Objective: This study aimed to assess the concordance between pre-operative imaging results and intra-operative findings in patients with endometrial carcinoma, based on pre-operative risk stratification and the clinical indication for obtaining imaging. Methods: We identified all patients who had surgery for newly diagnosed endometrial carcinoma at our institution from January 2017 to January 2021 and categorized them as low or high risk. We reviewed pre-operative images or radiologic evidence of extrauterine disease and correlated with intra-operative findings, and we categorized the findings as true positive or false negative. Results: We identified a total of 1970 patients for this analysis. Within the study cohort, 1247 tumors (63.3%) were low-risk histology, 719 (36.5%) were high-risk histology, and 4 patients (0.2%) had no pre-operative biopsy. Pre-operative imaging was performed for 676 low-risk (54.2%) and 700 high-risk patients (97.4%) (p <.001) and detected the presence of extra-uterine disease in 50 of 676 low-risk (7.4%) and 140 of 700 high-risk patients (20%). In the low-risk cohort, there was radiologic evidence of extra-uterine disease present for 12 of 60 patients (20%) with an abnormal exam or ultrasound, 16 of 126 patients (12.7%) with symptoms or a delayed diagnosis, and 16 of 490 asymptomatic patients (3.3%). In the high-risk cohort, there was radiologic evidence of extra-uterine disease present for 32 of 66 patients (48.5%) with symptoms or a delayed diagnosis, 21 of 46 patients (45.7%) with an abnormal exam or ultrasound, and 84 of 583 asymptomatic patients (14.4%). A total of 122 of 1376 patients (8.9%) had a change in clinical management based on pre-operative imaging findings: 32 of 676 (4.7%) low-risk and 90 of 700 high-risk patients (12.9%). Conclusions: Over 90% of patients with low-risk tumors had no findings suspicious for metastatic disease on pre-operative imaging, indicating limited utility in that cohort. However, pre-operative imaging yielded clinically meaningful information in patients with high-risk histology, particularly, those with symptoms, delayed diagnosis, or concerning exam findings.

Solid tumor antigen heterogeneity is a major challenge for cancer immunotherapies, including chimeric antigen receptor (CAR) T cells. Unlike CD19 for B cell malignancies, no target with pan-cellular expression in solid tumors and absence in normal vital cells has been identified. CD70 is a promising candidate, physiologically confined to immune cell subsets and aberrantly expressed in many cancers. We show that heterogeneous CD70 expression in tumors is epigenetically regulated, ranging from high to very low in individual cells, appearing negative by conventional detection methods. Using a highly sensitive CD70 receptor, HLA-independent T cell (HIT) receptor coexpressing CD80 and 4-1BBL for costimulation, we efficiently eliminated CD70-heterogeneous tumors that evade prototypic CAR T cells. These findings provide a potential strategy to treat a broad range of solid tumors.

