Faculty Bibliography

OBJECTIVE: To evaluate the results of multigene panel testing among Ashkenazi Jewish compared with non-Ashkenazi Jewish patients. METHODS: We reviewed the medical records for all patients who underwent multigene panel testing and targeted BRCA1/2 testing at a single institution between 6/2013-1/2015. Clinical actionability for identified pathogenic mutations was characterized based on the National Comprehensive Cancer Network (NCCN) guidelines and consensus statements and expert opinion for genes not addressed by these guidelines. RESULTS: Four hundred and fifty-four patients underwent multigene panel screening, including 138 Ashkenazi Jewish patients. The median patient age was fifty-two years. Three hundred and fifty-four patients (78%) had a personal history of cancer. Two hundred and fifty-one patients had breast cancer, 49, ovarian cancer, 26, uterine cancer and 20, colorectal cancer. We identified 62 mutations in 56 patients and 291 variants of uncertain significance in 196 patients. Among the 56 patients with mutations, 51 (91%) had actionable mutations. Twenty mutations were identified by multigene panels among Ashkenazi Jewish patients, 18 of which were in genes other than BRCA1/2. A review of targeted BRCA1/2 testing performed over the same study period included 103 patients and identified six mutations in BRCA1/2, all of which occurred in Ashkenazi Jewish patients. Among all Ashkenazi Jewish patients undergoing genetic testing, 25/183 (14%) had a mutation, 24/25 of which were actionable (96%) and 17/25 patients (68%) had mutations in non BRCA1/2 genes. CONCLUSIONS: With the rapid acceptance of multigene panels there is a pressing need to understand how this testing will affect patient management. While traditionally many Ashkenazi Jewish patients have undergone targeted BRCA1/2 testing, our data suggest consideration of multigene panels in this population as the majority of the results are clinically actionable and often in genes other than BRCA1/2.

Background: Lynch syndrome (LS) accounts for 2-6% of all endometrial cancers (EC), and women with a germline mutation in the mismatch repair (MMR) genes (MLH1, MSH2, MSH6, and PMS2) have an average lifetime risk of EC of 40%. As with breast and ovarian cancer syndromes, there are likely other genes implicated in the development of EC outside of the MMR genes. Multi-gene panel testing (MGPT) with next generation sequencing (NGS) allows for simultaneous analysis of numerous genes.We sought to evaluate the characteristics and incidence of gene mutations in women with newly diagnosed EC. Methods: We conducted a review of EC patients diagnosed from 6/2013 to 12/2016 who had MGPT at our institution. Demographics, family history, genetic testing results, and tumor characteristics were collected and analyzed using chi tests. Results: Of the 129 patients who had MGPT, 13 (10%) had a mutation and only 5 (38%) were in patients < 50 years old. The median age of EC diagnosis is 55 (31-100) years and median BMI = 27.5 (21-59). Majority were stage 1, 76 (59%) and grade 1, 50 (39%). Patients with additional primary cancers, breast or colon were not more likely to have a mutation. However, patients with a family history of gynecologic cancer were more likely to have a mutation identified, 10 (77%) mutation vs no mutation 34 (29%), p = 0.003. Among all patients tested, 8 (6%) had a mutation in LS genes, and 6 (5%) had mutations in other genes (BRCA1, BRCA2, RAD51C, MUTYH, CHEK2); 1 (0.8%) had both MSH2 and CHEK2 mutation. Three patients had prior testing for breast cancer; 2 were found to have a BRCA1/ 2 mutation and the other was on Tamoxifen and BRCA negative. IHC was performed on 7 of 13 patients, and 5 (71%) had a loss of MMR protein expression. Variants of uncertain significance were noted in 35/129 (27%) of patients tested. Conclusions: Majority of EC patients with a mutation detected with NGS were > age 50. We identified additional new mutations in non-LS genes including, CHEK2, RAD51C, and MUYTH with MGPT. These accounted for 29% of the mutations and would have not been not detected using classic LS gene testing. These genes are implicated in breast, ovary or colon cancer. MGPT testing is feasible and useful in identifying additional actionable gene mutations

