Concordance of a laparoscopic scoring algorithm with primary surgery findings in advanced stage ovarian cancer

To determine the concordance between the laparoscopic scoring assessment and extent of disease identified at primary tumor reductive surgery (TRS) in patients with advanced ovarian cancer.

From April 2013 to June 2017, we prospectively triaged patients with stage IIA to IVB ovarian cancer to laparoscopic scoring assessment. A validated predictive index value (PIV) score (range: 0-14) was assigned. Patients with PIV scores <8 were offered primary surgery and those with score ≥8 received NACT. Patients who underwent primary TRS had a second PIV score based on laparotomy findings. Concordance percentages were calculated between the two scores. Positive predictive value (PPV) was calculated to reflect the performance of the laparoscopic PIV score to predict R0 (complete gross resection) at TRS.

226 patients underwent laparoscopic scoring assessment, of which 139 (61.5%) had a PIV score <8 and 81 (35.8%) had a PIV score ≥8. 6 patients (2.7%) were unscoreable. There was 96% overall concordance between PIV scores at laparoscopy and primary TRS. Concordance scores by location were: bowel infiltration 74.7%, mesenteric disease 84.6%, liver surface involvement 86.5%, omental disease 89.7%, diaphragm disease 92.9%, stomach infiltration 94.7%, peritoneal carcinomatosis 94.8%. A laparoscopic PIV score of <8 had a PPV of 85.4% at predicting R0 at primary TRS.

Laparoscopic assessment of tumor burden is a feasible tool to predict R0 cytoreduction in patients with advanced ovarian cancer. Concordance between PIV scores at laparoscopy and primary TRS varied by anatomic location, with the lowest concordance seen in predicting bowel infiltration.