent. Data were accrued between January 2010 and June 2013. Patients underwent PET/CT and pelvic and abdominal lymphadenectomy. Two hundred seven of 215 enrolled patients had PET/CT and pathologic examination results for the abdomen and pelvis. Mean patient age was 62.7 years ± 9.6 (standard deviation). Data in all 23 patients with a positive abdominal examination and in 26 randomly selected patients with a negative abdominal examination were used for this central reader study. Seven independent blinded readers reviewed diagnostic CT and PET/CT results in different sessions 1 month apart. Accuracy was calculated at the participant level, correlating abdominal (right and left para-aortic and common iliac) and pelvic (right and left external iliac and obturator) LN regions with pathologic results, respecting laterality. Reader-average sensitivities, specificities, and areas under the receiver operating characteristic curve (AUCs) of PET/CT and diagnostic CT were compared. Power calculation was for sensitivity and specificity in the abdomen. Results Sensitivities of PET/CT versus diagnostic CT for the detection of LN metastasis were 0.65 (95% confidence interval [CI]: 0.57, 0.72) versus 0.50 (95% CI: 0.43, 0.58) (P = .01) in the abdomen and 0.65 (95% CI: 0.57, 0.72) versus 0.48 (95% CI: 0.41, 0.56) (P = .004) in the pelvis. Corresponding specificities were 0.88 (95% CI: 0.83, 0.92) versus 0.93 (95% CI: 0.89, 0.96) (P = .11) and 0.93 (95% CI: 0.86, 0.96) versus 0.89 (95% CI: 0.82, 0.94) (P = .27), and AUCs were 0.78 (95% CI: 0.66, 0.89) versus 0.74 (95% CI: 0.63, 0.86) (P = .39) and 0.82 (95% CI: 0.71, 0.92) versus 0.73 (95% CI: 0.63, 0.84) (P = .02). Conclusion FDG PET/CT has satisfactory diagnostic accuracy in the detection of abdominal LN metastasis in high-risk endometrial cancer. Compared with diagnostic CT alone, addition of PET to diagnostic CT significantly increased sensitivity in both the abdomen and pelvis while maintaining high specificity. © RSNA, 2017 Online supplemental material is available for this article.

Objectives Surgery is a cornerstone for patients with gynecologic malignancies. Surgical site infections (SSI) remain a source of post-operative morbidity. Consequences range from escalated costs, delay in adjuvant therapy, and increased morbidity. Our primary objective was to evaluate the effectiveness of a cyanoacrylate microbial sealant (CMS) to reduce post-operative SSI following laparotomy for suspected gynecologic malignancy. Methods Patients were randomized using a 1:1 allocation to receive either standard skin preparation or standard preparation with CMS and stratified by BMI. Patients were followed for 6weeks for SSI. Demographic data was collected through the EMR. Associations between SSI, use of CMS, and clinicopathologic factors were explored using descriptive statistics, chi-square and multivariate analysis. Results 300 patients underwent randomization. Median age of the cohort was 58. Arms were matched and there was no difference in rate of medical comorbidities. Mean BMI was 38.8kg/m 2 in patients randomized to BMI≥30 and 26.3kg/m 2 randomized to BMI p =0.18). Multivariate model demonstrated that BMI≥30 ( p p =0.010), transfusion in the OR ( p p =0.0005) were associated with post-operative SSI. Conclusions Patients presenting to a gynecologic oncologist for surgery frequently present with multiple risk factors for SSI and laparotomy is complicated by surgical-site complications in up to 30% of cases. The addition of CMS alone does not appear to reduce risk of overall SSI. Additional risk-reducing strategies including use of antimicrobial agents and optimization of modifiable risk factors prior to surgery should be explored as pathways for reducing this significant post-operative morbidity.

Objectives The ability to stratify a patient's risk of metastasis and survival permits more refined care. A proof of principle study was undertaken to investigate the relationship between single nucleotide polymorphisms (SNPs) in literature based candidate cancer genes and the risk of nodal metastasis and clinical outcome in endometrioid endometrial cancer (EEC) patients. Methods Surgically-staged EEC patients from the Gynecologic Oncology Group or Washington University School of Medicine with germline DNA available were eligible. Fifty-four genes represented by 384 SNPs, were evaluated by Illumina Custom GoldenGate array. Association with lymph node metastases was the primary outcome. Progression-free survival (PFS) and overall survival (OS) was also evaluated. Results 361 SNPs with high quality genotype data were evaluated in 337 patients with outcome data. Five SNPs in CXCR2 had an odds ratio (OR) between 0.68 and 0.70 (p-value ≤ 0.025). The A allele rs946486 in ABL had an OR of 1.5 (p-value = 0.01) for metastasis. The G allele in rs7795743 in EGFR had an OR for metastasis of 0.68 (p-value = 0.02) and hazard ratio (HR) for progression of 0.66 (p-value = 0.004). Importantly, no SNP met genome wide significance after adjusting for multiple test correcting and clinical covariates. The A allele in rs2159359 SNP in NME1 and the G allele in rs13222385 in EGFR were associated with worse OS. Both exhibited genome wide significance; rs13222385 remained significant after adjusting for prognostic clinical variables. Conclusion SNPs in cancer genes including rs2159359 SNP in NME1 and rs13222385 in EGFR may stratify risk in EEC and are prioritized for further investigation.

