Comparing survival of older ovarian cancer patients treated with neoadjuvant chemotherapy versus primary cytoreductive surgery: Reducing bias through machine learning.

OBJECTIVE: To develop and evaluate a multidimensional comorbidity index (MCI) that identifies ovarian cancer patients at risk of early mortality more accurately than the Charlson Comorbidity Index (CCI) for use in health services research. METHODS: We utilized SEER-Medicare data to identify patients with stage IIIC and IV ovarian cancer, diagnosed in 2010-2015. We employed partial least squares regression, a supervised machine learning algorithm, to develop the MCI by extracting latent factors that optimally captured the variation in health insurance claims made in the year preceding cancer diagnosis, and 1-year mortality. We assessed the discrimination and calibration of the MCI for 1-year mortality and compared its performance to the commonly-used CCI. Finally, we evaluated the MCI's ability to reduce confounding in the association of neoadjuvant chemotherapy (NACT) and all-cause mortality. RESULTS: We included 4723 patients in the development cohort and 933 in the validation cohort. The MCI demonstrated good discrimination for 1-year mortality (c-index: 0.75, 95% CI: 0.72-0.79), while the CCI had poor discrimination (c-index: 0.59, 95% CI: 0.56-0.63). Calibration plots showed better agreement between predicted and observed 1-year mortality risk for the MCI compared with CCI. When comparing all-cause mortality between NACT with primary cytoreductive surgery, NACT was associated with a higher hazard of death (HR: 1.13, 95% CI: 1.04-1.23) after controlling for tumor characteristics, demographic factors, and the CCI. However, when controlling for the MCI instead of the CCI, there was no longer a significant difference (HR: 1.05, 95% CI: 0.96-1.14). CONCLUSIONS: The MCI outperformed the conventional CCI in predicting 1-year mortality, and reducing confounding due to differences in baseline health status in comparative effectiveness analysis of NACT versus primary surgery.

Estrogen replacement therapy and non-hormonal medication use among patients with uterine cancer.

OBJECTIVE: As the prognosis for endometrial cancer is excellent, management of the effects of estrogen deprivation has an important influence on quality of life. We examined the trends in the use of estrogen replacement therapy (ERT) and non-hormonal medications among patients with uterine cancer following surgery. METHODS: The MarketScan Database was used to identify patients 18-49 years who underwent hysterectomy plus oophorectomy and those aged 50-75 years who underwent hysterectomy between 2008 and 2020. ERT and non-hormonal treatments of menopause were identified preoperatively and postoperatively. After propensity score balancing, difference-in-differences (DID) analyses were performed to compare the pre-and-postoperative changes in ERT and non-hormonal medication use between groups. The trends in postoperative use of ERT were assessed and tested using Cochran-Armitage trend tests. RESULTS: A total of 19,700 patients with uterine cancer and 185,150 controls were identified. Overall, postoperative ERT use decreased for both age groups and for patients with and without uterine cancer. The DID in ERT use between those with uterine cancer and those with benign pathology after hysterectomy was -37.1% (95% CI, -40.5 to -33.6%) for patients 18-49 years of age and - 10.4% (95% CI, -10.9 to -9.9%) for those 50-75 years. The DID for non-hormonal medication use between those with uterine cancer and those with benign pathology after hysterectomy was 11.2% (95% CI, 7.8 to 14.7%) for younger patients and 3.4% (95% CI, 2.9 to 4.0%) for those 50-75 years. The postoperative new ERT use has been declining over time in patients with uterine cancer in those 18-49 years of age (P = .02) and those 50-75 years of age (P < .001). CONCLUSIONS: The use of ERT is uncommon and has declined over time in patients with uterine cancer. Conversely, non-hormonal medications are more commonly used among patients with uterine cancer.

Cost-effectiveness of lenvatinib plus pembrolizumab versus chemotherapy for recurrent mismatch repair-proficient endometrial cancer after platinum-based therapy.