Objective: A minimally invasive surgical approach has proven to decrease perioperative complications and shorten duration of hospital stay; however, there are limited data evaluating the use of robotic surgery and early hospital discharge in the elderly population. As age is a well-known, independent risk factor for perioperative morbidity and gynecologists treat many elderly patients, we sought to evaluate outcomes in robotic-assisted gynecologic surgery in elderly compared to younger patients. Method: We reviewed the medical records of all patients undergoing robotic-assisted gynecologic surgery at a single institution from March 2015 to May 2016. We compared perioperative outcomes and discharge timing for patients b65 years to those N65 years using univariate and multivariate analyses. Results: A total of 1,143 patients were included (1,049 b65 years, 94 N65 years). Elderly patients had a statistically significant higher BMI and ASA class and were more likely to have malignancy as the indication for surgery when compared to younger patients. Surgery in elderly patients was longer (198 vs 166 minutes, P b 0.001) with a higher incidence of intraoperative complications (9% vs 3%, P = 0.003). Median hospital stay was 6.3 (range 2-1,386) vs 7.8 hours (range 2.5-103) in the b65 and N65 groups, respectively. Eighty-six percent of younger patients were discharged on the day of surgery versus 62% of elderly patients. Two percent of younger patients and 11% of elderly patients were admitted for N23 hours. Differences in hospital admission status remained significant on multivariate analysis adjusting for BMI, ASA class, malignant indication for surgery, length of surgery, and intraoperative complications. There was no difference in postoperative complications, emergency room visits, or hospital readmissions between elderly and younger patients. (See Table 1.) Conclusion: Ninety percent of elderly patients were discharged within 23 hours of surgery without a significant difference in postoperative complications, emergency roomvisits, or readmissions when compared to younger patients. Although the risks of surgery are increased in the elderly population and hospital admissions were greater, roboticassisted gynecologic surgery and early hospital discharge were safe in our cohort of elderly patients

Objective: Approximately 30% of ovarian cancer is attributable to germline mutations, and genetic assessment is recommended for all women with ovarian cancer. However, only 15-30% are currently being offered genetic evaluation. We sought to determine whether a patient-centered, facilitated genetics referral pathway, whereby all newly diagnosed ovarian cancer patients are contacted by a genetics navigator to schedule genetic assessment as part of routine care, could increase rates of genetic counseling and uptake of testing. Method: Patients with epithelial ovarian cancer were referred for genetic assessment by their gynecologic oncologist within 6 weeks of diagnosis and consented for participation in our institutional review board-approved facilitated genetics pathway. Enrolled patients were contacted by a genetics navigator to schedule a genetic counselor appointment within 6 weeks. Patients who did not schedule or missed sessions were recontacted by the navigator. The genetic counselors offered pre-and post-test counseling and multigene panel testing. Primary outcome was feasibility of this pathway as defined by presentation for genetic assessment or declining genetic evaluation. Results: From October 2015 to July 2016, 50 patients were enrolled. Thirty-six patients (72%) underwent genetic assessment and, of these patients, 34 (94%) had genetic testing. Three patients (6%) are currently scheduled for appointments. Eleven patients (22%) did not undergo genetic assessment for the following reasons: not interested (4), not feeling well (2), missed appointment (2), nervous about testing (1), unable to see genetics counselors within 6 weeks (1), and death (1). Median time from diagnosis to genetics appointment was 13 days (range 0-53). Among the 32 patients for whom results are available, 7 (22%) had pathogenic mutations (BRCA1, 4; BRCA2, 3). Conclusion: The genetic testing pathway we present, characterized by facilitated referral to genetic counselors at time of ovarian cancer diagnosis, is both effective and efficient, resulting in genetic assessment of 72% of patients with newly diagnosed ovarian cancer, testing in 94% of these patients, and discovery of pathogenic mutations in 22% of those tested. Because germline mutations have both prognostic and therapeutic implications, the time of diagnosis may present an idealwindow to offer genetic testing

