21TiP Molecular alterations predictive of outcome in early staged cervical cancer: A translational investigation in the international validation study of sentinel node biopsy in early cervical cancer SENTICOL III

Cervical cancer remains the second cause of cancer death in women worldwide. Despite of good histological and clinical features, some patients relapse after the treatment of early stage cervical cancer. Recent studies identified correlation between some genetic alterations and poor prognosis in advanced-staged cervical cancer. Yet, the literature remains scarce about molecular alterations and outcome correlation in early stage cervical cancer. Our translational study investigate recurrent molecular alterations predictive of outcome in early stage cervical cancer. We included the first 100 patients randomized in the Senticol III trial. This large multicentric, prospective, randomized, and international “validation study” tries to validate sentinel node biopsy as nodal staging of early cervical cancers (stage Ia – IIa1). Cervical tumor slides with contributive FFPE sample are analyzed and stratified based on well-established histological criteria (SEDLIS criteria). The immune microenvironment characteristics including TIL’s infiltration and PDL1 expression. PDL1 expression is assessed by IHC using the 22C3 antibody and quantified using CPS score. We made HPV detection and typing by PCR of the tumor samples. We performed DNA and RNA extractions from the FFPE tumor specimens. Using the dedicated gene panel developed by our team, we analyzed 571 genes commonly altered in cancer. We performed high throughput RNA sequencing to establish the gene expression profile of each tumor and its associated stroma. The genomic and transcriptomic analysis assessed the tumor mutational load, the most frequently altered genes and their expression. The biostatistical analyses will correlate molecular alterations, histopathological and clinical classical features, with patient outcome. The different parameters will be first analyzed independently (univariate analysis) and then in a multiparametric manner (logistic regression). NCT03386734. Fabrice Lecuru, MD PhD.