Deep Learning-Based Dynamic Risk Prediction of Venous Thromboembolism for Patients With Ovarian Cancer in Real-World Settings From Electronic Health Records.

PURPOSE: Patients with epithelial ovarian cancer (EOC) have an elevated risk for venous thromboembolism (VTE). To assess the risk of VTE, models were developed by statistical or machine learning algorithms. However, few models have accommodated deep learning (DL) algorithms in realistic clinical settings. We aimed to develop a predictive DL model, exploiting rich information from electronic health records (EHRs), including dynamic clinical features and the presence of competing risks. METHODS: We extracted EHRs of 1,268 patients diagnosed with EOC from January 2007 through December 2017 at the National Cancer Center, Korea. DL survival networks using fully connected layers, temporal attention, and recurrent neural networks were adopted and compared with multi-perceptron-based classification models. Prediction accuracy was independently validated in the data set of 423 patients newly diagnosed with EOC from January 2018 to December 2019. Personalized risk plots displaying the individual interval risk were developed. RESULTS: DL-based survival networks achieved a superior area under the receiver operating characteristic curve (AUROC) between 0.95 and 0.98 while the AUROC of classification models was between 0.85 and 0.90. As clinical information benefits the prediction accuracy, the proposed dynamic survival network outperformed other survival networks for the test and validation data set with the highest time-dependent concordance index (0.974, 0.975) and lowest Brier score (0.051, 0.049) at 6 months after a cancer diagnosis. Our visualization showed that the interval risk fluctuating along with the changes in longitudinal clinical features. CONCLUSION: Adaption of dynamic patient clinical features and accounting for competing risks from EHRs into the DL algorithms demonstrated VTE risk prediction with high accuracy. Our results show that this novel dynamic survival network can provide personalized risk prediction with the potential to assist risk-based clinical intervention to prevent VTE among patients with EOC.

Impact of insomnia on ovarian cancer risk and survival: a Mendelian randomization study.

BACKGROUND: Insomnia is the most common sleep disorder in patients with epithelial ovarian cancer (EOC). We investigated the causal association between genetically predicted insomnia and EOC risk and survival through a two-sample Mendelian randomization (MR) study. METHODS: Insomnia was proxied using genetic variants identified in a genome-wide association study (GWAS) meta-analysis of UK Biobank and 23andMe. Using genetic associations with EOC risk and overall survival from the Ovarian Cancer Association Consortium (OCAC) GWAS in 66,450 women (over 11,000 cases with clinical follow-up), we performed Iterative Mendelian Randomization and Pleiotropy (IMRP) analysis followed by a set of sensitivity analyses. Genetic associations with survival and response to treatment in ovarian cancer study of The Cancer Genome Atlas (TCGA) were estimated controlling for chemotherapy and clinical factors. FINDINGS: Insomnia was associated with higher risk of endometrioid EOC (OR = 1.60, 95% CI 1.05-2.45) and lower risk of high-grade serous EOC (HGSOC) and clear cell EOC (OR = 0.79 and 0.48, 95% CI 0.63-1.00 and 0.27-0.86, respectively). In survival analysis, insomnia was associated with shorter survival of invasive EOC (OR = 1.45, 95% CI 1.13-1.87) and HGSOC (OR = 1.4, 95% CI 1.04-1.89), which was attenuated after adjustment for body mass index and reproductive age. Insomnia was associated with reduced survival in TCGA HGSOC cases who received standard chemotherapy (OR = 2.48, 95% CI 1.13-5.42), but was attenuated after adjustment for clinical factors. INTERPRETATION: This study supports the impact of insomnia on EOC risk and survival, suggesting treatments targeting insomnia could be pivotal for prevention and improving patient survival. FUNDING: National Institutes of Health, National Cancer Institute. Full funding details are provided in acknowledgments.

Association between BMI and oncologic outcomes in epithelial ovarian cancer: a predictors-matched case-control study.