OBJECTIVE: The recent Study 309-KEYNOTE-775 showed improved survival for lenvatinib plus pembrolizumab compared to chemotherapy in patients with recurrent endometrial cancer. We created a decision model to compare the cost-effectiveness of lenvatinib plus pembrolizumab in patients with recurrent mismatch repair-proficient (pMMR) endometrial cancer who had progressed after first-line chemotherapy. METHODS: A Markov model was created to simulate the clinical trajectory of 10,000 patients with recurrent pMMR endometrial cancer. The initial decision point in the model was treatment with ether lenvatinib plus pembrolizumab or chemotherapy (doxorubicin or dose-dense paclitaxel). Model probabilities, utility values and costs were derived with assumptions drawn from published literature. A cycle length of 3 months and a time horizon of 2 years was used. The effectiveness was calculated in terms of average quality adjusted life years (QALYs) gained. The primary outcome was incremental cost-effectiveness ratios (ICERs), expressed in 2020 US dollars/QALYs. One-way, two-way and probabilistic sensitivity analyses were performed. RESULTS: Chemotherapy was the least costly strategy at $66,693 followed by lenvatinib plus pembrolizumab ($193,590). Lenvatinib plus pembrolizumab resulted in more patients being alive at 2 years (lenvatinib plus pembrolizumab: 367, chemotherapy: 109). Chemotherapy was cost-effective compared with lenvatinib plus pembrolizumab (ICER: $164,493/QALYs). Lenvatinib plus pembrolizumab became cost-effective when its cost was reduced by $1553 per month (7.8% reduction). CONCLUSION: For patients with recurrent pMMR endometrial cancer Lenvatinib plus pembrolizumab is associated with greater survival but is more costly than chemotherapy. The cost of lenvatinib and pembrolizumab would have to be reduced by approximately 7% to be considered cost-effective.

Cost-effectiveness of BRCA1 testing at time of obstetrical prenatal carrier screening for cancer prevention.

BACKGROUND: Improved technologies paired with an increase in access to genetic testing have led to the availability of expanded carrier screening evaluating hundreds of disorders. Currently, most autosomal dominant mutations, such as BRCA1, are not included in expanded carrier assays. Screening pregnant or preconception reproductive-aged women for BRCA1 may present a unique opportunity to perform population-based screening for patients at a time when precancer screening, chemoprevention, and/or risk-reducing surgery may be beneficial. OBJECTIVE: This study aimed to inform clinical decision-making as to whether the universal incorporation of BRCA1 testing at the time of obstetrical prenatal carrier screening is cost-effective. STUDY DESIGN: A decision analysis and Markov model was created. The initial decision point in the model was BRCA1 testing at the time of expanded carrier screening. Model probabilities, cost, and utility values were derived from published literature. For BRCA1-positive patients, the model simulated breast cancer screening and risk-reducing surgical interventions. A cycle length of 1 year and a time horizon of 47 years were used to simulate the lifespan of patients. The setting was obstetrical clinics in the United States, and the participants were a theoretical cohort of 1,429,074 pregnant patients who annually underwent expanded carrier screening. RESULTS: Among our cohort, BRCA1 testing resulted in the identification of an additional 3716 BRCA1-positive patients, the prevention of 1394 breast and ovarian cancer cases, and 1084 fewer deaths. BRCA1 testing was a cost-effective strategy compared with no BRCA1 testing with an incremental cost-effectiveness ratio of $86,001 per quality-adjusted life years. In a 1-way sensitivity analysis, we varied the prevalence of BRCA1 in the population from 0.00% to 20.00% and found that BRCA1 testing continued to be the cost-effective strategy until the prevalence rate was reduced to 0.16%. Multiple additional sensitivity analyses did not substantially affect the cost-effectiveness. CONCLUSION: The addition of BRCA1 testing to obstetrical prenatal carrier screening is a cost-effective management strategy to identify at-risk women at a time when cancer screening and preventive strategies can be effective. Despite the burden of additional genetic counseling, prenatal care represents a unique opportunity to implement population-based genetic testing.

Ambient particulate matter air pollution exposure and ovarian cancer incidence in the USA: An ecological study.

OBJECTIVE: To investigate associations between air particulate matter of ≤2.5 µm in diameter (PM2.5 ) and ovarian cancer. DESIGN: County-level ecological study. SETTING: Surveillance, epidemiology, and end results from a collection of state-level cancer registries across 744 counties. Data from the Environmental Protection Agency's network for PM2.5 monitoring was used to calculate trailing 5- and 10-year PM2.5 county-level values. County-level data on demographic characteristics were obtained from the American Community Survey. POPULATION: A total of 98 751 patients with histologically confirmed ovarian cancer as a primary malignancy from 2000 to 2016. METHODS: Generalised linear regression models were developed to estimate the association between PM2.5 and PM10 levels, over 5- and 10-year periods of exposure, and ovarian cancer risk, after accounting for county-level covariates. MAIN OUTCOME MEASURES: Risk ratios for associations between ovarian cancer (both overall and specifically epithelial ovarian cancer) and PM2.5 levels. RESULTS: For the 744 counties included, the average PM2.5 level from 1990 through 2018 was 11.75 µg/m3 (SD = 3.7) and the average PM10 level was 22.7 µg/m3 (SD = 5.7). After adjusting for county-level covariates, the overall annualised ovarian cancer incidence was significantly associated with increases in 5-year PM2.5 (RR = 1.11 per 10 units (µg/m3 ) increase, 95% CI 1.06-1.16). Similarly, when the analysis was limited to epithelial cell tumours and adjusted for county-level covariates there was a significant association with trailing 5-year PM2.5 exposure models (RR = 1.12 per 10 units increase, 95% CI 1.08-1.17). Likewise, 10-year PM2.5 exposure was associated with ovarian cancer overall and with epithelial ovarian cancer. CONCLUSIONS: Higher county-level ambient PM2.5 levels are associated with 5- and 10-year incidences of ovarian cancer, as measurable in an ecological study.