5561 Background: Low grade serous ovarian carcinoma (LGSOC) is characterized by MAPK pathway alterations (most commonly KRAS , which occurs in 1/3 of tumors) and poor chemosensitivity. Avutometinib (MEK/RAF clamp) in combination with defactinib (FAK inhibitor) has shown durable responses in patients with recurrent LGSOC harboring a KRAS mutation and was recently FDA approved for this indication. However, data is lacking regarding the activity of avutometinib + defactinib (A+D) vs single agent MEK inhibitors for treatment of women with recurrent LGSOC. Here we examine patients who were treated with A+D. We also examine the duration of treatment in patients who received a single agent MEK inhibitor followed by A+D. Methods: In this single-institution retrospective cohort study, patients with recurrent LGSOC treated with A+D between 2021 – 2025 were included and stratified by tumor KRAS mutation status. The primary outcome was median duration of treatment on A+D, compared using Wilcoxon-Rank sum test. Pair-wise comparison of treatment duration in patients who were treated with single agent MEK inhibitors and A+D was performed. Results: 26 patients were included in the final analysis. Median age at diagnosis was 49 years (range 20-85). Patients had a median of 4 (range 1-9) prior lines of therapy, including hormonal (92%, n=24), bevacizumab (38%, n=10), and MEK inhibitors trametinib and binimetinib (35%, n=8). Eight (31%) patients harbored a tumor KRAS mutation. Among these 26 patients, median duration of treatment with A+D was 16.8 months in the KRAS -mutant and 5.5 months in the KRAS -wild type cohorts, respectively (p=0.037). Nine patients were treated with both A+D and a MEK inhibitor, three of whom had KRAS -mutant tumors. All nine patients received the single agent MEK inhibitor prior to treatment with A+D, with a median of 7 (range 4-8) prior lines of therapy before treatment with A+D. Among the three patients with KRAS -mutant tumors, median duration of treatment with single agent MEK inhibitor was 4.4 months (range 0.4-24.2) while with A+D, it was 17.6 months (range 6.1-23.4). Conclusions: In this single-institution retrospective cohort, treatment with A+D was associated with a significantly longer median duration of therapy in patients with KRAS -mutant recurrent LGSOC compared with KRAS -wild type disease. Among patients who received both regimens, A+D appeared to provide longer treatment duration than prior single agent MEK inhibitors in the KRAS -mutant subset, though this is limited by small sample size. These findings support improved activity of A+D in KRAS -mutant LGSOC and warrant confirmation in larger prospective studies.

OBJECTIVE:To examine if preoperative serum albumin (SA) level is associated with worse survival following primary debulking surgery (PDS) for epithelial ovarian cancer (EOC). METHODS:In this retrospective study, we identified patients with stage III-IV EOC who underwent PDS at our institution from 1/1/2015-8/31/2022. Patients were assigned to the low albumin (LA; SA < 3.5 g/dL) or normal albumin (NA; SA ≥ 3.5 g/dL) group. Exclusion criteria included age ≥ 80 years, age 75-79 years with poor physical status or high likelihood of complex surgery, secondary malignancy, and missing preoperative SA level. Median and 5-year progression-free (PFS) and overall survival (OS) rates were calculated using Kaplan-Meier method. Log-rank test was used for categorical variables and Wald test for continuous variables. RESULTS:Overall, 636 patients were identified, 65 (10%) in the LA and 571 (90%) in the NA group. Median follow-up was 73 months (95% CI: 68-76). Comparing the LA versus NA group, 5-year PFS rate was 18.3% (95% CI: 9.2%-29.9%) vs 26.5% (95% CI: 22.6%-30.5%), respectively (P = 0.107); 5-year OS rate was 56.1% (95% CI: 41.5%-68.4%) vs 66.1% (95% CI: 61.6%-70.3%), respectively (P = 0.053). Complete gross resection (CGR) was achieved in 66% of patients with LA compared to 78% with NA (P = 0.029). On multivariate analysis of OS, age, histology, BRCA status, and CGR were statistically significant prognostic factors, but preoperative SA was not. CONCLUSIONS:Five-year survival rates were similar between patients with and without hypoalbuminemia. Relying on preoperative SA level alone, patients may be selected for neoadjuvant chemotherapy who would have equivalent survival outcomes with PDS.

OBJECTIVE:Noncompete clauses in employment contracts limiting the location and scope in which a physician can practice after leaving their current employer are common in the U.S. We assessed the prevalence, awareness of, and physicians' attitudes toward these agreements in gynecologic oncology. METHODS:A 10-item anonymous survey was circulated to 1221 physician members of the Society of Gynecologic Oncology (SGO). The survey included questions regarding demographics, practice setting, and knowledge of and opinions about noncompete agreements. Descriptive statistics were performed. RESULTS:A total of 207 (17%) of 1221 SGO members completed the survey; 91 (44%) represented academic practices, 53 (26%) hybrid practices, 37 (18%) private practices, and 23 (11%) other practice settings. Overall, 145 respondents (70%) reported their current employment contract contained a noncompete clause, 166 (80%) reported they had ever signed an employment contract in gynecologic oncology containing one, and 89 (43%) reported their decision where to seek employment opportunities was affected by a noncompete clause. Overall, 132 respondents (64%) were familiar with enforceability of noncompete agreements in their current jurisdiction, and 172 (83%) had a negative view of noncompete agreements. CONCLUSIONS:Our study demonstrated that noncompete clauses are common within gynecologic oncology. Nearly half of respondents indicated a noncompete clause affected their decision to seek employment. Most respondents reported familiarity with enforceability of these clauses and a negative view of noncompete clauses. The widespread use of noncompete clauses merits further evaluation for the potential impact on physician mobility and the potential to exacerbate geographic disparities within gynecologic oncology.