Objective: With the growing use of robotic surgery, there is an increased occurrence of port-site hernias requiring operative intervention. Currently there is limited literature, and prior studies have failed to find surgical or patient-related risk factors. We sought to identify patient and surgical risk factors, evaluate clinical presentation, and report management of this postoperative complication at a high-volume multispecialty robotic surgical center. Method: All robotic surgeries performed at a single institution from September 1, 2010, to September 1, 2015, were included. Univariate analysis was used to compare patient demographics and medical conditions for those who did and did not develop port-site hernias. Results: A total of 4,858 robotic surgeries were completed during the study period. A total of 37 (0.7%) port-site hernias requiring operative intervention were identified following urologic (23/1,888, 1.2%), gynecologic (13/2,661, 0.5%), and general surgery (1/309, 0.3%) procedures. Hernias occurred at the umbilical (n = 23) and 8-mm lateral port sites (n = 14). Only umbilical ports were closed under direct visualization. Median time from surgery to hernia diagnosis was 201 days (range 2-975). Presentation included bulge symptoms (n = 29) and nausea/vomiting (n = 6). The herniated contents included bowel/omentum (n =19), fat (n =14), or empty sac (n = 4). All cases were managed surgically, 21 with laparoscopy and 16 with laparotomy, with presentation within 30 days necessitating urgent surgery (n = 6). A total of 7/37 patients had complications from reoperation (bowel resection, n = 3; abscess formation, n=2; blood transfusion, n = 1). There was no difference between patients who did and did not develop a port-site hernia with regards to age, gender, BMI, smoking status, hypertension, diabetes, rheumatologic disease, HIV, prior hernia, or cancer diagnosis. (See Table 1.) Conclusion: Port-site hernias necessitating operative intervention following robotic surgery are rare, occurring in 0.7% of patients in our cohort. We found no patient or surgical variable to be predictive of this complication. Hernias occurred at both the umbilical and lateral ports. Despite needing a second surgery, all patients recovered and did not suffer significant long-term morbidity

Objective: Niraparib is a highly selective PARP1/2 inhibitor that induces synthetic lethality in tumor cells with homologous recombination DNA repair deficiencies (HRD). Niraparib is administered orally once daily (QD) and may be taken without regard to meals, because food does not significantly affect the absorption or metabolism of niraparib. In a phase 1 study, niraparib was clinically active and well tolerated in patients with ovarian cancer. To address the need for novel therapies for recurrence, a randomized, doubleblind phase 3 study is evaluating niraparib as maintenance treatment in patients with stage III/IV ovarian, fallopian tube, or primary peritoneal cancer who responded to front-line platinum-based chemotherapy but are at high risk for progressive disease, based on one of the following: FIGO stage IV regardless of outcome of surgery, FIGO stage III and presence of macroscopic residual disease after primary surgery, or use of neoadjuvant chemotherapy regardless of residual disease after interval debulking surgery. Method: This multicenter trial is enrolling =305 patients with ovarian, fallopian tube, or peritoneal cancer. Eligible females are required to have histologically diagnosed stage III/IV cancer, tumor available for HRD test (myChoiceTM HRD test), normal or N90% decrease in CA-125 levels during front-line therapy, and no prior PARP inhibitor use. Patients who have undergone debulking surgery must have residual disease after primary surgery unless they received neoadjuvant therapy. Patients with a known history of myelodysplastic syndrome (MDS) or a baseline cytogenetic test result with a risk for a diagnosis of MDS or acute myeloid leukemia are excluded. Patients must have completed >=4 cycles of platinumbased therapy and had a complete or partial RECIST response. Patients are randomized (2:1) to receive niraparib 300 mg QD or matched placebo in 28-day cycles. Tumors are assessed every 12 weeks per RECIST v1.1. The primary endpoint is progression-free survival (PFS) by blinded central review. After 258 PFS events, the study will have 90% power (one-sided a = 0.025) to detect an improvement in median PFS with HR = 0.65. Secondary endpoints are evaluation of overall survival, patient-reported outcomes, safety and tolerability, and time to progression on next therapy. Results: Not applicable (trial in progress). Conclusion: Not applicable (trial in progress)

Objective: Differences in complication rates and outcomes are expected when comparing surgery for benign and malignant indications; however, there are limited data addressing this issue in robotic-assisted gynecologic surgery. Because this distinction and its ramifications can influence patient counseling, surgical planning, and reimbursement as we move to value-based payment models, we sought to evaluate the perioperative outcomes for women undergoing gynecologic robotic-assisted surgery for benign versus malignant indications. Method:We reviewed the medical records of all patients undergoing robotic-assisted gynecologic surgery at a single institution from March 2015 to May 2016. Perioperative outcomes were evaluated using univariate and multivariate regression analysis to compare complications and outcomes for benign versus malignant surgical indications. Results: A total of 1,142 patientswere included (benign 962, malignant 180). Malignant cases were significantly older (58 vs 42 years, P b 0.001) with more medical comorbid conditions and higher BMI. There was no difference in estimated blood loss or intraoperative complications. Malignant cases had longer surgical time (3.6 vs 2.6 hours, P b 0.001), higher rates of conversion to laparotomy (2% vs 0.4%, P = 0.008), postoperative complications (14% vs 9%, P=0.05), and hospital readmission within 6 weeks (7% vs 3%, P = 0.01). Postoperative complications included fever (benign 3%, malignant 5%), urinary tract infection (2%, 5%), prolonged/severe pain (2%, 2%),wound infection (2%, 2%), abscess (1%, 1%), other infection (1%, 2%), port-site hernia (1%, 2%), small bowel obstruction (0%, 1%), and deep vein thrombosis (0%, 1%). On multivariate analysis adjusting for age, BMI, and medical comorbidity condition, malignancy remained associated with higher rates of conversion and hospital readmission. (See Table 1.) Conclusion: Robotic surgery performed for gynecologic malignancy is associated with higher rates of conversion to laparotomy and hospital readmission compared to robotic surgery performed for benign disease. This information is very important for physician and patient surgical expectations and should be accounted for when determining models of value-based reimbursement

