Natural killer cell genomic profile is associated with treatment outcome in patients with chronic HCV infection

Phenotypic and immunogenetic characteristics of NK cells have been recently associated with treatment outcome of HCV infected patients. However, thus far, no global NK patterns predictive of response to antiviral therapy in HCV infection have been identified. We here analyzed phenotypic and transcriptional characteristics of NK cells derived from prospectively followed patients with chronic HCV infection prior to treatment with PEG-IFN/RBV. The results report that the pre-treatment transcriptional patterns of purified NK cells from patients subsequently undergoing a sustained virologic response (SVR) clearly segregated from those of non-responder (NR) patients and surprisingly overlapped with the genomic profile of NK cells from healthy donors and from patients who spontaneously cleared the virus (SR). The treatment outcome was associated with differential surface expression of NKp30 and NKG2D and with the transcriptional expression of molecules involved in post-transcriptional modifications of RNA/protein trafficking and HLA class II signaling. With the development of a highly stringent prediction model (LOOCV p value < 0.00001) we identified a 33 gene signature whose expression predicted with 100% accuracy the outcome of IFN based treatment. None of the genes found to differentiate patients with SVR from NR were related to the IL28B polymorphism. Neither an IL28B CC homozygosis was significantly and clearly associated with SVR in our series of patients. Thus, NK cells are associated to different HCV disease courses and response to treatment and the NK cell evaluation in HCV patients might provide a comprehensive explanation of useful determinants of clinical response in HCV-infected patients subjected to treatment regimens that have IFN as their backbone.

40