OBJECTIVE: We aimed to study the association between obesity and survival in ovarian cancer (OC) patients, accounting for confounders as disease stage, histology, and comorbidities. METHODS: Retrospective matched case-control study of consecutive patients, with epithelial OC. Obese (body mass index [BMI] ≥ 35 kg m-2) patients were matched in a 1:4 ratio with patients having lower BMIs (BMI < 35 kg m-2) based on disease stage, cytoreduction state, tumor histology and ASA score. We compared the 3-year and total recurrence-free survival and overall survival through Kaplan-Meier survival curves and Cox proportional hazards. RESULTS: Overall, 153 consecutive patients were included, of whom 32 (20.9%) had a BMI ≥ 35. and 121 a BMI < 35. The median follow-up time was 39 months (interquartile range 18-67). Both study groups were similar in multiple prognostic factors, including American Society of Anesthesiologists physical status, completion of cytoreduction, histology and stage of disease (p = 0.981, p = 0.992, p = 0.740 and p = 0.984, respectively). Ninety-five (62.1%) patients underwent robotic surgery and conversion rate from robotic to laparotomy was similar in both groups 2 (6.3%) in obese group vs. 6 (5.0%) in lower BMI patients, p = 0.673. During the follow-up time, the rate of recurrence was similar in both groups; 21 (65.6%) in obese group vs. 68 (57.1%), p = 0.387 and the rate of death events was similar; 16 (50.0%) in obese group vs. 49 (40.5%), p = 0.333). The 3-year OS was higher in the obese group (log rank p = 0.042) but the 3-year RFS was similar in both groups (log rank p = 0.556). Median total OS was similar in both groups 62 months (95% confidence interval 25-98 months) in obese vs. 67 months (95% confidence interval 15-118) in the lower BMI group, log rank p = 0.822. Median RFS was similar in both groups; 61 months (95% confidence interval 47-74) in obese, vs. 54 (95% confidence interval 43-64), log rank p = 0.842. In Cox regression analysis for OS, including obesity, age, laparotomy and neoadjuvant treatment - only neoadjuvant treatment was independently associated with longer OS: odds ratio 1.82 (95% confidence interval 1.09-3.05) and longer RFS: odds ratio 2.16 (95% confidence interval 1.37-3.41). CONCLUSIONS: In the present study on consecutive cases of ovarian cancer, obesity did not seem to be associated with outcome, except for an apparent improved 3-year survival that faded away thereafter.

Incidence and risk factors of venous and arterial thromboembolic events among patients with ovarian cancer- data from a large Canadian database.

OBJECTIVES: To determine the incidence of thromboembolic events (TEEs) in ovarian cancer patients and to identify risk factors that are significantly associated with the development of venous thromboembolism (VTE), arterial thromboembolism (ATE), or overall TEEs in this population. METHODS: This is a retrospective cohort study of 4491 patients with epithelial ovarian cancer identified in the British Columbia cancer registry between 1996 and 2017. The presence of TEEs and risk factors were identified in administrative health records from fee-for-service provider visits and hospital data using ICD-9-CM and ICD-10-CM billing codes. Statistical analysis was performed using Chi-squared test and Fischer's exact test. RESULTS: Of 4491 patients with epithelial ovarian cancer included in this study, 1.74% experienced ATE and (9.44%) experienced VTE. There was a significant association found between the occurrence of TEEs and all-cause mortality. Sepsis was significantly associated with both venous and arterial thromboembolism. The top three risk factors for arterial thromboembolism included peripheral vascular disease (PVD), open wound, and aneurysm. CONCLUSIONS: Risk factors predictive of thrombosis in ovarian cancer patients are not consistent between ATE and VTE, thus thrombotic events should not be combined for analysis. Differential thrombosis risk assessment is needed to improve prevention strategies and guide thromboprophylaxis for these patients.

Visceral obesity determined by CT as a predictor of short-term postoperative complications in patients with ovarian cancer.

OBJECTIVE: To explore the association between visceral obesity and short-term postoperative complications in patients with advanced ovarian cancer undergoing cytoreductive surgery. METHODS: The medical records of patients with advanced epithelial ovarian cancer were reviewed. The visceral fat area, subcutaneous fat area and total fat area at the L3/4 level were measured on a preoperative single-slice CT scan. The receiver operating characteristic (ROC) curve was used to calculate the optimal cutoff value for the visceral fat area. The relationship between the visceral fat area and the characteristics of ovarian cancer patients were analyzed. Univariable and multivariable logistic regression analyses were performed to investigate relationship between perioperative characteristics and short-term complications. RESULTS: According to the ROC curve, the best cutoff value of the VFA was 93 cm2. Of the 130 patients, 53.8% (70/130) had visceral obesity. Patients with visceral obesity were older than those with nonvisceral obesity (58.4 years old vs. 52.1 years old, p < 0.001). The proportion of patients with hypertension was higher (35.7 vs. 13.3%, p = 0.003). The total fat area and subcutaneous fat area were larger in patients with visceral obesity (294.3 ± 75.5 vs. 176.2 ± 68.7, p < 0.001; 158.9 ± 54.7 vs. 121.7 ± 52.6, p < 0.001). Compared with patients in the nonvisceral obese group, patients in the visceral obese group were more likely to have postoperative fever (21/70 30.0% vs. 8/60 1.25%, p = 0.023), leading to a longer length of hospital stay (21 days vs. 17 days, p = 0.009). The time from surgery to adjuvant chemotherapy for patients with visceral obesity was shorter (24 days vs. 19 days, p = 0.037). Multivariate analysis showed that visceral obesity (OR = 6.451, p < 0.001) and operation time (OR = 1.006, p < 0.001) were independent predictors of postoperative complications. CONCLUSION: Visceral obesity is an important risk factor for short-term postoperative complications in patients with advanced ovarian cancer undergoing cytoreductive surgery.

