190 Identifying subgroups that benefit from ultra-radical upfront surgery in advanced epithelial ovarian cancer: auxiliary data from the Stockholm ovarian cancer project

Introduction/Background We previously reported that a shift to upfront ultra-radical surgery in a public health care setting increased complete macroscopic resection (CMR) rates without improving survival in women with advanced epithelial ovarian cancer (EOC). Still, CMR conferred the longest survival both before and after this shift. This paradox is likely explained by subgroups of women who may or may not benefit from increased surgical proficiency. Our objective was to identify these subgroups. Methodology This cohort study included all women diagnosed with EOC between 2009–2011 and 2014–2016, in the Stockholm/Gotland region, Sweden. The latter cohort was exposed to the upfront ultra-radical surgical treatment algorithm. The association between surgical extent defined by a surgical complexity score (SCS) and overall survival (OS) was analyzed with Cox regression models. Results The final analysis included 272 women with CMR and 150 women receiving chemotherapy only. Exposed women with stage III had an improved OS after 5-years, Hazard Ratio (HR) 0.37 (95% CI, 0.16–0.85); p=0.02, in contrast to stage IV, HR 1.23 (95% CI, 0.72–2.11); p=0.44. Exposed women where CMR was achieved with a low/medium SCS had an improved OS, adjusted HR (aHR) 0.56 (95% CI, 0.33–0.96); p=0.03 contrary to when a high SCS was required, aHR 0.84 (95% CI, 0.40–1.75); p=0.63. There was no absolute difference in 5-year OS in women with stage IV where a high SCS was needed to achieve CMR vs. the alternative (chemotherapy only), 5.7% (95% CI, -4.3 to 15.7). Conclusion With longer follow-up, an upfront ultra-radical treatment algorithm improves OS in women with stage III EOC and in whom CMR was achieved with a low/medium SCS. The effect of cytoreductive surgery is limited when a high SCS is required, particularly in stage IV. Clinical trials are warranted to investigate the effect of surgical resection altogether in this subgroup. Disclosures Nina Groes Kofoed, Oscar Peñaloza, Matteo Bottai and Sahar Salehi declare no conflict of interest, Henrik Falconer declares the potential conflict of interest; 'receipt of honoraria or consultation fees' from Surgical Science.