Tyrosine kinase inhibitor toxicities: A society of gynecologic oncology review and recommendations.

OBJECTIVE: Oral tyrosine kinase inhibitors (TKIs) have new indications for treatment in gynecologic malignancies. These targeted drugs have both unique and overlapping toxicities, which require careful attention and management. New combination therapies with immune-oncology agents have demonstrated promise in endometrial cancer. This review examines common adverse events associated with TKIs and provides readers with an evidence-based review on current uses and strategies for the management of these medications. METHODS: A comprehensive review of the medical literature on TKI use in gynecologic cancer was undertaken by a committee approach. Details of each drug, its molecular target, and relevant data on both clinical efficacy and side effects were compiled and organized for clinical use. Information on drug-related secondary effects and management strategies for specific toxicities, including dose reduction and concomitant medications, were gathered. RESULTS: TKIs can potentially offer improved response rates and durable responses for a group of patients who were previously without an effective standard second-line therapy. The combination of lenvatinib and pembrolizumab represents a more targeted approach to the drivers of endometrial cancer; however, there remains significant drug-related toxicity, and thus dose reduction and dose delay are frequently required. Toxicity management requires frequent check-ins and management strategies to help patients find the highest tolerable dose. TKIs are expensive and patient financial toxicity is as critical a measure of a drug's utility as any drug side effect. Many of these drugs have patient assistance programs, which should be fully utilized to minimize cost. CONCLUSIONS: Future studies are needed to expand the role of TKIs into new molecularly driven groups. Attention to cost, durability of response, and long-term toxicity management is needed to ensure all eligible patients have access to treatment.

Patterns of cervical cancer screening follow-up in the era of prolonged screening intervals.

INTRODUCTION: Little is known as to how prolonged screening recommendations for cervical cancer have affected compliance. OBJECTIVE: We examined compliance with repeat cervical cancer screening among U.S. women aged 30-64 who underwent index screening between 2013 and 2019. STUDY DESIGN: The IBM Watson Health MarketScan Database was used to identify commercially-insured women 30-64 years old who underwent cervical cancer screening from 2013 to 2019. The cohort was limited to women with continuous insurance 12 months before and ≥ 2 months after index testing. Patients with prior hysterectomy, more frequent surveillance needs, or a history of abnormal cytology, histology, or HPV test were excluded. Index screening included cytology, co-testing, or primary HPV testing. Cumulative incidence curves described screening intervals. Compliance was considered if repeat screening occurred 2.5-4 years after index cytology and 4.5-6 years after index co-testing. Cause-specific hazard models examined factors associated with compliance. RESULTS: Of 5,368,713 patients identified, co-testing was performed in 2,873,070 (53.5%), cytology in 2,422,480 (45.1%), and primary HPV testing in 73,163 (1.4%). The cumulative incidence of repeat screening among all women by seven years was 81.9%. Of those who underwent repeat screening, 85.7% with index cytology and 96.6% with index co-testing were rescreened early. Only, 12.2% with index cytology had appropriate rescreening and 2.1% had delayed rescreening. Among the index co-testing group, 3.2% had appropriate rescreening and 0.3% had delayed rescreening. CONCLUSION: Appropriate cervical cancer follow-up screening is highly variable. The cumulative incidence rate of repeat screening was 81.9% and among women rescreened, the vast majority are tested earlier than recommended by current guidelines.

Heterogeneity of outcomes of endometrial cancer patients included in prospective clinical trials.