What is the most pertinent definition of malnutrition in epithelial ovarian cancer to assess morbidity and mortality?

INTRODUCTION: Malnutrition is common in ovarian cancer and is a major cause of morbidity and mortality. We aimed to define the most pertinent way to assess malnutrition in patients with epithelial ovarian cancer (EOC) in order to study its impact on morbidity (intra and post-operative complications) and survival (OS, overall survival and RFS, recurrence-free survival). METHODS: We retrospectively included all patients with EOC from 2003 to 2020. Nutritional status was assessed using the weight loss at diagnosis (more or <5%), albuminemia, the Nutritional Risk Index (NRI), and the Malnutrition Universal Screening Tool (MUST). RESULTS: Six hundred and fifteen patients were included. Among them, 34% declared having lost >5% of their usual weight, 58% had an albuminemia <35 g/L, 86% presented an abnormal NRI and 29% an abnormal MUST score. After univariate analysis, weight loss>5% appeared to be significantly associated with RFS. An abnormal NRI or MUST score were significantly associated with a decrease in OS in univariate analysis. None of the markers of malnutrition studied were correlated with morbidity. CONCLUSION: We were not able to reach a consensus concerning the most accurate definition to define malnutrition and predict morbidity and mortality in EOC. However, this modifiable prognosis factor must be systematically assessed and managed accordingly.

High-fat diet and obesity are associated with differential angiogenic gene expression in epithelial ovarian cancer.

OBJECTIVE: We sought to evaluate the association between diet and angiogenic biomarkers in KpB mice, and the association between these markers, body mass index (BMI), and overall survival (OS) in high-grade serous cancers (HGSC). METHODS: Tumors previously obtained from KpB mice subjected to high-fat diets (HFD, n = 10) or low-fat diets (LFD, n = 10) were evaluated for angiogenesis based on CD-31 microvessel density (MVD). Data from prior microarray analysis (Agilent 244 K arrays) conducted in 10 mice were utilized to assess associations between diet and angiogenetic biomarkers. Agilent (mouse) and Affymetrix Human Genome U133a probes were linked to 162 angiogenic-related genes. The associations between biomarkers, BMI, and OS were evaluated in an HGSC internal database (IDB) (n = 40). Genes with unadjusted p < 0.05 were evaluated for association with OS in the TCGA-OV database (n = 339). RESULTS: There was no association between CD-31 and diet in mice (p = 0.66). Sixteen angiogenic-related genes passed the p < 0.05 threshold for association with HFD vs. LFD. Transforming growth factor-alpha (TGFA) demonstrated 72% higher expression in HFD vs. LFD mice (p = 0.04). Similar to the mouse study, in our HGSC IDB, higher TGFA expression correlated with higher BMI (p = 0.01) and shorter survival (p = 0.001). In the TCGA-OV dataset, BMI data was not available and there was no association between TGFA and OS (p = 0.48). CONCLUSIONS: HFD and obesity may promote tumor progression via differential modulation of TGFA. We were unable to confirm this finding in the TCGA dataset. Further evaluation of TGFA is needed to determine if this is a target unique to obesity-driven HGSC.

Incidence of and survival after surgical intervention for bowel obstruction in women with advanced ovarian cancer.