5527 Background: Treatment options for patients with recurrent cervical and endometrial cancer remain limited. Even with optimum care, median survival has stalled at 12-17 months. The PARP inhibitor rucaparib has demonstrated activity in both BRCA wild-type and mutant cancers. Furthermore, preclinical studies suggest a synergistic effect of PARP inhibitors and antiangiogenic agents. We hypothesized that the combination of rucaparib and the VEGF inhibitor bevacizumab would yield a clinically-significant anti-cancer effect in patients with persistent or recurrent cervical or endometrial carcinoma. Methods: NCT03476798 is a phase II trial of adults with histologically-documented carcinoma of the cervix or endometrium. Patients with evaluable lesions who had undergone at least one prior line of systemic therapy, had adequate performance status and organ function, with a life expectancy of at least three months were eligible. Biopsies were obtained prior to treatment initiation for assessment of baseline tumor biomarkers, including ARID1A mutation status. Each cycle comprised 21 days. Rucaparib was administered orally at 600 mg, twice daily. Bevacizumab was administered by IV at 15 mg/kg on day 1 of each cycle. The primary objective was to estimate the proportion of patients with persistent or recurrent cervical or endometrial cancer who survive progression-free for at least six months (PFS6). Kaplan-Meier analysis was used to estimate progression-free survival. Results: There were 28 evaluable patients; six had cervical and 22 had endometrial cancer. Median age was 60.5 years (range, 30-74). Self-reported patient races were White (82.1%), Black (10.7%), and Native American (7.1%). Self-identified Hispanic or Latina patients comprised 3.6% of the cohort. Twenty-two of 28 patients had progressive disease by six months [survival distribution function estimate = 0.214 (lower CI, 0.087; upper CI, 0.378)]. Of the six patients who achieved PFS6, one had cervical and five had endometrial cancer. Six patients had a mutation in the ARID1A gene and those patients achieved PFS6 at a rate of 66.7%. Conclusions: The study hypothesis was evaluated in a two-stage design, and the interim analysis occurred once 28 evaluable patients were enrolled. In order to move on to the second stage, at least seven patients needed to remain progression-free at six months, but only six did. Thus, the study was ended after the interim analysis. The combination of rucaparib and bevacizumab did not provide the expected clinical benefit in this cohort of patients, but may warrant further exploration in patients with ARID1A mutations. Clinical trial information: NCT03476798.

Short-form Ron (sfRon) is an understudied, alternative isoform of the full-length Ron receptor tyrosine kinase. In contrast to Ron, which has been shown to be an important player in many cancers, little is known about the role of sfRon in cancer pathogenesis. Here we report the striking discovery that sfRon expression is required for development of carcinogen-induced malignant ovarian tumors in mice. We also show that sfRon is expressed in several subtypes of human ovarian cancer including high-grade serous carcinomas, which is in contrast to no detectable expression in healthy ovaries. In addition, we report that introduction of sfRon into OVCAR3 cells resulted in epithelial-to-mesenchymal transition, activation of the PI3K and PDK1 pathway, and inhibition of the MAPK pathway. We demonstrated that sfRon confers an aggressive cancer phenotype in vitro characterized by increased proliferation and migration, and decreased adhesion of ovarian cancer cells. Moreover, the in vivo studies show that OVCAR3 tumors expressing sfRon exhibit significantly more robust growth and spreading to the abdominal cavity when compared with the parental sfRon negative OVCAR3 cells. These data suggest that sfRon plays a significant role in ovarian cancer initiation and progression, and may represent a promising therapeutic target for ovarian cancer treatment.

The bottleneck in current vector-based cancer therapy is the targeted and controlled release of therapeutics in tumors. Exosomes are submicron-sized vesicles that are secreted by all cell types and are involved in communication and transportation of materials between cells. Analogous in size and function to synthetic nanoparticles, exosomes offer many advantages, rendering them the most promising candidates for targeted drug or gene delivery vehicles. Patient-specific customized therapeutic strategies can be engineered using exosomes derived from the patient's own healthy cells. Therefore, exosome-based cancer therapy has the potential to become an important part of personalized medicine. Interest in exosomes as carrier organelles is relatively recent. Knowledge about exosomal biology and its applications remains limited. The present review is an attempt to describe the current status of the application of exosomes to cancer therapy and the potential challenges associated with their use.