OBJECTIVE: To assess heterogeneity in five-year overall survival of patients with endometrial cancer using a large retrospective database with cohorts defined by recent prospective clinical trials. METHODS: The National Cancer Database was used to identify patients with endometrial cancer who underwent hysterectomy from 2004 to 2016. The reported inclusion criteria for GOG-249, PORTEC-3, and GOG-258 were used to define the respective cohorts. Five-year overall survival for each cohort was stratified by tumor characteristics and adjuvant therapy regimens. RESULTS: A total of 89,133 patients were identified who would have fulfilled the entry criteria to GOG-249, PORTEC-3, or GOG-258. When stratified by tumor characteristics, irrespective of adjuvant therapy, five-year overall survival ranged from 59.9%-81.7% for patients meeting GOG-249 inclusion criteria, 40.2%-81.8% for patients meeting PORTEC-3 inclusion criteria, and 17.5%-75.0% for those meeting GOG-258 inclusion criteria. Analysis of subgroups by adjuvant therapy regimen revealed significant improvement in five-year overall survival for chemoradiotherapy compared to chemotherapy or radiotherapy alone for endometroid stage III and stage IVA disease and for some stages of serous and clear cell histology. CONCLUSIONS: Recent prospective trials of adjuvant therapy for endometrial cancer have included heterogeneous cohorts of patients based on five-year overall survival rates when the populations are stratified by tumor characteristics. The variation in expected five-year overall survival for subsets of patients may result in underpowered studies or misleading results.

Dostarlimab for recurrent mismatch repair-deficient endometrial cancer: A cost-effectiveness study.

OBJECTIVE: Patients with recurrent endometrial cancer treated with carboplatin and paclitaxel whose disease progresses have few effective treatment options. Based on promising clinical trial data, the anti-programmed cell death 1 (anti-PD-1) antibody dostarlimab was recently granted accelerated approval for endometrial cancer by the US Food and Drug Administration. We developed a decision model to examine the cost-effectiveness of dostarlimab for patients with progressive/recurrent deficient mismatch repair (dMMR) endometrial cancer whose disease has progressed with first-line chemotherapy. DESIGN: Cost-effectiveness study. POPULATION: Hypothetical cohort of 6000 women with progressive/recurrent dMMR endometrial cancer. METHODS: The initial decision point in the Markov model was treatment with dostarlimab, pembrolizumab or pegylated liposomal doxorubicin (PLD). Model probabilities, and cost and utility values were derived with assumptions drawn from published literature. Effectiveness was estimated as average quality-adjusted life years (QALYs) gained. One-way, two-way and probabilistic sensitivity analyses were performed to vary the assumptions across a range of plausible values. MAIN OUTCOME MEASURES: The primary outcome was the incremental cost-effectiveness ratio (ICER). RESULTS: Pegylated liposomal doxorubicin (PLD) was the least costly strategy, at $55,732, followed by dostarlimab ($151,533) and pembrolizumab ($154,597). Based on a willingness-to-pay threshold of $100,000/QALY, PLD was cost-effective compared with dostarlimab, with an ICER of $331,913 per QALY gained for dostarlimab, whereas pembrolizumab was ruled out by extended dominance (less effective, more costly), compared with dostarlimab. In one-way sensitivity analyses, dostarlimab was cost-effective when its cost was reduced to $4905 (52% reduction). These results were robust in a variety of sensitivity analyses. CONCLUSIONS: Dostarlimab is associated with greater survival compared with other treatments for women with recurrent dMMR endometrial cancer. Although the agent is substantially more costly, dostarlimab became cost-effective when its cost was reduced to $5489 per cycle.

Utilization and outcomes of adjuvant therapy for stage II and III uterine leiomyosarcoma.

OBJECTIVES: The optimal adjuvant therapy for uterine leiomyosarcoma (uLMS) remains uncertain. We analyzed the utilization of adjuvant chemotherapy and radiation therapy for stage II and III uLMS and explored the association between use of adjuvant therapy and survival. METHODS: Patients with stage II or III uLMS treated from 2004 to 2016 and recorded in the National Cancer Database were identified. Multivariable regression models were fit to estimate predictors of use of either adjuvant radiation therapy or chemotherapy. To analyze the impact of chemotherapy on all-cause mortality, an inverse probability of treatment weighted (IPTW) propensity score method was used to account for measured confounders, and the receipt of radiation therapy was adjusted in the outcome model. The process was repeated to analyze the impact of radiation therapy on all-cause mortality by using an IPTW propensity score method and adjusting for the receipt of adjuvant chemotherapy. RESULTS: A total of 890 patients were identified. Adjuvant chemotherapy use increased from 62.2% in 2010 to 70.4% in 2016, whereas radiation usage decreased from 26.7% in 2010 to 10.4% in 2016. Patients with stage III (vs. stage II) disease were less likely to receive radiation therapy. After propensity score weighting, chemotherapy was associated with a 30% decreased risk of all-cause mortality in stage III patients (HR 0.70, 95% CI 0.45-0.98) but had no effect on mortality for stage II patients (HR 0.93, 95% CI 0.70-1.20). Radiation therapy was associated with a 26% decreased risk of mortality for stage II tumors (HR 0.74; 95% CI, 0.53-0.99) and a 57% decrease in mortality for stage III disease (HR 0.43, 95% CI 0.18-0.99). CONCLUSIONS: Among women with stage II-III uLMS, use of chemotherapy is increasing while use of radiation therapy is decreasing. Radiation therapy is associated with improved survival in both stage II and III disease, while there was no association between use of adjuvant chemotherapy and survival in stage II patients.

