Validation of the KELIM score as a predictor of response to neoadjuvant treatment in patients with advanced high grade serous ovarian cancer

Abstract

ObjectiveThe objective of this study is to externally validate the KELIM (rate of elimination of CA-125 elimation) score in patients with high grade serous ovarian cancer(HGSC)undergoing NACT and determine its relation to outcome of cytoreduction, platinum sensitivity, progression free(PFS) and overall survival(OS). MethodsThis is a retrospective cohort study of patients with Stage III-IV HGSC diagnosed between January 1, 2010 and December 31, 2019 and treated with NACT. KELIM score was calculated using at least 3 CA-125 values within the first 100 days of chemotherapy. Demographic parameters were collected and Kaplan Meier survival analyses were performed for PFS and OS. This study was approved by local ethics board. Results217 patients met inclusion criteria. Median follow-up was 28.93 months(range 2.86–135.06). There was no significant difference in stage, functional status, cytoreductive outcome or BRCA status(germline or somatic) between patients with a KELIM ≥ 1 and <1. Patients with a KELIM<1 had a lower median PFS (13.58 vs 19.69, p < 0.001), median platinum free interval(PFI) (7.66 vs 13.64, p < 0.001) and 5-year OS (57% vs 72%, p = 0.0140) compared to patients with KELIM≥1 . After adjusting for stage, treatment delays, bevacizumab or poly adenosine diphosphate-ribose polymerase(parp)-inhibitor use, and BRCA status, patients with KELIM<1 had a high risk of disease progression(HR = 1.57 (95% CI 1.08–2.28) and death(HR = 1.99 (95% CI 1.01–3.95) compared to KELIM≥1. BRCA status was independently associated to an increase on KELIM score (OR = 1.917, 95% CI 1.046–3.512, p = 0.035). ConclusionPatients with advanced HGSC undergoing NACT with a KELIM <1 were more likely to have platinum-resistant disease, worse PFS and worse OS when compared to patients with KELIM≥1. The KELIM score can be a helpful tool to predict chemo-response and aid in treatment decision making.