A multivariate analysis of the prognostic impact of tumor burden, surgical timing and complexity after complete cytoreduction for advanced ovarian cancer

Supragastric lesser sac: an insidious site for surgical exploration during the debulking surgery in advanced ovarian cancer

OBJECTIVE

Metastases in the supragastric lesser sac (SGLS) are not only occult but are also barriers to complete resection of ovarian cancer. We describe a cohort of patients with SGLS disease undergoing debulking surgery.

METHODS

We identified all patients who underwent evaluation and eventual resection of SGLS disease as part of cytoreductive surgery for stage IIIC-IVB high-grade epithelial ovarian cancer at our institution from January 2018 to August 2022.

RESULTS

Thirty-three of 286 patients (11.5%) underwent resection of SGLS disease. Metastases in the SGLS were identified by preoperative imaging in 4 of 33 patients (12.1%). The median peritoneal cancer index score was 22 (range, 9-33). Through surgical exploration, metastases were frequently seen in the right diaphragm (100%), hepatorenal recess (97%), lesser omentum (81.8%), left diaphragm (78.8%), supracolic omentum (75.8%), anterior transverse mesocolon (72.7%), splenic hilum (63.6%), ligamentum teres hepatis (60.6%), and gallbladder fossa (51.5%). The lesser omentum was normal in 6 of 33 (18.2%) patients, despite metastases within the SGLS. A total of 54.5% of patients underwent complex surgery (surgical complexity scores; median, 8; range, 3-14). Complete resections were achieved in 19 (57.6%) patients. No complications were related to the resection of SGLS disease. The median length of progression-free survival was 24.8 months (95% confidence interval=16.6-32.9).

CONCLUSION

Metastases to the SGLS are not uncommon in advanced ovarian cancer, particularly those with widely disseminated disease. Disease in this recess is rarely identified by preoperative imaging and deserves systematic surgical exploration to attain complete cytoreduction.

Open abdominal vacuum pack technique for the management of severe abdominal complications after cytoreductive surgery in ovarian cancer

INTRODUCTION

The aim of this study was to evaluate the indications and management of grade III-IV postoperative complications in patients requiring vacuum-assisted open abdomen after debulking surgery for ovarian carcinomatosis.

METHODS

Retrospective study of prospectively collected data from patients who underwent a cytoreductive surgery by laparotomy for an epithelial ovarian cancer that required postoperative management of an open abdomen. An abdominal vacuum-assisted wound closure (VAWC) was applied in cases of abdominal compartmental syndrome (ACS) or intra-abdominal hypertension, to prevent ACS. The fascia was closed with a suture or a biologic mesh. The primary aim was to achieve primary fascial closure. Secondary outcomes considered included complications of cytoreductive surgery (CRS) and open abdominal wounds (hernia, fistula).

RESULTS

Two percent of patients who underwent CRS required VAWC during the study's patient inclusion period. VAWC indications included: (i) seven cases of gastro-intestinal perforation, (ii) three necrotic enterocolitis, (iii) two intestinal ischemia, (iv) three anastomotic leakages and (v) four intra-abdominal hemorrhages. VAWC was used to treat indications (i) to (iv) (which represented 73.7% of cases), to prevent compartmental syndrome. Primary fascia closure was achieved in 100% of cases, in four cases (21.0%) a biologic mesh was used. Median hospital stay was 65 days (range: 18-153). Four patients died during hospitalization, three of these within 30 days of VAWC completion.

CONCLUSION

VAWC for managing open abdominal wounds is a reliable technique to treat surgical post-CRS complications in advanced ovarian cancer and reduces the early post-operative mortality in cases presenting with severe complications.

Response to treatment and prognostic significance of supradiaphragmatic disease in patients with high-grade serous ovarian cancer

OBJECTIVES

This study was designed to investigate the response to chemotherapy of supradiaphragmatic disease diagnosed by preoperative imaging. As secondary objectives, oncologic outcomes of patients affected by supradiaphragmatic disease and their pattern of recurrence were also evaluated.