Adoption of minimally invasive surgery after neoadjuvant chemotherapy in women with metastatic uterine cancer.

OBJECTIVE: Utilization of neoadjuvant chemotherapy (NACT) for advanced stage uterine cancer is increasing. We analyzed the use and outcomes of open versus minimally invasive surgery (MIS) for women with stage IV uterine cancer who received NACT and underwent IDS. METHODS: The National Cancer Database was used to identify women with stage IV uterine cancer diagnosed from 2010 to 2017 and treated with NACT. Among women who underwent IDS, overall survival (OS) was compared between those who underwent laparotomy vs a minimally invasive approach. To account for imbalances in confounders, a propensity score analysis using inverse probability of treatment weighting (IPTW) was performed. RESULTS: A total of 1618 women were identified. Minimally invasive IDS was performed in 31.1% and increased from 16.2% in 2010 to 40.4% in 2017 (P < 0.001). More recent year of diagnosis and performance of surgery at a comprehensive cancer center were associated with increased use of MIS (P < 0.05). Women with serous and clear cell tumors, and carcinosarcomas (compared to endometrioid tumors), as well as Medicaid coverage (compared to commercial insurance) were less likely to undergo an MIS approach (P < 0.05). The median OS was 28 months (95% CI 23.7-30.7) and 24.3 months (95% CI 22.3-26.1) for MIS and laparotomy, respectively. After propensity score balancing, there was no association between the use of MIS and survival (HR = 0.90, 95% CI 0.71-1.14). CONCLUSIONS: Among women with stage IV uterine cancer treated with NACT performance of minimally invasive debulking surgery is increasing. Compared to laparotomy, MIS does not appear to negatively impact survival.

Pathologic characteristics, patterns of care, and outcomes of Asian-Americans and Pacific islanders with uterine cancer.

OBJECTIVES: To compare the patterns of care and outcomes of Asian-Americans/Pacific Islanders (AAPI) to non-Hispanic White (NHW) women with uterine cancer, and examine differences across Asian country of origin. METHODS: National Cancer Database was used to identify AAPI and NHW women with uterine cancer diagnosed from 2004 to 2017. Marginal multivariable log-linear regression models and Cox proportional-hazards models were developed to estimate differences in quality-of-care and all-cause mortality between AAPI and NHW women and across AAPI ethnic groups. RESULTS: We identified 13,454 AAPI and 354,693 NHW women. Compared to NHW women, AAPI patients were younger at diagnosis (median age 57 vs. 62 years), had fewer comorbidities, more often had serous or sarcoma histologic subtypes and stage III/IV cancer. AAPI women had a slightly higher rate of receiving pelvic lymphadenectomy for deeply invasive or high-grade tumors (77.6% vs. 74.3%), and a lower rate of undergoing minimally invasive surgery (70.4% vs. 74.8%) for stage I-IIIC tumors. Among patients undergoing hysterectomy, AAPI women had a lower mortality compared with NHW women for cancer stage I/II/III, and a 28% reduction for type I (grade 1 or 2 endometrioid cancers) disease (aHR = 0.72; 95% CI, 0.64-0.81). Among AAPI subgroups, Pacific Islanders had the worst survival across different cancer stage and disease type. CONCLUSION: AAPI women are diagnosed with uterine cancer at a younger age and have more aggressive histologic subtypes and advanced stage than their White counterparts. They have a similar level of quality-of-care as NHW women, and an improved survival for early stage and type I disease.

Long-term outcomes of vaginal hysterectomy for endometrial cancer.