Background: Endometrial cancer is the most common gynecologic malignancy and incidence and mortality continue to increase. While active chemotherapy regimens are available, the median time to progression remains six to eight months and mortality in this setting is approaches 100%. Rapid progression and lack of response to systemic therapies are significant challenges in this disease and underscore the need for identification of molecular markers and targets of drug resistance. Alterations in non-coding micro-RNA (miRNA) have been shown to be necessary for cellular proliferation and differentiation in normal and malignant phenotypes. miRNA signatures unique to malignant, specifically treatment resistant cancers are described in many solid tumors; however, knowledge of these changes in resistant endometrial cancer is limited. Identification of molecular signatures unique to drug resistant endometrial cancer may inform novel therapeutic targets to predict and/or reverse resistance to chemotherapy. One source of tumor miRNA is in exosomes derived from tumor cells that transport tumor derived products including non-coding RNAs. They are readily identified in peripheral body fluids and are known to act at distant sites to facilitate tumor growth and metastasis. As such, exosomes are a potential source of tumor-specific genetic material that may characterize disease and predict drug response. Results: In characterized endometrial cancer cell lines, we determined chemo-sensitivity to platinum-based therapy. Exosomes were isolated and purified from the culture media and exosomal miRNA profiles were sequenced, aligned and cross-referenced with published datasets. Our preliminary data have identified a unique exosomal miRNA signature that differentiates chemotherapy sensitive and resistant endometrial cancer in vitro. Upregulation of oncomirs, miR155, miR98 and downregulation of tumor suppressor miRNAs, miR25, miR27a, miR200a were observed. Two unique miRNA candidates were also identified miR3125 and miR6068. The clinical expression of these candidate miRNAs was interrogated within the Kaplan Meier plotter RNAseq dataset and decreased overall survival was observed in samples exhibiting dysregulation of the candidate miRNA panel1. These results were confirmed in vitro with rtPCR. We have since isolated exosomes from the sera of 14 endometrial cancer patients with known clinical outcomes data. Seven patients exhibited rapid disease progression on standard chemotherapy while 7 demonstrated a complete clinical response of at least 6 months. The exosomal miRNA was isolated, quantified and sequencing is ongoing. Conclusions: Our data implicate a unique grouping of miRNAs in the endometrial cancer chemo-resistance. The clinical impact of these miRNAs remains unknown, but 5 are associated with a significant decrease in overall survival1. The ability to detect a unique miRNA profile in the body fluids of patients, provides an accessible resource for study before and during treatment. Such data could significantly impact treatment of endometrial cancer both as a predictive biomarker of response and source of novel therapeutic targets. The high rate of recurrence, inevitable development of a chemo-resistant disease and the clinical challenges associated with repeated tumor sampling, make exosomal miRNA a valuable candidate for exploration in this disease. Citation Format: Katherine M. Moxley. Exosomes in predicting response to chemotherapy in endometrial carcinoma [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 1418.

Background: Ovarian cancer is the fifth leading cause of cancer death in the United States in large part because more than 85% of patients present with metastatic disease. The majority of these are epithelial of high-grade serous (HGSOC) histology where the 5-year overall survival remains poor despite advances in chemotherapy and improved understanding of molecular tumor biology. As a heterogeneous, p53-driven malignancy, HGSOC employs a variety of molecular and metabolomic changes that promote preferential tumor growth within the peritoneal cavity. The miliary spread patterns of ovarian cancers within the peritoneal cavity create distinct clinical challenges, particularly in the preoperative assessment of surgical resectability. Therefore, it is clinically imperative to develop better imaging modalities capable of accurately predicting location, volume, and consistency of tumor implants that will inform treatment planning without exposing the patient to unnecessary harm. To this end, recent research efforts have focused on identifying novel clinical and imaging biomarkers capable of predicting optimal surgical resection, tumor response, and disease specific outcomes. Unfortunately, the clinical utility of such markers in individualizing patient specific treatment strategies remains limited. In the absence of unique molecular biomarkers, the use of advanced quantitative imaging features that utilize novel computer-aided detection (CAD) algorithms and can train artificial intelligence platforms will improve the accuracy with which current imaging modalities predict surgical outcome in HGSOC. Results: Using an established imaging platform established at Tempus Labs, Inc. we evaluated preoperative images of women undergoing cytoreductive surgery for advanced epithelial ovarian cancer. The CT images of forty-five retrospectively collected cases, with complete clinical outcomes data, are undergoing analysis. Lesions have been identified, contoured from CT images, and approved by a trained radiologist. Over 2,000 quantitative imaging (radiomic) features will be extracted from these images, and using the Tempus proprietary imaging platform, feature selection will be performed to identify factors predictive of surgical resectability and compare them to those that do not correlate with surgical outcome. These selected features will be used to model surgical outcomes data. Conclusions: Prior research has demonstrated the utility of quantitative CT image analysis in predicting response to chemotherapy and platinum sensitivity. This same technology, however, has not been applied to the surgical management of women with advanced HGSOC. In this study, we assessed the utility of similar “radiomics algorithms” in order to establish a set of radiomic features that characterize tumors amenable to optimal (R0) surgical resection. The selected features identified in this retrospective training cohort are now be tested in a larger prospective cohort of metastatic HGSOC cancer patients. Citation Format: Paul A. DiSilvestro, Jessica DiSilvestro, Jonathan S. Ball, Stephen Yip, Anna Schwarzbach, Katherine M. Moxley. Analysis of advanced quantitative computed tomography imaging features in predicting the surgical resectability of advanced epithelial ovarian cancer [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 A12.