INTRODUCTION: Women with advanced ovarian cancer commonly present with peritoneal disease both at primary diagnosis and relapse, with risk of subsequent bowel obstruction. The aims of this study were to assess the cumulative incidence of and survival after intervention for bowel obstruction in women with advanced ovarian cancer, to identify factors predictive of survival and the extent to which the intended outcome of the intervention was achieved. MATERIAL AND METHODS: Women diagnosed with advanced ovarian cancer stages III and IV in 2009-2011 and 2014-2016 in the Stockholm-Gotland Region in Sweden were identified in the Swedish Quality Registry for Gynecologic Cancer. Through hospital records, types of intended and executed interventions for bowel obstruction were assessed, and as well as when in the course of oncologic treatment, the intervention was performed. Time from first intervention to death was analyzed with survival methodology and proportional hazard regression was used. RESULTS: Of 751 identified women, 108 had an intervention for bowel obstruction. Laparotomy was the most prevalent intervention and was used in 87% (94/108) of all women, with a success rate of 87% (82/94). An intervention for bowel obstruction was performed before or during first line treatment in 32% (35/108) with a cumulative incidence in the whole cohort of 14% (108/751, 95% confidence interval [CI] 11-16). Median survival after intervention for bowel obstruction was 4 months (95% CI 3-6). The hazard of death increased when the intervention was performed after completion of primary treatment (HR 4.46, 95% CI 1.61-12.29, P < 0.01), with a median survival of 3 months. In women subjected to radical surgery during primary treatment, the hazard of death after intervention for bowel obstruction decreased (hazard ratio [HR] 0.54, 95% CI 0.32-0.91, P = 0.02). CONCLUSIONS: Women with advanced ovarian cancer undergoing intervention for bowel obstruction have a dismal prognosis, regardless of which line of oncologic treatment the intervention was performed. In the majority of women an intervention for bowel obstruction was performed in a relapse situation with an even worse survival. Our findings emphasize the importance of a holistic approach in the decision-making before an intervention for bowel obstruction in women with advanced ovarian cancer.

Association of inflammation-related exposures and ovarian cancer survival in a multi-site cohort study of Black women.

BACKGROUND: An association was observed between an inflammation-related risk score (IRRS) and worse overall survival (OS) among a cohort of mostly White women with invasive epithelial ovarian cancer (EOC). Herein, we evaluated the association between the IRRS and OS among Black women with EOC, a population with higher frequencies of pro-inflammatory exposures and worse survival. METHODS: The analysis included 592 Black women diagnosed with EOC from the African American Cancer Epidemiology Study (AACES). Cox proportional hazards models were used to compute hazard ratios (HRs) and 95% confidence intervals (CIs) for the association of the IRRS and OS, adjusting for relevant covariates. Additional inflammation-related exposures, including the energy-adjusted Dietary Inflammatory Index (E-DIITM), were evaluated. RESULTS: A dose-response trend was observed showing higher IRRS was associated with worse OS (per quartile HR: 1.11, 95% CI: 1.01-1.22). Adding the E-DII to the model attenuated the association of IRRS with OS, and increasing E-DII, indicating a more pro-inflammatory diet, was associated with shorter OS (per quartile HR: 1.12, 95% CI: 1.02-1.24). Scoring high on both indices was associated with shorter OS (HR: 1.54, 95% CI: 1.16-2.06). CONCLUSION: Higher levels of inflammation-related exposures were associated with decreased EOC OS among Black women.

The Association of Kidney Function and Inflammatory Biomarkers with Epithelial Ovarian Cancer Risk.

BACKGROUND: One of the mechanisms of ovarian tumorigenesis is through inflammation. Kidney dysfunction is associated with increased inflammation; thus, we assessed its relationship with ovarian cancer risk. METHODS: In prospectively collected samples, we evaluated the association of kidney function markers and C-reactive protein (CRP) with ovarian cancer risk in the UK Biobank. We used multivariable-adjusted Cox proportional hazards models to evaluate quartiles of serum and urine markers with ovarian cancer risk overall and by histology. We assessed effect modification by CRP (≤3.0, >3.0 mg/L). RESULTS: Among 232,908 women (1,110 ovarian cancer cases diagnosed from 2006-2020), we observed no association between estimated glomerular filtration rate and ovarian cancer risk (Q4 vs. Q1: HR, 1.00; 95% confidence intervals, 0.83-1.22). Potassium was associated with endometrioid (Q4 vs. Q1: 0.33, 0.11-0.98) and clear cell (4.74, 1.39-16.16) tumors. Poor kidney function was associated with a nonsignificant increase in ovarian cancer risk among women with CRP>3.0 mg/L (e.g., uric acid Q4 vs. Q1; 1.23, 0.81-1.86), but not CRP≤3.0 mg/L (0.83, 0.66-1.05). Other associations did not vary across CRP categories. CONCLUSIONS: Kidney function was not clearly associated with ovarian cancer risk. Larger studies are needed to evaluate possible histology specific associations. Given the suggestive trend for increased ovarian cancer risk in women with poor kidney function and high CRP, future work is needed, particularly in populations with a high prevalence of inflammatory conditions. IMPACT: This study provided the first evaluation of markers of kidney function in relation to ovarian cancer risk.