BACKGROUND: Total vaginal hysterectomy (TVH) has been proposed as an alternative to laparoscopic (TLH) and abdominal hysterectomy (TAH), particularly for women with medical comorbidities. We examined the use and long-term outcomes of vaginal hysterectomy for women with early-stage endometrial cancer. METHODS: The Surveillance, Epidemiology, and End Results-Medicare database was used to identify women with stage I-II endometrial cancer treated with primary hysterectomy from 2000 to 2015. Multivariable regression models were developed to examine clinical, demographic, and pathologic factors associated with performance of TVH. The association between route of hysterectomy and cancer-specific and overall survival was examined using multivariable Cox proportional hazards models. RESULTS: A total of 19,212 patients including 837 (4.6%) who underwent TVH were identified. Performance of TVH declined from 4.5% in 2000 to 2.2% in 2015 (P < 0.0001). Compared to patients 65-69 years of age, patients 75-79 years old (aRR = 1.46; 95% CI, 1.19-1.79) and those >80 years old (aRR = 1.60; 95% CI, 1.30-1.97) were more likely to undergo TVH. Women with high grade tumors were less likely to undergo TVH. Five-year overall and cancer specific survivals were similar for TAH, TLH, and TVH. In multivariable models, there was no association between TVH and either cancer-specific survival (HR = 0.89; 95% CI, 0.65-1.22) compared to laparoscopic hysterectomy. CONCLUSION: Use of TVH for stage I and II endometrial cancer has decreased in the U.S. Chronologic age is the greatest predictor of performance of TVH. Performance of TVH does not negatively impact survival for women with early-stage endometrial cancer.

Circulating tumor DNA monitoring for early recurrence detection in epithelial ovarian cancer.

OBJECTIVE: Epithelial ovarian cancer (EOC) is the most lethal gynecologic malignancy. We examined the utility of circulating tumor DNA (ctDNA) as a prognostic biomarker for EOC by assessing its relationship with patient outcome and CA-125, pre-surgically and during post-treatment surveillance. METHODS: Plasma samples were collected from patients with stage I-IV EOC. Cohort A included patients with pre-surgical samples (N = 44, median follow-up: 2.7 years), cohort B and C included: patients with serially collected post-surgically (N = 12) and, during surveillance (N = 13), respectively (median follow-up: 2 years). Plasma samples were analyzed using a tumor-informed, personalized multiplex-PCR NGS assay; ctDNA status and CA-125 levels were correlated with clinical features and outcomes. RESULTS: Genomic profiling was performed on the entire cohort and was consistent with that seen in TCGA. In cohort A, ctDNA-positivity was observed in 73% (32/44) of presurgical samples and was higher in high nuclear grade disease. In cohort B and C, ctDNA was only detected in patients who relapsed (100% sensitivity and specificity) and preceded radiological findings by an average of 10 months. The presence of ctDNA at a single timepoint after completion of surgery +/- adjuvant chemotherapy and serially during surveillance was a strong predictor of relapse (HR:17.6, p = 0.001 and p < 0.0001, respectively), while CA-125 positivity was not (p = 0.113 and p = 0.056). CONCLUSIONS: The presence of ctDNA post-surgically is highly prognostic of reduced recurrence-free survival. CtDNA outperformed CA-125 in identifying patients at highest risk of recurrence. These results suggest that monitoring ctDNA could be beneficial in clinical decision-making for EOC patients.

Patterns of cervical cancer screening among Medicaid beneficiaries.

OBJECTIVE: Cervical cancer screening guidelines have evolved over time with the incorporation of human papillomavirus (HPV) testing along with cytology. Current screening guidelines recommend cytological screening every 3 years or HPV testing with or without cytology every 5 years in women age 30-65 years. We examined the use of cervical cancer screening among average-risk Medicaid beneficiaries. DESIGN: Retrospective cohort study. POPULATION: Women age 30-64 years at average risk for cervical cancer who underwent cervical cancer screening with cytology, co-testing or primary HPV testing from 2013 to 2016. METHODS: The IBM Watson Health Multi-State Medicaid MarketScan Database was used. Subsequent screening rates within 3 years of the index test were examined. MAIN OUTCOME MEASURE: The rate of repeat cervical cancer screening was analysed using a cumulative incidence function. RESULTS: A total of 265 083 patients were identified. Overall, 43.1% (n = 114 312) had index co-testing, 55.2% (n = 146 309) had cytology and 1.7% (n = 4462) had primary HPV testing. The cumulative incidence of early, repeat cervical cancer screening was 3.9% at 12 months, 22.7% at 24 months and 33.3% at 36 months. During the period from 12 to 24 months after follow up, 20.9% of women underwent repeat screening while 19.4% underwent repeat screening 24-36 months after the index test. Among women who did not undergo repeat cervical cancer screening, a yearly gynaecological examination was performed in only 16 627 (10.7%) during year 2 and in 11 116 (8.8%) during year 3. CONCLUSION: Among average-risk Medicaid beneficiaries, cervical cancer screening is frequently overused. Women who do not undergo cervical cancer screening are unlikely to undergo routine gynaecological examination. TWEETABLE ABSTRACT: Among average-risk Medicaid beneficiaries, cervical cancer screening is frequently overused.