Objective: With the increasing age of the population, more elderly patients undergo gynecologic surgery. While multiple studies demonstrate the advantages of minimally invasive surgery (MIS), including reduced morbidity and hospital stay, there is a paucity of data in the elderly population, especially in patients older than 80 years. We sought to evaluate outcomes among elderly patients undergoing MIS by a gynecologic oncologist. Method: We reviewed the medical records of patients 65 years of age and older who underwent MIS (robotic, laparoscopic, or vaginal surgery) by a gynecologic oncologist at a single institution between 2009 and 2016. We compared outcomes among "younger-elderly" (65-79) and "older-elderly" (80 and older) patients. Results: A total of 298 patients underwent MIS by a gynecologic oncologist (younger-elderly, 268; older-elderly, 29). The median age was 69 years (range 65-79) in the younger-elderly and 83 years (range 80-93) in the older-elderly. The older-elderly had more medical comorbidity conditions (median Charlson index 7 vs 5, P b 0.001). There was no significant difference between the 2 groups with respect to surgical approach, underlying malignancy, conversion to laparotomy and hospital stay (Table 1). Two hundred and nineteen (81%) younger-elderly patients and 22 (76%) older-elderly patients had ambulatory surgery, defined as an admission of less than 24 hours (P = 0.31). Surgical complications were rare (8/298, 3%), and there was no difference between the younger-and olderelderly patients in rates of complications or 30-and 90-day readmissions. Conclusion: We found MIS with early discharge to be a safe approach in elderly patients undergoing surgery by gynecologic oncologists. While patients older than 80 years were poorer surgical candidates, there were no differences in surgical outcomes or hospital stay when compared to the younger-elderly group. Elderly age should not prohibit consideration of MIS with early hospital discharge

Objective: Borderline ovarian tumors (BOT) are not uncommon and affect a younger age group than epithelial ovarian cancers. The majority of BOT are stage I at presentation and have an excellent long-term prognosis; appropriately, fertility-sparing surgery is the norm. However, there is a wide variance in the reported rate of recurrence following fertility-sparing surgery, and the risk of reoperation secondary to increased surveillance has not been well reported. We sought to evaluate the rate of recurrence and reoperation among women with BOT. Method: Patients were identified using pathology results and electronic medical records from two medical centers. Demographic, clinical, and pathologic data were obtained from medical records. Comparative analyses were performed to determine the effects of histology and type of surgery on risk of recurrence and reoperation. Statistical tests were performed with R Studio v0.99.442. Results: Between January 1, 2005, and December 31, 2015, we identified 134 patients who underwent surgical management for BOT. Median age at diagnosis was 40, and 82% presented with stage I disease. The majority of index cases were either serous (70/134, 52%) or mucinous (46/134, 34%) BOT. Median follow-up was 22 months. Following initial surgery, 86 patients (65%) had at least 1 ovary in situ. Of these patients, 34 (40%) underwent additional ovarian surgery: 21/34 (64%) were diagnosed with BOT; 9/34 (27%) had benign disease; and 2/34 (6%) were diagnosed with cancer. Median time to the second surgery was 18 months following the initial surgery. Of the 86 patients with fertility-sparing surgery, 20 (23%) delivered a live infant. The risk of reoperation or recurrence was not influenced by histology (serous, mucinous, endometrioid) or type of surgery (unilateral oophorectomy, ovarian cystectomy). The risk of reoperation was increased with increasing stage of BOT: stage I = 23/90 (25%) versus stage III = 3/8 (37.5%) (P= 0.043). Conclusion: Patients who undergo fertility sparing surgery for BOT are at high risk of both recurrent disease and reoperation for benign lesions. There is no association between histology or type of surgery and risk of recurrence or reoperation. Patients should be counseled about these risks at the time of initial surgery