METHODS

Data of consecutive patients with newly diagnosed FIGO stage IV (for supradiaphragmatic disease) epithelial ovarian cancer undergoing either primary debulking surgery or neoadjuvant chemotherapy plus interval debulking surgery between 2004 and 2021, were retrospectively collected. All patients were preoperatively evaluated by chest/abdominal CT scan or ¹⁸F-FDG PET/CT preoperatively and at follow-up to evaluate response to chemotherapy. At follow-up visits, site of recurrence diagnosed by imaging techniques was systematically recorded as it occurred. Progression-free and overall survival were measured by using Kaplan-Meier and Cox models.

RESULTS

A total of 130 patients was included in this study with a median (range) follow-up of 32.9 (12.8-176.7) months. Complete or partial response was achieved in most of the patients after 3 cycles (77.7%) and 6 cycles (85.4%) of chemotherapy. At follow-up, recurrence occurred in 96 (73.8%) patients and the main site of recurrence was abdomen only in 64 (66.7%) patients. At multivariate analysis, residual disease after surgery was the only variable influencing survival outcomes.

CONCLUSIONS

Supradiaphragmatic disease respond to chemotherapy in most patients affected by advanced EOC and recurrence mainly occurs in the abdomen. Results from this study confirms that abdominal optimal cytoreduction is the main surgical goal in the treatment of women affected by FIGO stage IV EOC.

Deep learning magnetic resonance imaging predicts platinum sensitivity in patients with epithelial ovarian cancer

Objective

The aim of the study is to develop and validate a deep learning model to predict the platinum sensitivity of patients with epithelial ovarian cancer (EOC) based on contrast-enhanced magnetic resonance imaging (MRI).

Methods

In this retrospective study, 93 patients with EOC who received platinum-based chemotherapy (≥4 cycles) and debulking surgery at the Sun Yat-sen Memorial Hospital from January 2011 to January 2020 were enrolled and randomly assigned to the training and validation cohorts (2:1). Two different models were built based on either the primary tumor or whole volume of the abdomen as the volume of interest (VOI) within the same cohorts, and then a pre-trained convolutional neural network Med3D (Resnet 10 version) was transferred to automatically extract 1,024 features from two MRI sequences (CE-T1WI and T2WI) of each patient to predict platinum sensitivity. The performance of the two models was compared.

Results

A total of 93 women (mean age, 50.5 years ± 10.5 [standard deviation]) were evaluated (62 in the training cohort and 31 in the validation cohort). The AUCs of the whole abdomen model were 0.97 and 0.98 for the training and validation cohorts, respectively, which was better than the primary tumor model (AUCs of 0.88 and 0.81 in the training and validation cohorts, respectively). In k-fold cross-validation and stratified analysis, the whole abdomen model maintained a stable performance, and the decision function value generated by the model was a prognostic indicator that successfully discriminates high- and low-risk recurrence patients.

Conclusion

The non-manually segmented whole-abdomen deep learning model based on MRI exhibited satisfactory predictive performance for platinum sensitivity and may assist gynecologists in making optimal treatment decisions.

Nomogram to predict postoperative complications after cytoreductive surgery for advanced epithelial ovarian cancer: A multicenter retrospective cohort study

Objective

To establish nomograms to predict the risk of postoperative complications following cytoreductive surgery in patients with advanced epithelial ovarian cancer (AEOC).

Methods

A multicenter retrospective cohort study that included patients with FIGO stage IIIC-IV epithelial ovarian cancer who underwent cytoreductive surgery was designed. By using univariate and multivariate analyses, patient preoperative characteristics were used to predict the risk of postoperative complications. Multivariate modeling was used to develop Nomograms.