Incidence of venous thromboembolism in patients with advanced stage ovarian cancer undergoing neoadjuvant chemotherapy: Is it time for thromboprophylaxis?

OBJECTIVES: Our objectives were to determine the incidence, timing, and risk factors for venous thromboembolisms (VTEs) in patients with advanced stage epithelial ovarian cancer (EOC) who received neoadjuvant chemotherapy (NACT). We explored the utilization of direct-acting oral anticoagulants (DOACs) for VTE treatment. METHODS: This retrospective cohort study included patients with advanced stage EOC receiving NACT followed by interval cytoreductive surgery (ICS) at a single institution. Risk factors were compared between patients with versus without VTE between EOC diagnosis and 180 days after ICS. Bleeding complications were compared between patient who received a DOAC versus non-DOAC. RESULTS: VTE cases occurred amongst 33 of the 154 (21.4%) patients with 4 (2.6%) concurrent with EOC diagnosis, 9 (5.8%) between EOC diagnosis and NACT start, 13 (8.4%) between NACT start and ICS, and 7 (4.5%) within 180 days after ICS. There were no statistically significant differences in risk factors assessed (age, body mass index, functional status, histology, Khorana score, and smoking history) between patients with versus without VTE. Eleven patients (33.3%) received a DOAC for VTE treatment. There were no significant differences in number of intraoperative blood transfusions (p = 0.38), blood loss (p = 0.95), or bleeding complications (p = 0.53) between patients treated with a DOAC versus a non-DOAC. CONCLUSION: There is a high incidence of VTE events (21.4%) in patients with advanced stage EOC undergoing NACT. Two-thirds of the VTEs may have been prevented with thromboprophylaxis as they occurred between EOC diagnosis and ICS. These data support consideration of thromboprophylaxis in all patients with advanced stage EOC undergoing NACT.

Multiple types of distress are prospectively associated with increased risk of ovarian cancer.

BACKGROUND: Few modifiable risk factors for epithelial ovarian cancer have been identified. We and other investigators have found that individual psychosocial factors related to distress are associated with higher risk of ovarian cancer. The present study examined whether co-occurring distress-related factors are associated with ovarian cancer risk. METHODS: Five distress-related factors were measured repeatedly over 21 years of follow-up: depression, anxiety, social isolation, widowhood, and, in a subset or women, posttraumatic stress disorder (PTSD). Cox proportional hazards models estimate relative risks (RR) and 95% confidence intervals (CI) of ovarian cancer for a time-updated count of distress-related factors, in age-adjusted models, then further adjusted for ovarian cancer risk factors and behavior-related health risk factors. RESULTS: Across 1,193,927 person-years of follow-up, 526 incident ovarian cancers occurred. Women with ≥3 versus no distress-related psychosocial factors demonstrated increased ovarian cancer risk (HRage-adjusted = 1.71; 95% CI = 1.16, 2.52). No significant difference in ovarian cancer risk was observed in women with one or two versus no distress-related psychosocial factors. In the subsample with PTSD assessed, ≥3 versus no distress-related psychosocial factors was associated with twofold greater ovarian cancer risk (HRage-adjusted = 2.08, 95% CI = 1.01, 4.29). Further analysis suggested that women at highest ovarian cancer risk had PTSD co-occurring with any other distress-related factor (HR = 2.19, 95% CI = 1.20, 4.01). Adjusting for cancer risk factors and health behaviors minimally impacted risk estimates. CONCLUSIONS: Presence of multiple indicators of distress was associated with risk of ovarian cancer. When including PTSD as an indicator of distress, the association was strengthened.

Stratifying the risk of ovarian cancer incidence by histologic subtypes in the Korean Epithelial Ovarian Cancer Study (Ko-EVE).