Background: Ovarian cancer is the fifth leading cause of cancer death in the United States, with an anticipated 13,000 deaths in the United States in 2017. High-grade serous (HGSOC) sub-type is both the most common and the most lethal histopathology. As more than 85% of patients present with metastatic disease, the 5-year overall survival remains dismal at only 30%. It is known that human cancers are initiated and maintained by self-renewing stem cells. Cancer stem cell related protein Doublecortin-like kinase 1 (DCLK1) is a major regulator of stemness and EMT, promoting drug resistance, tumor dormancy, progression, and metastasis in solid tumors. The molecular mechanisms of DCLK1 in tumorigenesis and drug resistance remain limited but DCLK1 expression is increased in cancer, especially in the drug-resistant setting. Results: Our lab has made the discovery that DCLK1 is overexpressed in HGSOC where it regulates tumor progression, tumor stemness, and EMT via noncoding RNA regulation of downstream oncogenic pathways. Our study involved assessment of DCLK1 expression in tissue culture and xenografts created from primary patient-derived cell lines (PDACs). Expression of the DCLK1 protein was assessed by immunoblot analysis in lysates of cell lines representing HGSOC and PDACs obtained from the ascites of patients at the University of Oklahoma at the time of diagnosis, prior to initiation of chemotherapy. FTE-187, FTE-188, and OSE cells were used as normal controls. Elevated levels of DCLK1 expression were observed in the panel of HGSOC cell lines and in PDAC lines. Our in vitro studies confirmed DCLK1 upregulation in the tissue and stroma of PDAC lines and our in vivo studies with PDAC xenografts revealed that siRNA knockdown of DCLK1 negatively regulated tumor suppressor miRNAs including mi-143/145, let-7a, and miR-200a-c resulting in downregulation of pluripotency, EMT, and proangiogenic markers including OCT4, SOX2, KLF4, NANOG, LIN28B, NOTCH, KRAS, and c-MYC. This was most pronounced in the setting of platinum resistance. For this, lysates from paired cells lines with known platinum-sensitive and platinum-resistant variants were assessed (Tyk-nu/Tyk-nu CPR and A2780/A2780.cp). An increase in DCLK1 expression was observed in both of the platinum-resistant lines compared to the platinum-sensitive lines. This was confirmed in PDACs where clinical platinum status correlated with DCLK1 expression, suggesting the role of DCLK1 in inherent platinum-resistant tumor clones. Conclusions: Our data implicate DCLK1 in ovarian cancer pathogenesis and platinum resistance and indicate that DCLK1 knockdown via may suppress malignant progression and resensitize tumors to platinum chemotherapy. These data could have a significant impact on future therapeutic strategies involving DCLK1 inhibitors in treating platinum-resistant patients. Given the high recurrence rate and inevitable development of platinum-resistant disease, such compounds have high potential to impact clinical outcomes in this lethal malignancy. Citation Format: Katherine M. Moxley, JiHee Ha, Daniel Andrade, Courtney Houchen, Danny Dhanasekaran. DCLK1 mediates tumor stemness and platinum resistance in high-grade serous epithelial ovarian cancer [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 A55.

