Combined immunochemotherapy in selected patients with metastatic renal cell carcinoma: HLA class II genotype can help to predict response to therapy.

Bierer S, Hoffmeister I, Gerss J, Herrmann E, Wülfing C, Sibrowski W, Hertle L, Kelsch R

Abstract

A number of new agents have been approved for systemic therapy of metastatic renal cell carcinoma (mRCC) recently. Thereby, prognostic factors may aid in predicting the effectiveness of various treatment modalities in individual cases. Aim of this study was to determine the value of human leukocyte antigen (HLA) class II characteristics in predicting response of mRCC to combined immunochemotherapy (ICT). A retrospective study of 29 patients with mRCC treated with ICT was performed: 17 patients (group A) with long-term remission and 12 (group B) with progressive disease after ICT. DNA was used for high resolution typing of HLA-DRB1, -DRB3, -DRB4, -DRB5, -DQA1, and -DQB1. Statistical evaluation started with Classification and Regression Trees analysis. The assignment of single alleles to the groups was then aggregated to create a classification on a patients' basis. Finally, the accuracy of this test algorithm was evaluated. HLA-DRB1 (DRB1*0301*0401*0402*0407*1101*1501=progression) was the strongest discriminator between the 2 groups. The test algorithm defined all patients with at least one of these DRB1 alleles to be progressive after ICT. Thus, 12 of 12 patients of group B could have been identified as progressive (sensitivity=100%). However, only 10 of 17 patients of group A would have been identified as responding (specificity=58%). Thus, the test had a positive and negative predictive value of 63% and 100%, respectively. Approximately 5% to 10% of all patients with mRCC are able to benefit from ICT with long-term remission. HLA class II characteristics may aid in identifying this small subgroup of patients with mRCC.