Results

Overall, 585 AEOC patients were included for analysis (training cohort = 426, extrapolation cohort = 159). According to the findings, the training cohort observed an incidence of postoperative overall and severe complications of 28.87% and 6.10%, respectively. Modified frailty index (mFI) (OR 1.96 and 2.18), FIGO stage (OR 2.31 and 3.22), and Surgical Complexity Score (SCS) (OR 1.16 and 1.23) were the clinical factors that were most substantially associated to the incidence of overall and severe complications, respectively. The resulting nomograms demonstrated great internal discrimination, good consistency, and stable calibration, with C-index of 0.74 and 0.78 for overall and severe complications prediction, respectively. A satisfactory external discrimination was also indicated by the extrapolation cohort, with the C-index for predicting overall and severe complications being 0.92 and 0.91, respectively.

Conclusions

The risk of considerable postoperative morbidity exists after cytoreductive surgery for AEOC. These two nomograms with good discrimination and calibration might be useful to guide clinical decision-making and help doctors assess the probability of postoperative complications for AEOC patients.

Who really benefits from intraperitoneal chemotherapy for advanced ovarian cancer? A treatment-free survival analysis of the AICE trial

OBJECTIVE

To investigate whether peritoneal disease extent can predict the survival benefit of intraperitoneal/intravenous (IP/IV) chemotherapy in ovarian cancer.

DESIGN

A treatment-free survival (TFS) analysis.

SETTING

Five-centre trial.

POPULATION

An extended follow-up of the Additional Intraperitoneal Cisplatin and Etoposide in ovarian cancer (AICE) trial (NCT01669226), with data cut-off on 27 August 2020. Patients were categorised into subgroups with high tumour burden (HTB) and low tumour burden (LTB).

METHODS

Overall survival (OS) was divided into time on protocol treatment exposure (T), time free of subsequent treatment or death (TFS) and time after the first subsequent therapy (REL). TFS analyses and quality-adjusted OS were calculated by multiplying the mean time in each health state by its assigned utility: quality-adjusted OS = u_t  × T + TFS + u_rel  × REL.

MAIN OUTCOME MEASURES

The area under each Kaplan-Meier curve was estimated using the 96-month restricted mean time, with threshold utility analyses used to illustrate quality-adjusted OS comparisons.

RESULTS

In the HTB subgroup, the restricted mean TFS was 33.9 months and 18.7 months in the IP/IV and IV groups, respectively (p = 0.005), with a significant quality-adjusted OS gain (13.2-16.0 months). In the LTB subgroup, IP/IV therapy yielded no survival benefit in either TFS (p = 0.268) or quality-adjusted OS (range: 1.4-6.3 months).

CONCLUSIONS

Both TFS and quality-adjusted OS was longer across all utility weight values with IP/IV than with standard IV therapy in the HTB subgroup, whereas patients in the LTB subgroup did not benefit from the therapy. The tumour burden of ovarian cancer should be assessed before deciding on IP/IV versus IV treatment.

Prognostic influence of an early time to chemotherapy following primary cytoreductive surgery for advanced epithelial ovarian cancer

Objective

The current investigation analyzes the prognostic role of the time to chemotherapy (TTC) interval following primary cytoreductive surgery for patients with advanced epithelial ovarian cancer.

Methods

Characteristics and outcome data for 509 consecutive patients with stage IIIB–IVB ovarian, fallopian tube, and peritoneal cancer who had primary cytoreductive surgery between January 2000 and December 2019 are utilized. A univariate Cox regression determined the association of categorical variables with progression-free survival (PFS) and overall survival (OS). Significant variables (p≤0.05) on univariate analysis were applied to Cox proportional hazard regression.

