The impact of care by gynecologic oncologists on the survival of primary epithelial ovarian cancer: a population-based study

Abstract

Objectives: Timely treatment of primary epithelial ovarian cancer (EOC) by gynecologic oncologists (GOs) with a combination of surgery and/or platinum/taxane-based chemotherapy at expert centers has been advocated. Nonetheless, some patients have their oncologic surgery by non-GOs or are not assessed by GOs prior to starting their initial cancer treatment. The aim of the study was to evaluate the impact of care by gynecologic oncologists on the survival of primary non-mucinous EOC in Ontario. Methods: In this population-based retrospective cohort study, patients diagnosed with non-mucinous EOC between 2007-2018 were identified using province-wide administrative databases. Assessment by GOs was defined as any consultation 6 months prior to diagnosis date and up to first treatment (surgery or chemotherapy). Type of surgeon performing surgeries related to ovarian cancer were identified (general gynecologists, general surgeons, or GOs). Multivariate cox proportional hazards regression models were used to evaluate the impact of GO consultations prior to initiating treatment or having surgery done by a GO on the survival of ovarian cancer. Age, socioeconomic status (SES), year of diagnosis, comorbidity score and days from diagnosis to surgery were covariates in the models. FIGO stage was available in 70% of our cohort. Sensitivity analysis was performed adjusting for stage and the other covariates. Results: A total of 10,086 EOC patients were included between 2007-2018 with a median age of 63 years (range 53-72). Primary treatment included surgery alone in 26% of patients (2,593/10,086), while 63% (6,397/10,086) had neoadjuvant chemotherapy and/or adjuvant chemotherapy and 11% (1,096/10,086) had chemotherapy alone. A total of 8,990 ovarian cancer patients underwent surgery as part of their primary treatment. During the study period, there was an increase in GO consultations (79% in 2007 to 87% in 2018-19) and surgeries performed by GOs (69% in 2007 to 88% in 2018-19). On multivariate analysis, there was an increase in all-cause mortality for patients not assessed by GOs before first treatment (Hazard ratio (HR): 2.18; 95% CI 1.98-2.41). There was also an increase in all-cause mortality if ovarian cancer surgery was performed by non-GOs (HR 2.37; 95% CI 2.11- 2.67) after adjusting for age, SES, year of diagnosis, comorbidity score and days from diagnosis to surgery. In a sensitivity analysis where FIGO stage was added to the models, there remained a significant increase in all-cause mortality for patients not assessed by GOs before first treatment (HR 1.44; 95% CI 1.26-1.66) and an increase in all-cause mortality if ovarian cancer surgery was performed by non-GOs (HR 2.23; 95% CI 1.88-2.65). Conclusions: Assessment by GO before starting initial treatment (chemotherapy or surgery) was associated with improved survival in women with non-mucinous EOC. Type of surgeons performing primary ovarian cancer surgery had a significant impact on survival.