Transcriptomic and immunohistochemical approaches identify HLA-G as a predictive biomarker of gestational choriocarcinoma resistance to monochemotherapy

International audience; Objective: Using a transcriptional approach on tissue samples, we sought to identify predictive biomarkers of post molar malignant transformation, and of choriocarcinoma chemosensitivity to mono- (methotrexate or actinomycin D) or polychemotherapy [EMA(Etoposide, Methotrexate, Actinomycin D)-CO(Cyclophosphamide, Vincristine) and EMA-EP(Etoposide, Cisplatine)] regimens.Methods: We studied the expression of a 760-gene panel (PanCancer Pathway) related to oncogenesis and immune tolerance in tissue samples of complete hydatidiform moles and gestational choriocarcinoma.Results: We did not identify any differentially expressed gene between moles with post molar malignant transformation in choriocarcinoma (n = 14) and moles with remission (n = 20). In monochemoresistant choriocarcinoma (n = 34), four genes (HLA-G, COL27A1, IL1R2 and GLI3) had a significantly reduced expression and one (THEM4) had an increased expression [FDR (false discovery rate) adjusted p-value ≤ 0.05] when compared to monochemosensitive choriocarcinoma (n = 9). The proportion of trophoblast cells and the intensity of immunohistochemical HLA-G expression were reduced in monochemoresistant choriocarcinoma (p < 0.05). In polychemoresistant choriocarcinoma (n = 20) we did not identify differentially expressed genes with an FDR adjusted p-value ≤ 0.05 when compared to polychemosensitive choriocarcinoma (n = 15). Gene pathway analysis revealed a predicted activation of IFN ᵞ in monochemoresistant choriocarcinoma and inhibited IL2 and TNF in polychemoresistant choriocarcinoma. The main biological functions predicted to be altered in chemoresistant choriocarcinoma were related to immunological homeostasis and leukopoiesis.Conclusion: HLA-G is a strong candidate gene to predict choriocarcinoma resistance to monochemotherapy and that further studies are required to implement its routine quantification in the decision process for the management of gestational choriocarcinoma.