OBJECTIVE:To determine factors impacting successful bilateral sentinel lymph node mapping and empty nodal packet rate using indocyanine green for uterine cancer staging. METHODS:All women with clinical early-stage uterine cancer who underwent hysterectomy with sentinel lymph node mapping between January 1, 2014, and December 31, 2022, were included. The primary endpoint was the successful bilateral sentinel lymph node mapping rate. Secondarily, we assessed the empty nodal packet rate and identified factors impacting successful sentinel lymph node mapping and empty nodal packet rates. Appropriate statistical tests were used. RESULTS:A total of 2690 patients were included, with a median body mass index of 30.9 kg/m2 (range; 14.0-70.2). Of these, 1893 (70%) had low-grade endometrioid histology, 260 (10%) had high-grade endometrioid histology, and 537 (20%) had non-endometrioid histology. A total of 2499 (93%) cases were completed with minimally invasive surgery, and 191 (7%) via laparotomy. Bilateral sentinel lymph node mapping was successful in 2340 of 2690 patients (87%). The rate of bilateral mapping by body mass index group was 89% for body mass index <30 kg/m2, 86% for body mass index 30 to 39.9 kg/m2, 85% for body mass index 40 to 49.9 kg/m2, 83% for body mass index 50 to 59.9 kg/m2, 69% for body mass index ≥60 kg/m2 (p =.04). A total of 2196 (88%) minimally invasive surgery cases had bilateral mapping compared with 144 (75%) laparotomy cases (p <.001). On multi-variate analysis, age, body mass index, and surgical approach were independently associated with successful sentinel lymph node mapping. Empty nodal packets occurred in 98 of 2633 patients (3.7%). Increasing age and body mass index were associated with a higher empty nodal packet rate. Bilateral mapping increased from 82.5% in 2014 to 91.3% in 2022 (p =.03), whereas the empty nodal packet rate did not change (p =.2). CONCLUSIONS:High rates of successful sentinel lymph node mapping can be achieved across body mass index categories using indocyanine green. A decreased rate was noted in patients with body mass index ≥60 kg/m2 and in cases completed via laparotomy. Sentinel lymph node mapping success rates improved with continued experience.

Background Hepatic involvement in recurrent ovarian cancer has historically been considered a relative contraindication to cytoreductive surgery. Advances in hepatobiliary surgery and multidisciplinary collaboration have expanded the role of liver resection (Farges et al., 2012, Fong et al., 2005, Hiatt et al., 1994, Curley and Glazer, 2025. This video demonstrates the technical considerations and feasibility of hepatic resection during secondary cytoreductive surgery for ovarian cancer. Case A 61-year-old woman with early-stage ovarian cancer and a long disease-free interval developed an isolated hepatic recurrence. Imaging revealed a centrally necrotic mass in the right hepatic lobe with contiguous involvement of the right adrenal gland. Biopsy confirmed metastatic high-grade Müllerian carcinoma. Following systemic chemotherapy with radiographic response, the patient was selected for secondary cytoreductive surgery. Surgical technique The procedure included partial hepatectomy of segments V and VI with en bloc right adrenalectomy and cholecystectomy. Small-volume pelvic disease was removed via low anterior rectal resection, achieving complete gross tumor resection. The video highlights hepatic mobilization, identification of key anatomic landmarks including the hepatoduodenal ligament, Glissonian pedicles, and hepatic venous structures, and parenchymal transection using the clamp-crush technique with vascular control via the Pringle maneuver. Intraoperative ultrasound confirmed vascular anatomy and perfusion of the remnant liver.OutcomeEstimated blood loss was 450 mL with a Pringle time of 25 min. Pathology confirmed metastatic high-grade Müllerian adenocarcinoma with negative hepatic margins. The patient was discharged on postoperative day ten without major postoperative complications. Conclusion Hepatic resection during secondary cytoreductive surgery for ovarian cancer is safe and feasible in carefully selected patients when performed by experienced multidisciplinary teams with detailed knowledge of hepatic anatomy.