6003 Background: Combination cediranib (C) and olaparib (O) improved progression-free survival (PFS) in patients (pts) with relapsed platinum (plat)-sensitive high-grade ovarian cancer (ovca) compared to O alone in a Phase 2 trial (NCT01116648). We conducted this randomized, open-label Phase 3 trial (NCT02446600) to assess whether combination C+O, or O alone, was superior to standard of care (SOC) plat-based therapy in relapsed plat-sensitive ovca. Methods: Eligible pts had recurrent plat-sensitive [ > 6-month plat-free interval (PFI)] high-grade serous or endometrioid, or BRCA-related, ovca. One prior non-plat therapy and unlimited prior plat-therapies were allowed; prior anti-angiogenics in the recurrent setting or prior PARP inhibitor were exclusions. Pts were randomized 1:1:1 to SOC (carboplatin/paclitaxel; carboplatin/gemcitabine; or carboplatin/liposomal doxorubicin), O (300mg twice daily), or C+O (C 30mg daily + O 200mg twice daily). Randomization was stratified by g BRCA status, PFI (6-12 vs > 12 months), and prior anti-angiogenic therapy. Target sample size was 549 pts; primary analysis occurred 2 years after the last pt enrolled. The primary endpoint was PFS. Type 1 error = 0.025 was controlled by a gatekeeping hierarchy that assessed C+O vs SOC, then O alone vs SOC, and finally C+O vs O. All maintenance therapy was prohibited. Results: Between 4FEB2016 and 13NOV2017, 565 pts enrolled (187 SOC, 189 O, 189 C+O), and 528 pts initiated treatment (166 SOC, 183 O, 179 C+O). 23.7% of patients had g BRCAmut. Median follow-up was 29.1 months. 53 pts on SOC initiated non-protocol therapy (predominantly PARP inhibitor maintenance) before disease progression. The hazard ratio (HR) for PFS was 0.856 (95% CI 0.66-1.11, p = 0.08, 1-tail) between C+O and SOC and 1.20 (95% CI 0.93-1.54) between O and SOC, with median PFS of 10.3, 8.2, and 10.4 months for SOC, O, and C+O, respectively. Response rates were 71.3% (SOC), 52.4% (O), and 69.4% (C+O). In gBRCA pts, HR for PFS was 0.55 (95% CI 0.73-1.30) for C+O vs SOC, and 0.63 (95% CI 0.37-1.07) for O vs SOC. In non-g BRCA pts, HR for these comparisons was 0.97 (95% CI 0.73-1.30) and 1.41 (1.07-1.86). No OS differences between arms were observed at 44% events. Pts receiving C+O (vs SOC) had more frequent Grade 3 or higher gastrointestinal (30.1% vs 8.4%), hypertension (31.7% vs 1.8%), and fatigue events (17.5% vs 1.8%). Conclusion: C+O demonstrated similar activity to SOC in relapsed plat-sensitive ovca but did not meet the primary endpoint of improved PFS. Clinical trial information: NCT02446600.

6087 Background: SAP (TAK-228, MLN0128) is a selective dual inhibitor of mammalian target of rapamycin complexes 1 and 2. In endometrial tumor xenograft models, SAP+PAC exhibited stronger antitumor efficacy than PAC alone. Methods: Female pts with histologic/cytologic diagnosis of endometrial cancer were randomized to receive SAP 4 mg by mouth (days [d] 2–4, 9–11, 16–18, 23–25) plus PAC 80 mg/m2 intravenously (d 1, 8, 15), or PAC alone, in 28-day cycles until unacceptable toxicity or disease progression. Randomization was stratified by histologic subtype, lines of prior chemotherapy (1 vs. 2), and prior taxane therapy. The primary endpoint was progression-free survival (PFS); secondary endpoints included overall survival (OS), overall response rate (ORR), clinical benefit rate (CBR; ORR + stable disease), and safety. Additional treatment arms of SAP alone (weekly dosing) and SAP+TAK-117 were closed after futility analyses. Results: 180 pts were randomized to SAP+PAC (n=90) or PAC (n=90); 86 and 87 pts received SAP+PAC and PAC, respectively; 3 pts from each arm were ongoing on treatment at data cut (30 July 2019). Baseline characteristics were balanced between arms. After a median follow-up of 17.2 vs. 14.4 mos with SAP+PAC vs. PAC, median PFS was 5.6 mos vs. 3.7 mos (hazard ratio [HR] 0.82; 95% CI 0.58–1.15). In pts with endometrioid histology (n=116), median PFS was 5.7 mos with SAP+PAC vs 3.3 mos with PAC (HR 0.66; 95% CI 0.43–1.03). In pts with nonendometrioid histology (n=64), median PFS was 3.6 mos with SAP+PAC vs. 5.4 mos with PAC (HR 1.09; 95% CI 0.62–1.90). Median OS was 13.7 mos with SAP+PAC vs. 14.6 mos with PAC (HR 1.01; 95% CI 0.67–1.53). Confirmed ORR was 24% with SAP+PAC vs. 18% with PAC (endometrioid, 23% vs. 16%; nonendometrioid, 28% vs. 22%); CBR was 80% vs. 58% (endometrioid, 84% vs. 55%; nonendometrioid, 72% vs. 63%). Median number of cycles received was 5 (range 1–23) with SAP+PAC and 4 (range 1–37) with PAC. Rates of grade ≥3 treatment-emergent adverse events (TEAEs) were 90% with SAP+PAC vs. 54% with PAC; the most common included anemia (21% vs.12%), neutropenia (12% vs. 3%), fatigue (12% vs. 5%), hypophosphatemia (12% vs. 1%), and pulmonary embolism (11% vs. 3%). Conclusions: Median PFS was longer with SAP+PAC vs. PAC in pts with endometrial cancer but did not reach statistical significance. PFS was particularly longer in the endometrioid subtype but again was not significant, and further studies are warranted. Incidence of grade ≥3 TEAEs was higher with SAP+PAC vs. PAC, but SAP+PAC toxicity was manageable, with no new safety signals. Clinical trial information: NCT02725268.