Results

The median TTC was 19 days and overall follow-up was 62.2 months. The PFS and OS were 25.5 months and 78.4 months for the study cohort plus 28.4 months and OS 84.5 months for patients rendered grossly disease-free. An early TTC (7–14 vs. 15–21 vs. 22–28 vs. >28 days) was associated with an improved PFS (41.7 vs. 30.6 vs. 18.9 vs. 17.9 months; p<0.001) and OS (132.7 vs. 104.6 vs. 56.5 vs. 48.0 months; p<0.001). The performance status, histology, disease distribution, dimension of residual disease, and categorical plus continuous TTC were predictors of PFS and OS. The use of maintenance therapy was also a predictor of PFS, and the route of chemotherapy administration was a predictor of OS.

Conclusions

For advanced epithelial ovarian cancer, a TTC of less than 21-days was observed to independently improve the PFS and OS. A 7–14 days TTC trended towards a further extension of the OS.

Chemotherapy was administered as early as possible after extensive primary cytoreductive surgery for patients with stage IIIB-IV epithelial ovarian cancer. Early time to chemotherapy improved survival. Complete recovery from surgery before chemotherapy is not essential.

Feature Selection is Critical for 2-Year Prognosis in Advanced Stage High Grade Serous Ovarian Cancer by Using Machine Learning

Introduction

Accurate prediction of patient prognosis can be especially useful for the selection of best treatment protocols. Machine Learning can serve this purpose by making predictions based upon generalizable clinical patterns embedded within learning datasets. We designed a study to support the feature selection for the 2-year prognostic period and compared the performance of several Machine Learning prediction algorithms for accurate 2-year prognosis estimation in advanced-stage high grade serous ovarian cancer (HGSOC) patients.

Methods

The prognosis estimation was formulated as a binary classification problem. Dataset was split into training and test cohorts with repeated random sampling until there was no significant difference (p = 0.20) between the two cohorts. A ten-fold cross-validation was applied. Various state-of-the-art supervised classifiers were used. For feature selection, in addition to the exhaustive search for the best combination of features, we used the-chi square test of independence and the MRMR method.

Results

Two hundred nine patients were identified. The model's mean prediction accuracy reached 73%. We demonstrated that Support-Vector-Machine and Ensemble Subspace Discriminant algorithms outperformed Logistic Regression in accuracy indices. The probability of achieving a cancer-free state was maximised with a combination of primary cytoreduction, good performance status and maximal surgical effort (AUC 0.63). Standard chemotherapy, performance status, tumour load and residual disease were consistently predictive of the mid-term overall survival (AUC 0.63–0.66). The model recall and precision were greater than 80%.

Conclusion

Machine Learning appears to be promising for accurate prognosis estimation. Appropriate feature selection is required when building an HGSOC model for 2-year prognosis prediction. We provide evidence as to what combination of prognosticators leads to the largest impact on the HGSOC 2-year prognosis.

Effect of tumor burden and radical surgery on survival difference between upfront, early interval or delayed cytoreductive surgery in ovarian cancer

OBJECTIVE

We sought to evaluate the impact on survival of tumor burden and surgical complexity in relation to the number of cycles of neoadjuvant chemotherapy (NACT) in patients with advanced ovarian cancer (OC) with minimal (CC-1) or no residual disease (CC-0).

METHODS

This retrospective study included patients with International Federation of Gynaecology and Obstetrics IIIC-IV stage OC who underwent debulking surgery at 4 high-volume institutions between January 2008 and December 2015. We assessed the overall survival (OS) of primary debulking surgery (PDS group), early interval debulking surgery after 3-4 cycles of NACT (early IDS group) and delayed debulking surgery after 6 cycles (DDS group) with CC-0 or CC-1 according to peritoneal cancer index (PCI) and Aletti score.

RESULTS

Five hundred forty-nine women were included: 175 (31.9%) had PDS, 224 (40.8%) early IDS and 150 (27.3%) DDS. Regardless of Aletti score, median OS after PDS was significantly higher than after early IDS or DDS, but the survival difference was higher in women with an Aletti score <8. Among patients with PCI ≤10, median OS after PDS was significantly higher than after early IDS or DDS. In women with PCI >10, there were no differences between PDS and early IDS, but DDS was associated with decreased OS.