INTRODUCTION: This study aimed to verify the association between ovarian cancer (OC) and reproductive- and lifestyle-related risk factors stratified by the subtype of OC. METHODS: In this matched case-control study derived from the Korean epithelial ovarian cancer study (Ko-EVE), we calculated the risk of OC subtypes using odds ratios (ORs) and 95% confidence intervals (95% CIs) in a logistic regression model. RESULTS: As a result of matching, 531 cases and 2,124 controls were selected. Smoking had positive association with high-grade serous (HGS) OC (OR = 2.69, 95% CI = 1.15-6.30), whereas alcohol consumption had positive association with mucinous type (MUC) (OR = 3.63, 95% CI = 1.39-9.49). Obesity (≥30 kg/m2 ) was associated with clear cell type (CLC) (OR = 4.57, 95% CI = 1.06-19.77). Spontaneous abortion was negatively associated with CLC (OR = 0.34, 95% CI = 0.13-0.90), in contrast to HGS (OR = 1.43, 95% CI = 0.96-2.15). Tubal ligation, hysterectomy, and oophorectomy were associated with decreased risk of HGS (OR = 0.14, 95% CI = 0.05-0.39; OR = 0.23, 95% CI = 0.07-0.73; OR = 0.28, 95% CI = 0.08-0.97, respectively). Early menarche was strongly associated with increased risk of CLC, but not MUC (OR = 6.11, 95% CI = 1.53-24.42; OR = 3.23, 95% CI = 0.98-10.86). Further, childbirth (≥2 times) was negatively associated with endometrioid type OC and CLC (OR = 0.11, 95% CI = 0.04-0.35; OR = 0.12, 95% CI = 0.02-0.37, respectively). Oral contraceptives and hormone replacement therapy were negatively associated with OC (OR = 0.61, 95% CI = 0.40-0.93; OR = 0.51, 95% CI = 0.32-0.80, respectively), and similar negative associations were also observed in HGS (OR = 0.69; OR = 0.60, respectively). Associations between family history of breast cancer and OC, regular exercise (≥5/week), and artificial abortion and OC were similar across all subtypes (OR = 3.92; OR = 0.41; OR = 0.72, respectively). CONCLUSION: A heterogeneous association between some risk factors and the incidence of each subtype of epithelial OC was observed, suggesting that the carcinogenic mechanisms of each subtype may be partly different.

Assessing the need for venous thromboembolism prophylaxis at the time of neoadjuvant chemotherapy for ovarian cancer: A literature review.

OBJECTIVE: Gynecologic cancers, especially ovarian cancer, are associated with a high incidence of venous thromboembolism (VTE). Recent data have shown the risk of VTE development is not only limited to the postoperative period; there also appears to be an increased risk during neoadjuvant chemotherapy (NACT) administration, prompting the need for better risk stratification in this setting. We sought to assess the risk of VTE development in patients with ovarian cancer undergoing NACT. METHODS: We performed a PubMed literature review using the following medical terms: advanced ovarian cancer, advanced peritoneal cancer, advanced fallopian tube cancer, thrombosis, thromboembolic events, and neoadjuvant chemotherapy. Eligible studies included patients with advanced ovarian, fallopian tube, or peritoneal cancer who underwent NACT and had VTE. VTE was defined as either a deep venous thrombosis or a pulmonary embolism. RESULTS: Seven relevant studies were identified; all 7 were published between 2017 and 2021. Across these studies, we identified 1427 patients who underwent NACT and either had VTE at presentation or developed VTE during their treatment course. Of these patients, 1171 underwent NACT and were at risk for VTE development and were included in our pooled analysis. Of these patients, 144 (12.3%) developed VTE. CONCLUSIONS: VTE prophylaxis may be considered in patients with ovarian cancer undergoing NACT.

The application of risk models based on machine learning to predict endometriosis-associated ovarian cancer in patients with endometriosis.

INTRODUCTION: There is currently no satisfactory model for predicting malignant transformation of endometriosis. The aim of this study was to construct and evaluate a risk model incorporating noninvasive clinical parameters to predict endometriosis-associated ovarian cancer (EAOC) in patients with endometriosis. MATERIAL AND METHODS: We enrolled 6809 patients with endometriosis confirmed by pathology, and randomly allocated them to training (n = 4766) and testing cohorts (n = 2043). The proportion of patients with EAOC in each cohort was similar. We extracted a total of 94 demographic and clinicopathologic features from the medical records using natural language processing. We used a machine learning method - gradient-boosting decision tree - to construct a predictive model for EAOC and to evaluate the accuracy of the model. We also constructed a multivariate logistic regression model inclusive of the EAOC-associated risk factors using a back stepwise procedure. Then we compared the performance of the two risk-predicting models using DeLong's test. RESULTS: The occurrence of EAOC was 1.84% in this study. The logistic regression model comprised 10 selected features and demonstrated good discrimination in the testing cohort, with an area under the curve (AUC) of 0.891 (95% confidence interval [CI] 0.821-0.960), sensitivity of 88.9%, and specificity of 76.7%. The risk model based on machine learning had an AUC of 0.942 (95% CI 0.914-0.969), sensitivity of 86.8%, and specificity of 86.7%. The machine learning-based risk model performed better than the logistic regression model in DeLong's test (p = 0.036). Furthermore, in a prospective dataset, the machine learning-based risk model had an AUC of 0.8758, a sensitivity of 94.4%, and a specificity of 73.8%. CONCLUSIONS: The machine learning-based risk model was constructed to predict EAOC and had high sensitivity and specificity. This model could be of considerable use in helping reduce medical costs and designing follow-up schedules.