Objective Preclinical data suggest elesclomol increases oxidative stress and enhances sensitivity to cytotoxic agents. The objective of this prospective multicenter phase 2 trial was to estimate the activity of IV elesclomol plus weekly paclitaxel in patients with platinum-resistant recurrent ovarian, tubal or peritoneal cancer through the frequency of objective tumor responses (ORR). Methods Patients with measurable disease, acceptable organ function, performance status ≤ 2, and one prior platinum containing regimen were eligible. A two-stage design was utilized with a target sample size of 22 and 30 subjects, respectively. Prior Gynecologic Oncology Group studies within the same population involving single agent taxanes showed an ORR of approximately (20%) and served as a historical control for direct comparison. The present study was designed to determine if the regimen had an ORR of ≥40% with 90% power. Results Fifty-eight patients were enrolled, of whom 2 received no study treatment and were inevaluable. The median number of cycles was 3 (268 total cycles, range 1–18). The number of patients responding was 11 (19.6%; 90% CI 11.4% to 30.4%) with one complete response. The median progression-free survival and overall survival was 3.6 months and 13.3 months, respectively. The median ORR duration was 9.2 months. Percentages of subjects with grade 3 toxicity included: Neutropenia 9%; anemia 5%; metabolic 5%; nausea 4%; infection 4%; neurologic (mostly neuropathy) 4%; and vascular (mostly thromboembolism) 4%. There were no grade 4 toxicities reported. Conclusions This combination was well tolerated but is unworthy of further investigation based on the proportion responding [ ClinicalTrials.gov Identifier: NCT00888615 ].

Deep convolutional neural networks (CNNs) based transfer learning is an effective tool to reduce the dependence on hand-crafted features for handling medical classification problems, which may mitigate the problem of the insufficient training caused by the limited sample size. In this study, we investigated the discrimination power of the features at different CNN levels for the task of classifying epithelial and stromal regions on digitized pathologic slides which are prepared from breast cancer tissue. We extracted the low level and high level features from four different deep CNN architectures namely, AlexNet, Places365-AlexNet, VGG, and GoogLeNet. These features are used as input to train and optimize different classifiers including support vector machine (SVM), random forest (RF), and k-nearest neighborhood (KNN). A number of 15000 regions of interest (ROIs) acquired from the public database are employed to conduct this study. The result was observed that the low-level features of AlexNet, Places365-AlexNet and VGG outperformed the high-level ones, but the situation is in the opposite direction when the GoogLeNet is applied. Moreover, the best accuracy was achieved as 89.7p by the relatively deep layer of max pool 4 of GoogLeNet. In summary, our extensive empirical evaluation may suggest that it is viable to extend the use of transfer learning to the development of high-performance detection and diagnosis systems for medical imaging tasks.

Although hypoxia has been shown to reprogram cancer cells toward glycolytic shift, the identity of extrinsic stimuli that induce metabolic reprogramming independent of hypoxia, especially in ovarian cancer, is largely unknown. In this study, we use patient-derived ovarian cancer cells and high-grade serous ovarian cancer cell lines to demonstrate that lysophosphatidic acid (LPA), a lipid growth factor and GPCR ligand whose levels are substantially increased in ovarian cancer patients, triggers glycolytic shift in ovarian cancer cells. Inhibition of the G protein α-subunit Gαi2 disrupted LPA-stimulated aerobic glycolysis. LPA stimulated a pseudohypoxic response via Rac-mediated activation of NADPH oxidase and generation of reactive oxygen species, resulting in activation of HIF1α. HIF1α in turn induced expression of glucose transporter-1 and the glycolytic enzyme hexokinase-2 (HKII). Treatment of mice bearing ovarian cancer xenografts with an HKII inhibitor, 3-bromopyruvate, attenuated tumor growth and conferred a concomitant survival advantage. These studies reveal a critical role for LPA in metabolic reprogramming of ovarian cancer cells and identify this node as a promising therapeutic target in ovarian cancer. Significance: These findings establish LPA as a potential therapeutic target in ovarian cancer, revealing its role in the activation of HIF1α-mediated metabolic reprogramming in this disease. Cancer Res; 78(8); 1923–34. ©2018 AACR.