Hyperthermic intraperitoneal chemotherapy (HIPEC) is a technique for delivering heated chemotherapy directly into the peritoneal cavity. As peritoneal carcinomatosis is a hallmark of ovarian cancer, intraperitoneal chemotherapy delivery has long been explored as a strategy to maximize local cytotoxicity through enhanced tissue penetration while limiting systemic toxicity. HIPEC, administered as a single hyperthermic infusion at time of cytoreductive surgery, was developed to optimize drug delivery while avoiding the challenges and toxicities associated with multiple cycles of postoperative intraperitoneal chemotherapy. This review synthesizes recent clinical trials evaluating HIPEC in epithelial ovarian cancer, with a focus on defining the optimal disease setting for its use. We also outline perioperative and multidisciplinary considerations for the safe and successful implementation of HIPEC programs and highlight ongoing areas of controversy and future research needed to clarify its role in gynecologic cancers.

Objective: To assess the safety, efficacy, and diagnostic adequacy of ultrasound-guided endometrial sampling (UGES) in patients with known or suspected cervical stenosis. Methods: We reviewed records of patients who underwent UGES (under general anesthesia) due to cervical stenosis at our center between 1999 and 2024. Of 181 patients identified, 155 were included. Key outcomes included success rate, complication rates, and histopathological findings. Results: The most common indications for sampling included postmenopausal bleeding (72 of 155, 46.5%), abnormal imaging findings (40 of 155, 25.8%), abnormal Papanicolaou test (16 of 155, 10.3%), and neocervix creation after treatment for cervical cancer (10 of 155, 6.5%). The overall success rate of UGES in gaining access to the endometrial canal for assessment of underlying pathology was 93.5% (145 of 155). Of 155 patients, 150 underwent UGES for histopathologic diagnosis; 131 (84.5%) provided adequate tissue for pathologic analysis, 15 (10%) had a sample obtained but had insufficient tissue for pathologic analysis, and 4 (2.7%) failed to provide a tissue sample due to inability to access the endometrial cavity. Overall, 109 of 155 samples (70.3%) were benign and 18 (11.6%) were malignant. All 5 procedures performed for therapeutic reasons were successful. Seven patients (4.5%) experienced a minor adverse event from UGES. No major complications occurred. Conclusion(s): UGES appears to be an effective and safe technique for patients with cervical stenosis. We observed a low complication rate and a high diagnostic yield, suggesting UGES may be a valuable alternative for endometrial evaluation in this challenging setting.

Background:In the management of advanced ovarian cancer, complex surgery is often necessary to achieve the goal of complete gross resection. Diaphragm resections and intrathoracic procedures are safe and feasible in select patients with advanced ovarian cancer. In our institution, up to 68% of cytoreductive surgery requires diaphragm peritonectomy, and 11% involve intrathoracic cytoreduction (Kahn et al., 2023, Kahn et al., 2024). The left diaphragm is often involved, requiring a peritonectomy. The most common complications include pleural effusion and pneumothorax, although a rare complication of left-sided diaphragmatic herniation has been reported (Ehmann et al., 2021a, Ehmann et al., 2021b). Methods:With this video, our goal is to introduce the novel idea of placing delayed absorbable mesh to reduce the risk of left diaphragmatic herniation following left diaphragm peritonectomy, as well as demonstrate the surgical technique of polyglactin 910 mesh placement during cytoreductive surgery for ovarian cancer. Results:We demonstrate a left diaphragm mesh placement during cytoreductive surgery for ovarian cancer, with the steps as follows: measure the diaphragmatic defect to the peritoneal edges with muscle fibers in-between and cut the mesh along the contour; suture in the right center and left center of mesh from the intraperitoneal side to the diaphragm side and then out to initially anchor to the diaphragm; continue to throw interrupted stitches circumferentially around the woven mesh beginning from the center and extending outwards. Conclusion:Delayed absorbable mesh placement is a feasible technique that may help reduce the risk of left diaphragmatic herniation following peritonectomy in cytoreductive surgery for ovarian cancer.