Racial differences in the association of body mass index and ovarian cancer risk in the OCWAA Consortium.

BACKGROUND: Obesity disproportionately affects African American (AA) women and has been shown to increase ovarian cancer risk, with some suggestions that the association may differ by race. METHODS: We evaluated body mass index (BMI) and invasive epithelial ovarian cancer (EOC) risk in a pooled study of case-control and nested case-control studies including AA and White women. We evaluated both young adult and recent BMI (within the last 5 years). Associations were estimated using multi-level and multinomial logistic regression models. RESULTS: The sample included 1078 AA cases, 2582 AA controls, 3240 White cases and 9851 White controls. We observed a higher risk for the non-high-grade serous (NHGS) histotypes for AA women with obesity (ORBMI 30+= 1.62, 95% CI: 1.16, 2.26) and White women with obesity (ORBMI 30+= 1.20, 95% CI: 1.02, 2.42) compared to non-obese. Obesity was associated with higher NHGS risk in White women who never used HT (ORBMI 30+= 1.40, 95% CI: 1.08, 1.82). Higher NHGS ovarian cancer risk was observed for AA women who ever used HT (ORBMI 30+= 2.66, 95% CI: 1.15, 6.13), while in White women, there was an inverse association between recent BMI and risk of EOC and HGS in ever-HT users (EOC ORBMI 30+= 0.81, 95% CI: 0.69, 0.95, HGS ORBMI 30+= 0.73, 95% CI: 0.61, 0.88). CONCLUSION: Obesity contributes to NHGS EOC risk in AA and White women, but risk across racial groups studied differs by HT use and histotype.

Splenectomy at the time of primary or interval cytoreductive surgery for epithelial ovarian carcinoma: A review of outcomes.

OBJECTIVES: We describe post-operative complications after cytoreductive surgery with and without splenectomy for Stage III or IV epithelial ovarian cancer, and identify areas for quality improvement in post-splenectomy care. METHODS: All patients with ovarian cancer cytoreductive surgery from 2008 to 2018 were identified using an institutional database Gynecologic Oncology Longitudinal Data Collection and Utilization Program (GOLD CUP). We compared patients who had and did not have splenectomy as part of cytoreductive surgery by demographics, comorbidities, stage, operative and post-operative data, readmission rates, progression free survival, overall survival and death from disease. Quality metrics reported include receipt of post-splenectomy education handouts and encapsulated-organism vaccines. Statistical analysis was completed in STATA SE 16.0. RESULTS: We identified 47 patients who underwent splenectomy and 454 who did not during primary or interval cytoreductive surgery. Final stage was IIIB in 1 (2.1%), IIIC in 26 (55.3%), IVA in 7 (14.9%), and IVB in 13 (27.7%) patients. Those with splenectomy had significantly higher stage. Surgery duration and hospital length of stay were longer and blood transfusion more common after splenectomy, but there were no differences in post-operative infection, readmission, or overall survival. Pancreatic leaks were seen in 4/47 (8.5%) patients. Post-splenectomy vaccinations were documented in 42/47 (89.4%) patients. Only 2/47 (4.3%) received post-splenectomy discharge instructions and 3/7 (42.9%) received aspirin for platelets 1 million or more. CONCLUSIONS: While splenectomy adds morbidity, it continues to offer benefit in those patients who can achieve optimal cytoreduction. Areas for quality improvement in post-splenectomy care include receipt of vaccinations, patient discharge information, and timely pancreatic fistula management.

Moderate to severe distress in half of ovarian cancer patients undergoing treatment highlights a need for more proactive symptom and psychosocial management.