Exosomes have great potential to serve as a source of diagnostic and prognostic biomarkers for endometrial cancer (EC). Urine-derived exosomes from patients with EC and patients with symptoms of EC, but without established EC, were used to evaluate a unique miRNA expression profile. Of the 84 miRNA studied, 57 were amplified in qPCR, suggesting the differential packaging of miRNA in exosomes. Further, hsa-miR-200c-3p was identified to be enriched the most. Various bioinformatics and in silico tools were used to evaluate the biological significance of hsa-miR-200c-3p in EC. We conclude that differential miRNA in exosomes can be utilized for discovery of biomarker signatures and EC diagnosis; hsa-miR-200c-3p is one such candidate. Urine-derived exosomes pave the way for the development of non-invasive biomarkers.

Exosomes are now realized as important contributors in cellular communication, which can transport many biological molecules and perturb physiological changes in the recipient cells. Several researchers have suggested that the content of exosomes changes when they are released from cells under physiological (i.e., disease) or physical stress. Because noncoding RNAs, such as miRNA (miRNA) and long noncoding RNA, form a major component of exosome cargo, they have been regarded as a vital source of biomarkers and therapeutic cargo. This chapter discusses in detail the noncoding RNAs present in the exosome and the literature exploring the use of exosome miRNAs as biomarkers for diagnosis and treatment response in major cancer types. The technology and newer methods of studying exosome RNAs are also reviewed.

Objective This two-stage phase II study assessed activity of single agent dalantercept in patients with recurrent/persistent endometrial carcinoma (EMC). Methods Eligible patients had persistent/recurrent EMC after 1–2 prior cytotoxic regimens, measurable disease (RECIST 1.1), and GOG performance≤2. Dalantercept 1.2mg/kg subcutaneous was administered once every 3weeks until disease progression (PD)/development of prohibitory toxicity. Primary objectives were to estimate the proportion of patients with persistent/recurrent EMC, who survive progression-free without receiving non-protocol therapy (TPFS) for at least 6months and to estimate the proportion having objective tumor response. Results All 28 enrolled patients were eligible and evaluable. Median age: 62years. Most common histologies: 32% Grade 1/2 endometrioid and 54% serous tumors. Prior treatment: 1 or 2 regimens in 82% and 18% of patients, respectively. Eighteen patients received prior radiation therapy. Patients received 1–12 cycles of dalantercept, and 46% of patients received ≤2cycles. The most common adverse events (AE) were fatigue, anemia, constipation and peripheral edema. Grade 3/4 AEs occurred in 39% and 4% of patients. One grade 5 gastric hemorrhage in a patient with a history of radiation fibrosis/small bowel obstruction was deemed possibly dalantercept-related. All patients are off study: 86% for PD. No ORs were observed; 57% had stable disease and 11% had TPFS>6 mos. Median progression-free and overall survival: 2.1months (90% CI: 1.4–3.2) and 14.5months (90% CI: 7.0–17.5), respectively. Conclusions Dalantercept has insufficient single agent activity in recurrent EMC to warrant further investigation at this dose level and schedule.

Background The polycomb group protein, BMI1 plays important roles in chromatin modification, stem cell function, DNA damage repair and mitochondrial bioenergetics. Such diverse cellular functions of BMI1 could be, in part, due to post-translational modifications, especially phosphorylation. To date, AKT has been reported as a kinase that by site specific phosphorylation of BMI1 modulates its oncogenic functions. Methods Immunoprecipitation in conjunction with kinase assay and mass spectrometry was used to determine association with and site specific phosphorylation of BMI1 by CK2α. Functional implications of the BMI1/CK2α axis was examined in cancer cells utilizing siRNA and exogenous gene expression followed by biochemical and phenotypic studies. Correlations between expression of CK2α and BMI1 were determined from cell lines and formalin fixed paraffin embedded tissues representing the normal fallopian tube epithelium and high grade serous ovarian cancer samples. Results Here we report that CK2α, a nuclear serine threonine kinase, phosphorylates BMI1 at Serine 110 as determined by in-vitro/ex-vivo kinase assay and mass spectrometry. In ovarian cancer cell lines, expression of CK2α correlated with the phospho-species, as well as basal BMI1 levels. Preventing phosphorylation of BMI1 at Serine 110 significantly decreased half-life and stability of the protein. Additionally, re-expression of the phosphorylatable but not non-phosphorylatable BMI1 rescued clonal growth in endogenous BMI1 silenced cancer cells leading us to speculate that CK2α-mediated phosphorylation stabilizes BMI1 and promotes its oncogenic function. Clinically, compared to normal fallopian tube epithelial tissues, the expression of both BMI1 and CK2α were significantly higher in tumor tissues obtained from high-grade serous ovarian cancer patients. Among tumor samples, the expression of BMI1 and CK2α positively correlated (Spearman coefficient = 0.62, P = 0.0021) with each other. Conclusion Taken together, our findings establish an important regulatory role of CK2α on BMI1 phosphorylation and stability and implicate the CK2α/BMI1 axis in ovarian cancer. Electronic supplementary material The online version of this article (doi:10.1186/s12943-017-0617-8) contains supplementary material, which is available to authorized users.