Uptake and outcomes of sentinel lymph node mapping in women undergoing minimally invasive surgery for endometrial cancer.

OBJECTIVE: To examine the patterns and outcomes of sentinel lymph node (SLN) assessment in women with endometrial cancer. DESIGN: Retrospective cohort study. SETTING: United States inpatient and outpatient hospital services. POPULATION: Women with endometrial cancer who underwent a laparoscopic or robotic-assisted hysterectomy. METHODS: The Perspective Database from 2012 to 2018 was used. Performance of lymph node dissection was classified as SLN mapping, lymph node dissection or no nodal evaluation. Adjusted regression models were developed to examine the association between SLN mapping and morbidity and cost. MAIN OUTCOME MEASURES: Utilisation rates, morbidity and cost of both lymph node dissection and SLN mapping. RESULTS: Among 45 381 patients, SLN mapping was performed for 7768 patients (17.1%), lymph node dissection was performed for 23 214 patients (51.2%) and no lymphatic evaluation was performed for 14 399 patients (31.7%). SLN mapping increased from 1.8% in 2012 to 35.3% in 2018, whereas the rate of lymph node dissection decreased from 63.5% to 39.1% (p < 0.001). Among women who underwent nodal evaluation, residence in the west, White race and use of robotic-assisted hysterectomy were associated with SLN mapping (p < 0.05 for all). The complication rate was 5.9% for SLN mapping, compared with 7.3% in those that underwent lymph node dissection (aRR 0.85, 95% CI 0.77-0.95). The median hospital costs for women who underwent SLN mapping ($10 479) and lymph node dissection ($10 747) were higher than for those who did not undergo nodal assessment ($9149) (p < 0.001). CONCLUSIONS: The performance of SLN mapping is increasing for endometrial cancer. Compared with lymph node dissection, SLN mapping is associated with lower morbidity. SLN mapping significantly increases the costs compared with hysterectomy alone. TWEETABLE ABSTRACT: SLN mapping is increasing rapidly for endometrial cancer and is associated with decreased perioperative morbidity.

Phase I trial of daily subcutaneous SPL-108 injections in combination with paclitaxel in patients with platinum resistant CD44+ advanced ovarian epithelial cancer.

OBJECTIVE: Preclinical evidence and early clinical trials have demonstrated the activity of SPL-108, a targeted agent that inhibits CD44 mediated induction of multidrug resistance specifically to paclitaxel and platinum agents. We conducted a phase I, open label, dose escalation study of the safety and tolerability of the combination of SPL-108 with weekly paclitaxel in patients with platinum resistant CD44+ ovarian, primary peritoneal, or fallopian tube cancer. METHODS: Patients with platinum resistant histologically proven epithelial ovarian, primary peritoneal, or fallopian tube cancers and measurable disease according to RECIST (Response Evaluation Criteria in Solid Tumours) version 1.1 were selected. Tumors were tested for CD44 expression for eligibility, defined as strong (+++) or moderate (++) staining in ≥20% of the tumor tissue or diffuse + staining. Patients were treated with daily and then twice daily SPL-108 subcutaneous injections and weekly intravenous paclitaxel on days 1, 8, and 15 of a 28 day cycle. Endpoints included safety, determination of maximum tolerated dose, and efficacy. Tumors underwent comprehensive genomic profiling, and cell lines and western blotting were used to study markers of response. RESULTS: We screened 16 patients, and 14 were enrolled based on CD44+ expression. A total of 86% of patients had high grade serous tumors and all had received multiple prior therapies. There were no grade 4-5 toxicities. One patient had grade 3 peripheral sensory neuropathy attributed to paclitaxel and one patient developed presumed colonic perforation attributed to the study drug. No dose reductions or treatment discontinuations were required. All patients tolerated the maximum planned dose; no maximum tolerated dose was reached. Overall response rate was 36%; 5 (36%) patients had partial response and 5 (36%) patients had stable disease. CONCLUSIONS: The combination of SPL-108 with weekly paclitaxel was safe and well tolerated. Encouraging antitumor activity was observed, with 72% of patients deriving a clinical benefit. TRIAL REGISTRATION: NCT03078400.

Perioperative Morbidity and Mortality of Patients With COVID-19 Who Undergo Urgent and Emergent Surgical Procedures.