Upfront treatment for patients with advanced high-grade serous ovarian cancer (HGSOC) includes a multi-hour cytoreductive surgery. Although the procedure is necessary for maximal tumor cytoreduction, understanding of the biology of systemic and intratumoral responses induced by surgical cytoreduction is limited. Through analysis of matched tumor and normal tissues and peripheral blood collected at multiple time points during cytoreductive surgery in patients with HGSOC, we demonstrate that surgery leads to rapid induction of systemic inflammatory response and activation of inflammatory signaling in the tumor and normal tissue, with interleukin-6 emerging as a dominant inflammatory pathway. A parallel study in a syngeneic murine HGSOC model recapitulated these findings and demonstrated accelerated tumor growth in response to surgery. This study highlights the previously unappreciated impact of specimen collection timing on the tumor signaling networks and provides insights into stress pathways activated by surgery, generating rationale for perioperative therapeutic interventions to reduce protumorigenic effects.

Small cell carcinoma of the ovary, hypercalcemic type (SCCOHT) is a rare and highly aggressive ovarian neoplasm, predominantly affecting young women, often in their second or third decade of life. Despite its distinctive clinical and pathological features, diagnosis is frequently delayed due to overlapping characteristics with other small round blue cell tumors. A hallmark of SCCOHT is the biallelic inactivation of the SMARCA4 gene, which leads to loss of BRG1 protein expression and disrupts epigenetic regulation via the SWI/SNF chromatin-remodeling complex. Unlike many other malignancies, SCCOHT exhibits low mutational burden and diploid karyotype, suggesting that epigenetic dysregulation, rather than genomic instability, is the underlying oncogenic mechanism. Clinically, SCCOHT often presents with nonspecific abdominal or pelvic symptoms and is uniquely associated with paraneoplastic hypercalcemia in up to two-thirds of cases. Diagnosis requires a combination of imaging, laboratory evaluation, histopathology, and immunohistochemistry. Treatment is not standardized but typically involves a multimodal approach, including radical surgery and platinum-based chemotherapy, often with multi-agent regimens. The role of radiotherapy is less well defined but may be considered for local control or palliation. Prognosis remains poor, with high recurrence rates and limited response to salvage therapy. Emerging molecular insights have prompted investigations into targeted therapies and immunotherapy, though clinical data are limited. Given the frequent presence of germline SMARCA4 mutations, genetic counseling is strongly recommended, and ongoing research is essential to improve diagnostic accuracy, personalize treatment, and enhance outcomes for this devastating malignancy.

To evaluate four methods of injecting indocyanine green (ICG) around the adnexa and identify the optimal technique for ovarian sentinel lymph node (SLN) mapping. Patients presenting for management of an adnexal mass were prospectively enrolled. Patients with peritoneal carcinomatosis were excluded. Four injection methods were evaluated: 1) intratubal, 2) paraovarian peritoneum, 3) infundibulopelvic (IP) ligament after resection of the adnexal mass, and 4) IP ligament before resection of the adnexal mass. Two mL of ICG was injected, and at least 10 minutes of transit time was allowed. The ipsilateral and contralateral pelvic and para-aortic lymph node beds were evaluated for ICG uptake. Retroperitoneal nodal resection was performed if indicated. Forty patients were enrolled, 10 in each group. For method 1, 20.0% of SLNs mapped, all to the ipsilateral para-aortic lymph node bed. For method 2, 10.0% mapped, only to the ipsilateral para-aortic lymph node bed. For method 3, 50.0% mapped to the ipsilateral para-aortic lymph node bed (n=3), contralateral para-aortic lymph node bed (n=1), or ipsilateral pelvic lymph node bed (n=1). For method 4, 70.0% mapped to the ipsilateral para-aortic lymph node bed (n=5) or ipsilateral pelvic lymph node bed (n=2). Surgeons reported methods 1 and 2 as cumbersome, and excessive peritoneal staining made SLN identification difficult. No injection-related complications were reported. Injection of ICG into the IP ligament before or after adnexal mass resection led to similar rates of SLN mapping and was deemed feasible by surgeons. Only one SLN was identified contralateral to the adnexal mass, and all but three mapped to the para-aortic region. Injection into the IP ligament should be evaluated in patients with likely adnexal malignancy, including SLN resection.