Predicting metastatic tumor response to chemotherapy at early stage is critically important for improving efficacy of clinical trials of testing new chemotherapy drugs. However, using current response evaluation criteria in solid tumors (RECIST) guidelines only yields a limited accuracy to predict tumor response. In order to address this clinical challenge, we applied Radiomics approach to develop a new quantitative image analysis scheme, aiming to accurately assess the tumor response to new chemotherapy treatment, for the advanced ovarian cancer patients. During the experiment, a retrospective dataset containing 57 patients was assembled, each of which has two sets of CT images: pre-therapy and 4-6 week follow up CT images. A Radiomics based image analysis scheme was then applied on these images, which is composed of three steps. First, the tumors depicted on the CT images were segmented by a hybrid tumor segmentation scheme. Then, a total of 115 features were computed from the segmented tumors, which can be grouped as 1) volume based features; 2) density based features; and 3) wavelet features. Finally, an optimal feature cluster was selected based on the single feature performance and an equal-weighed fusion rule was applied to generate the final predicting score. The results demonstrated that the single feature achieved an area under the receiver operating characteristic curve (AUC) of 0.838±0.053. This investigation demonstrates that the Radiomic approach may have the potential in the development of high accuracy predicting model for early stage prognostic assessment of ovarian cancer patients.

Accurate tumor segmentation is a critical step in the development of the computer-aided detection (CAD) based quantitative image analysis scheme for early stage prognostic evaluation of ovarian cancer patients. The purpose of this investigation is to assess the efficacy of several different methods to segment the metastatic tumors occurred in different organs of ovarian cancer patients. In this study, we developed a segmentation scheme consisting of eight different algorithms, which can be divided into three groups: 1) Region growth based methods; 2) Canny operator based methods; and 3) Partial differential equation (PDE) based methods. A number of 138 tumors acquired from 30 ovarian cancer patients were used to test the performance of these eight segmentation algorithms. The results demonstrate each of the tested tumors can be successfully segmented by at least one of the eight algorithms without the manual boundary correction. Furthermore, modified region growth, classical Canny detector, and fast marching, and threshold level set algorithms are suggested in the future development of the ovarian cancer related CAD schemes. This study may provide meaningful reference for developing novel quantitative image feature analysis scheme to more accurately predict the response of ovarian cancer patients to the chemotherapy at early stage.

The study aimed to investigate the role of applying quantitative image features computed from computed tomography (CT) images for early prediction of tumor response to chemotherapy in the clinical trials for treating ovarian cancer patients.A dataset involving 91 patients was retrospectively assembled. Each patient had two sets of pre- and post-therapy CT images. A computer-aided detection scheme was applied to segment metastatic tumors previously tracked by radiologists on CT images and computed image features. Two initial feature pools were built using image features computed from pre-therapy CT images only and image feature difference computed from both pre- and post-therapy images. A feature selection method was applied to select optimal features, and an equal-weighted fusion method was used to generate a new quantitative imaging marker from each pool to predict 6-month progression-free survival. The prediction accuracy between quantitative imaging markers and the Response Evaluation Criteria in Solid Tumors (RECIST) criteria was also compared.The highest areas under the receiver operating characteristic curve are 0.684 ± 0.056 and 0.771 ± 0.050 when using a single image feature computed from pre-therapy CT images and feature difference computed from pre- and post-therapy CT images, respectively. Using two corresponding fusion-based image markers, the areas under the receiver operating characteristic curve significantly increased to 0.810 ± 0.045 and 0.829 ± 0.043 (P 0.05), respectively. Overall prediction accuracy levels are 71.4%, 80.2%, and 74.7% when using two imaging markers and RECIST, respectively.This study demonstrated the feasibility of predicting patients' response to chemotherapy using quantitative imaging markers computed from pre-therapy CT images. However, using image feature difference computed between pre- and post-therapy CT images yielded higher prediction accuracy.