OBJECTIVE: Distress screening and management is a recommended component of oncology care. Our objective was to evaluate distress rate, sources, and compliance with psychosocial follow-up among ovarian cancer patients receiving chemotherapy. METHODS: We reviewed patient distress surveys completed by ovarian cancer patients receiving chemotherapy from 10/2017-6/2019. Lay or nurse navigators conducted screening with the NCCN Distress Thermometer from 0 (none) to 10 (highest distress). A distress score ≥ 4 (moderate/severe) was considered a positive screen. A recommendation for psychosocial follow-up was automatically generated in the treatment care plan based upon a yes response to any depression-related concern, independent of distress score. Documentation of referral to a mental health professional or social worker for counseling was considered compliant with psychosocial follow-up. We performed descriptive statistics and bivariate analyses. RESULTS: 97/211 (46%) ovarian cancer patients screened positive for distress. Average score was 6.1 for those who screened positive and 3.3 for the entire cohort (range 0-10). Unmarried status (p < 0.01) was associated with positive screen, whereas non-white race (p = 0.26) and recurrent disease (p = 0.21) were not. Median age was older for patients with a positive distress screen (p < 0.01). Among screened patients, the most frequent sources of distress were: cognitive/physical (87%), psychosocial (62%), practical (84%), and family concerns (40%). Of 50 patients recommended to have psychosocial referral, 4 (8%) patients had documented psychiatric follow-up and 19 (38%) patients had documented psychosocial counseling by a social worker. CONCLUSIONS: Nearly half of ovarian cancer patients screened positive for moderate/severe distress. Cancer/treatment-related cognitive/physical symptoms were the most frequent sources. Improved methods of symptom monitoring and management during treatment and resources to address psychosocial concerns are needed to improve distress management of ovarian cancer patients.

Pre-diagnosis and post-diagnosis dietary patterns and survival in women with ovarian cancer.

BACKGROUND: Evidence is limited on inflammation-related dietary patterns and mortality in ovarian cancer survivors. METHODS: We examined the associations between pre- and post-diagnosis dietary patterns, including change in diet from before to after diagnosis, and mortality among 1003 ovarian cancer survivors in two prospective cohort studies. Dietary pattern scores for empirical dietary inflammatory pattern (EDIP) and Alternative Healthy Eating Index (AHEI) were calculated based on food frequency questionnaires. We used Cox proportional hazard models to calculate hazard ratios (HRs) and 95% confidence intervals (CIs) for ovarian cancer-specific and all-cause mortality. RESULTS: Pre-diagnosis EDIP score and AHEI were not associated with mortality. Among non-high grade serous cases, a higher post-diagnosis EDIP score was associated with increased risk of all-cause mortality (HR5th vs 1st quintile = 1.95, 95% CI = 1.04-3.67, p-trend = 0.06). Compared to survivors consuming a low EDIP score diet before and after diagnosis, high post-diagnosis EDIP was associated with increased risk of ovarian cancer specific mortality (pre-to-post diagnosis low/high, HR = 1.38, 95% CI = 0.99-1.92; high/high HR = 1.58, 95% CI = 1.09-2.30) and all-cause mortality (low/high HR = 1.44, 95% CI = 1.06-1.95; high/high HR = 1.55, 95% CI = 1.10-2.19). CONCLUSION: Consuming a more inflammatory dietary pattern post-diagnosis was associated with increased mortality in ovarian cancer survivors, suggesting limiting the inflammatory potential of diet post-diagnosis could lead to enhanced survivorship.

Development and Validation of a Nomogram to Predict the Probability of Venous Thromboembolism in Patients with Epithelial Ovarian Cancer.

OBJECTIVE: To identify predictive factors and develop a nomogram to predict the probability of venous thromboembolism for epithelial ovarian cancer patients. Methods: Our study cohort was composed of 208 EOC patients who had received initial treatment in Sun Yat-sen Memorial Hospital from January 2016 to March 2020. Clinicopathological variables predictive of VTE were identified using univariate logistic analysis. A multivariate logistic regression model was used to select the predictive factors used for nomogram. The accuracy of nomogram was evaluated by the Concordance index (C-index), the area under the receiver-operator characteristic (ROC) curve, area under concentration-time curve (AUC) and the calibration curve. Results: Advancing age (hazard ratio [HR], 1.042; 95% confidence interval [CI], 1.000-1.085; P = .048), higher D-dimer level (HR, 1.144; 95%CI, 1.020-1.283; P = .022), lower PR immunohistochemical positive rate (HR, 0.186; 95%CI, 0.034-1.065; P = .059) and higher Ki67 immunohistochemical positive rate (HR, 4.502; 95%CI, 1.637-12.380; P = .004) were found to be independent risk factors for VTE, and were used to construct the nomogram. The C-index for VTE prediction of the nomogram was 0.75. Conclusions: We constructed and validated a nomogram able to quantify the risk of VTE for EOC patients, which can be applied in recognizing EOC patients with high risk of VTE.