Objective To compare cytoreductive outcomes of 2 published algorithms used to triage patients with ovarian cancer to primary debulking surgery versus diagnostic laparoscopy or neoadjuvant chemotherapy. Methods This prospective comparative study was conducted from August 1, 2021, to January 31, 2025. All patients with suspected advanced ovarian cancer who were eligible for primary debulking surgery on initial evaluation were identified. Imaging was reviewed utilizing a standardized radiology synoptic report. Data from the synoptic report were combined with clinical factors to determine a Resectability Score using 1 of 2 algorithms, Resectability Score 1.0 (RS1.0) and Resectability Score 2.0 (RS2.0). The algorithms include different clinical and radiologic variables; both generate low- and high-risk scores, with high-risk scores indicating a greater likelihood of suboptimal primary debulking surgery. In high-risk cases, laparoscopic evaluation of resectability was recommended, but management was based on surgeon’s discretion. Results Of 237 patients identified, 200 had primary debulking surgery with final pathology confirming epithelial ovarian carcinoma. Of 144 patients (72%) who underwent laparotomy and primary debulking surgery, 110 (76%) were triaged directly to surgery and 34 (24%) first underwent diagnostic laparoscopy; 120 (83%) had complete gross resection, 135 (94%) had residual disease ≤1 cm, and 9 (6%) had residual disease >1 cm. Of 56 patients (28%) who underwent neoadjuvant chemotherapy, 43 (77%) first underwent diagnostic laparoscopy, and 13 (23%) were triaged directly to chemotherapy. Among all patients, 44 (22%) had high-risk scores using RS1.0 and 54 (27%) using RS2.0. RS2.0 more frequently predicted the ability to and inability to achieve complete gross resection (p > .05). RS2.0 more accurately identified high-risk disease warranting neoadjuvant chemotherapy (p = .035). Most surgeons (73%) preferred RS2.0, citing ease of use and faster calculation time. Conclusions RS2.0 demonstrated favorable predictive accuracy for complete gross resection and was preferred among surgeons. The favorable complete gross resection rate (83%) highlights the value of individualized preoperative triage for primary debulking surgery versus neoadjuvant chemotherapy.

BACKGROUND:To evaluate four methods of injecting indocyanine green (ICG) around the adnexa and identify the optimal technique for ovarian sentinel lymph node (SLN) mapping. METHOD:Patients presenting for management of an adnexal mass were prospectively enrolled. Patients with peritoneal carcinomatosis were excluded. Four injection methods were evaluated: 1) intratubal, 2) paraovarian peritoneum, 3) infundibulopelvic (IP) ligament after resection of the adnexal mass, and 4) IP ligament before resection of the adnexal mass. Two mL of ICG was injected, and at least 10 minutes of transit time was allowed. The ipsilateral and contralateral pelvic and para-aortic lymph node beds were evaluated for ICG uptake. Retroperitoneal nodal resection was performed if indicated. EXPERIENCE:Forty patients were enrolled, 10 in each group. For method 1, 20.0% of SLNs mapped, all to the ipsilateral para-aortic lymph node bed. For method 2, 10.0% mapped, only to the ipsilateral para-aortic lymph node bed. For method 3, 50.0% mapped to the ipsilateral para-aortic lymph node bed (n=3), contralateral para-aortic lymph node bed (n=1), or ipsilateral pelvic lymph node bed (n=1). For method 4, 70.0% mapped to the ipsilateral para-aortic lymph node bed (n=5) or ipsilateral pelvic lymph node bed (n=2). Surgeons reported methods 1 and 2 as cumbersome, and excessive peritoneal staining made SLN identification difficult. No injection-related complications were reported. CONCLUSION:Injection of ICG into the IP ligament before or after adnexal mass resection led to similar rates of SLN mapping and was deemed feasible by surgeons. Only one SLN was identified contralateral to the adnexal mass, and all but three mapped to the para-aortic region. Injection into the IP ligament should be evaluated in patients with likely adnexal malignancy, including SLN resection.

