CA-125 KELIM as an Alternative Predictive Tool to Identify Which Patients Can Benefit from PARPi in High-Grade Serous Advanced Ovarian Cancer: A Retrospective Pilot Diagnostic Accuracy Study.
Dimitrios ZouzoulasDimitrios TsolakidisPanagiotis TzitzisKimon ChatzistamatiouVasilis TheodoulidisIliana SofianouGrigoris GrimbizisEleni TimotheadouPublished in: International journal of molecular sciences (2024)
BRCA mutation and homologous recombination deficiency (HRD) are the criteria for the administration of PARP inhibitor (PARPi) maintenance therapy. It is known that PARPi efficacy is related to platinum sensitivity and that the latter can be demonstrated from the CA-125 elimination rate constant (KELIM). This study aims to investigate if KELIM can be another tool in the identification of patients that could be benefit from PARPi therapy. Retrospective analysis of patients with high-grade serous advanced ovarian cancer that underwent cytoreduction and was further tested for HRD status. The HRD status was tested either by myChoice HRD CDx assay or by RediScore assay. KELIM score was measured in both neoadjuvant and adjuvant settings with the online tool biomarker-kinetics.org. A total of 39 patients had available data for estimating both HRD status and KELIM score. When assuming KELIM as a binary index test with the value 1 as the cut-off point, the sensitivity was 0.86, 95% CI (0.64-0.97) and the specificity was 0.83, 95% CI (0.59-0.96). On the other hand, when assuming KELIM as a continuous index test, the area under the curve (AUC) was 81% and the optimal threshold, using the Youden index, was identified as 1.03 with a sensitivity of 85.7% and a specificity of 83.3%. KELIM score seems to be a new, cheaper, and faster tool to identify patients that can benefit from PARPi maintenance therapy.
Keyphrases
- high grade
- end stage renal disease
- chronic kidney disease
- newly diagnosed
- ejection fraction
- peritoneal dialysis
- prognostic factors
- dna damage
- low grade
- stem cells
- squamous cell carcinoma
- randomized controlled trial
- radiation therapy
- clinical trial
- machine learning
- dna repair
- oxidative stress
- lymph node
- social media
- locally advanced
- patient reported
- high throughput
- breast cancer risk