CONCLUSION

The benefit of complete PDS compared with NACT was maximal in patients with a low complexity score. In patients with low tumor burden, there was a survival benefit of PDS over early IDS or DDS. In women with high tumor load, DDS impaired the oncological outcome.

Timing of surgery in patients with partial response or stable disease after neoadjuvant chemotherapy for advanced ovarian cancer

OBJECTIVE

The ideal number of neoadjuvant chemotherapy (NACT) cycles prior to interval tumor-reductive surgery (iTRS) for advanced ovarian cancer is poorly defined. We sought to assess survival stratified by number of NACT cycles and residual disease following iTRS in patients with advanced ovarian cancer with partial response (PR) or stable disease (SD) following 3-4 cycles of NACT.

METHODS

We retrospectively identified patients with advanced high-grade ovarian cancer (diagnosed 2/1/2013 to 2/1/2018) who received at least 3 cycles of NACT and iTRS and had a PR or SD. The population was divided into four groups based on the number of NACT cycles prior to iTRS and residual disease status after (CGR [complete gross residual] or incomplete resection [any amount of residual disease]): 1) 3-4 NACT cycles/CGR, 2) 3-4 NACT cycles/incomplete resection, 3) > 4 cycles/CGR, and 4) >4 cycles/incomplete resection. Overall survival (OS) and progression-free survival (PFS) were estimated using a Kaplan-Meier product-limit estimator and modeled using univariable and multivariable Cox proportional hazards analysis.

RESULTS

The cohort consisted of 265 patients with advanced high-grade ovarian cancer with a median age at diagnosis of 65 years. Most were White (87%), had serous histology (89%), and stage IV disease (57%), with an overall CGR rate of 81%. In a multivariable analysis receipt of >4 NACT cycles was not associated with worse PFS or OS (adjusted hazard ratio [aHR] 1.02, 95% CI 0.74-1.42; aHR 1.12, 95% CI, 0.73-1.72 respectively) than was receipt of 3-4 cycles. Any amount of residual disease was associated with worse PFS and OS regardless of the number of NACT cycles (aHR 1.56, 95% CI 1.09-2.22; aHR 2.38, 95% CI 1.52-3.72 respectively).

CONCLUSIONS

Residual disease was associated with worse survival outcomes regardless of the number of NACT cycles in patients with PR or SD after NACT for advanced high-grade ovarian cancer. These data suggest that the ability to achieve CGR should take precedence in decision-making regarding the timing of surgery.

Adherence to French and ESGO Quality Indicators in Ovarian Cancer Surgery: An Ad-Hoc Analysis from the Prospective Multicentric CURSOC Study

BACKGROUND

Quality Indicators for ovarian cancer (OC) have been developed by the European Society of Gynaecological Oncology (ESGO) and by the French National Cancer Institute (Institut National du Cancer, INCa). The aim of the study was to characterize OC care distribution in France by case-volume and to prospectively evaluate the adherence of high-volume institutions to INCa/ESGO quality indicators.

METHODS

The cost-utility of radical surgery in ovarian cancer (CURSOC) trial is a prospective, multicenter, comparative and non-randomized study that includes patients with stage IIIC-IV epithelial OC treated in nine French health care tertiary institutions. Adherence to institutional quality indicators were anonymously assessed by an independent committee. OC care distribution in France were provided by the nationwide database of hospital procedures.

RESULTS

More than half of patients are treated in low-volume institutions. Among the nine high-volume centers participating in the study, four (44.4%) met all institutional INCa/ESGO quality indicators. The other five (55.6%) did not fulfil one of the quality indicator criteria.

CONCLUSIONS

Access to high-volume OC providers in France is restricted to a minority of patients, and yet half of the referral institutions included in this study failed to meet all recommended institutional quality indicators. It is mandatory that national authorities work both to improve OC centralization and to incorporate quality assurance programs into certified centers.