OBJECTIVE: To evaluate the perioperative morbidity and mortality of patients with COVID-19 who undergo urgent and emergent surgery. SUMMARY BACKGROUND DATA: Although COVID-19 infection is usually associated with mild disease, it can lead to severe respiratory complications. Little is known about the perioperative outcomes of patients with COVID-19. METHODS: We examined patients who underwent urgent and emergent surgery at 2 hospitals in New York City from March 17 to April 15, 2020. Elective surgical procedures were cancelled throughout and routine, laboratory based COVID-19 screening was instituted on April 1. Mortality, complications, and admission to the intensive care unit were compared between patients with COVID-19 detected perioperatively and controls. RESULTS: Among 468 subjects, 36 (7.7%) had confirmed COVID-19. Among those with COVID-19, 55.6% were detected preoperatively and 44.4% postoperatively. Before the routine preoperative COVID-19 laboratory screening, 7.7% of cases were diagnosed preoperatively compared to 65.2% after institution of screening (P = 0.0008). The perioperative mortality rate was 16.7% in those with COVID-19 compared to 1.4% in COVID-19 negative subjects [aRR = 9.29; 95% confidence interval (CI), 5.68-15.21]. Serious complications were identified in 58.3% of COVID-19 subjects versus 6.0% of controls (aRR = 7.02; 95%CI, 4.96-9.92). Cardiac arrest, sepsis/shock, respiratory failure, pneumonia, acute respiratory distress syndrome, and acute kidney injury were more common in those with COVID-19. The intensive care unit admission rate was 36.1% in those with COVID-19 compared to 16.4% of controls (aRR = 1.34; 95%CI, 0.86-2.09). CONCLUSIONS: COVID-19 is associated with an increased risk for serious perioperative morbidity and mortality. A substantial number of patients with COVID-19 are not identified until after surgery.

Precision medicine testing in ovarian cancer: The growing inequity between patients with commercial vs medicaid insurance.

INTRODUCTION: Precision medicine technologies have significant impact in the care of patients with ovarian cancer. Compared to affluent patients, socioeconomically vulnerable patients are less likely to have access to this testing. There is little data that demonstrate this inequity over time. METHODS: We used the IBM Truven Health MarketScan Research Database to identify patients in the United States who underwent surgery for ovarian cancer between 2011 and 2017. The presence of claims for precision medicine testing within six months of surgery was assessed for each patient. Precision medicine testing included both molecular genetic testing (BRCA limited or full sequencing, somatic and germline testing) as well as ancillary pathology tests (immunohistochemistry, microsatellite instability). Demographic data was extracted. RESULTS: We identified 27,181 patients who met eligibility. Of these, 88.6% had commercial insurance, and 11.4% had Medicaid. While the proportion of patients who underwent precision medicine testing increased over time for both cohorts (47.0% to 66.6% for commercially insured, 41.4% to 57.6% for Medicaid insured, p < 0.0001), the inequity in testing rates widened (5.6% disparity to 9.0%, p < 0.0001). This was driven by growing inequity in germline and somatic genetic testing (7.6% disparity to 21.3%, p < 0.0001). CONCLUSIONS: There is widening inequity in precision medicine testing rates between commercially insured and Medicaid insured poate patients with ovarian cancer.

Trends in ovarian conservation and association with survival in premenopausal patients with stage I leiomyosarcoma.

OBJECTIVE: To explore trends of ovarian conservation (OCN) over time in young women with early stage leiomyosarcoma (LMS) and examine the association between OCN and survival. METHODS: Patients under the age of 50 who were diagnosed with stage I LMS who underwent hysterectomy with and without oophorectomy between 2010 and 2016 were identified in the National Cancer Database (NCDB). Performance of oophorectomy vs. OCN was determined using surgery codes. Trends of OCN were reported. Multivariable regression models were fit to estimate predictors of OCN. An inverse probability of treatment weighted propensity score method was used to examine the association between all-cause mortality and OCN. RESULTS: Overall, 225 patients (28%) underwent OCN. Rates of OCN decreased from 41.2% (2010) to 14.3% (2016); this finding was consistent across age groups: <35, 35-39, 40-44, and 45-49 years. Race, insurance, and stage did not affect performance of OCN. Women with poorly differentiated tumors were less likely to undergo OCN compared to well-differentiated tumors (aRR 0.59; 95% CI 0.40-0.86). After propensity score weighting, there was no association between OCN and mortality (HR 1.19, 95% CI 0.80-1.77). Five-year survival for the OCN group was 67.1% (95% CI 59.8-75.2%) compared to 72.2% for the oophorectomy group (95% CI 67.2-77.5%). CONCLUSIONS: OCN for early stage LMS in premenopausal women has decreased over time. There was no association between OCN and mortality among women with stage I LMS. OCN should be considered in premenopausal women with stage I LMS given the health benefits.