OBJECTIVE:To determine the incidence rate and risk factors associated with the development of diaphragm hernias following left diaphragm procedures at the time of ovarian cancer cytoreduction. METHODS:We retrospectively reviewed data from patients diagnosed with epithelial ovarian, fallopian tube, or primary peritoneal carcinoma who underwent any timeframe of cytoreductive surgery (primary, interval, secondary, tertiary) at our institution from December 2010 to September 2024. Patients were included if they underwent left diaphragm peritonectomy or resection as part of their cytoreductive surgery. The diagnosis of left diaphragm hernia was made by computed tomography of the chest either as an incidental finding during follow-up surveillance or during work-up for symptomatology. We utilized statistical analysis with descriptive proportions with interquartile ranges. RESULTS:A total of 267 patients with ovarian cancer underwent a left diaphragm peritonectomy or resection as part of their cytoreductive surgery at our institution and were included in the study. The overall median age was 62 years (interquartile range; 52-70) and body mass index 24.9 kg/m2 (interquartile range; 22.1-28.5). Overall, 2.2% (6/267) of patients developed a left diaphragmatic hernia, with 33% (2/6) following a primary cytoreduction and 67% (4/6) following an interval cytoreduction; in addition, 83% (5/6) had a splenectomy at the time of cytoreduction. The median time from date of surgery to time of diagnosis was 12.3 months (range; 4.8-24.8). On univariate analysis, the OR of developing a left diaphragm hernia after undergoing splenectomy was 3.2 (p = .24). CONCLUSIONS:Diaphragmatic hernias occurred most often following the setting of concurrent splenectomy; however, splenectomy was not a statistically significant risk factor. These findings provide meaningful insight into the frequency and clinical context in which this complication may arise, addressing a critical knowledge gap in the surgical management of patients with ovarian cancer who require upper abdominal and thoracic procedures.

OBJECTIVE:To evaluate whether the addition of nivolumab to standard neoadjuvant chemotherapy (NACT) in previously untreated advanced-stage epithelial ovarian cancer (EOC) is safe and feasible. METHODS:In this single-institution pilot study (NCT03245892), patients with advanced EOC for whom NACT was considered the most appropriate initial therapy option were treated with intravenous nivolumab 360 mg with standard carboplatin and dose-dense paclitaxel NACT every 3 weeks for 3-6 cycles. Patients then underwent interval cytoreductive surgery (ICS) followed by another 3 cycles of post-operative treatment. RESULTS:Safety results were consistent with the known profile of the study drugs for the 21 patients treated. Seven (33 %) patients had grade 3/4 adverse events attributed to nivolumab, with one case of grade 3 infusion reaction making that patient unevaluable. Fifteen of the 20 (75 % [95 % CI, 50.9-91.3 %]) evaluable patients had complete gross resection at ICS, and 7 of 20 (35 % [95 % CI, 15.4-59.2 %]) achieved an optimal pathologic chemotherapy response score (CRS) of 3. Median progression-free survival was 15.6 months (95 % CI, 11.6-26.4 months), with a 2-year rate of 32.5 % (95 % CI, 12.8-54.1 %). Median overall survival was 50.7 months (95 % CI, 26.5 months-not evaluable), with a 2-year rate of 85 % (95 % CI, 60.4-94.9 %). CONCLUSIONS:The combination of nivolumab with carboplatin and paclitaxel was safe and tolerable. Most patients achieved a complete gross resection and one-third had an optimal CRS at ICS. Further studies are needed to identify patients with ovarian cancer most likely to benefit from upfront immune checkpoint blockade.