Developing a platform to investigate the heterogeneity of outcomes for patients with ovarian cancer.

Background: The geographical variation in treatment patterns for patients with ovarian cancer is profound, long-standing and worrying. Although these variations were highlighted in a recent UK registry audit, granular data to provide explanations for these variations have been lacking.

Methods: A consortium of six UK centres was generated to curate and submit data for all patients treated at their centre for a 2-year period. Descriptive statistics were combined with Cox regression and Kaplan-Meier analysis to confirm the findings from the national registry audit and identify possible drivers of the heterogeneity previously described.

Results: Records for 1117 patients treated in six centres in 2018 and 2019 were collated. Although there were differences in the clinical characteristics of patients between centres, these were not enough to account for the significant variation in survival outcomes between centres (p<0.001). Treatment rates varied between centres with between 30% and 76% of patients receiving combination therapy but in Cox models 'treatment centre' remained a predictor of 1 year survival independent of patient, tumour factors and treatment choice.

Conclusion: Variations in outcome seen between UK centres are not related solely to casemix but rather to the approach and ethos of each centre towards advanced ovarian cancer treatment options. Although important, differences in treatment patterns do not completely explain the variations seen and further work is required to understand the drivers of difference seen.