HLA allele, TCR V- and J-gene segment usage combinations and their association with survival in neuroblastoma.

Abstract

10042 Background: Neuroblastoma has variable outcomes across different risk groups. In children with stage 4 neuroblastoma, five-year overall remains around 50% in high-risk children despite the emergence of anti-GD2 antibodies. T- and NK-cell infiltration is prognostic in therapy-resistant neuroblastoma, and higher HLA class I expression is linked to better overall survival (OS). In other cancers, specific HLA alleles and T-cell receptor (TCR) V, J- gene segments have been associated with survival. Thus, we conducted a retrospective study in stage 4S neuroblastoma patients to assess whether specific HLA allele, TCR V- and J-gene segment usage combinations correlated with OS in NBL. Among combinations that were associated with OS, we also identified changes in expression of immune marker genes. Methods: We obtained HLA allele data from exome files of the TARGET-NBL dataset using the xHLA software. The TCR recombination reads were obtained from the TARGET-NBL RNAseq files representing tumor specimens from 99 cases, utilizing a high-stringency search algorithm. The TCR recombination reads were translated, and the complementarity determining region-3 (CDR3) amino acid sequences were obtained. HLA and TCR datasets were integrated to assess OS probabilities, comparing cases with and without specific HLA allele, TCR V- or J-gene usage combinations. Significance was determined only if independent HLA allele or V- and J-gene usage assessments were not statistically significant, but significant in the corresponding HLA allele, TCR V- or J-gene segment usage combinations. HLA allele and TCR usage combinations were grouped by association with better or worse OS probabilities, and immune marker gene expression correlations were assessed via Student’s t-test and Mann-Whitney U test with a Bonferroni-corrected threshold of p = 0.00114. Results: We identified 73 HLA allele, TCR V- and J-gene usage combinations with significant OS distinctions: 20 associated with improved OS and 53 with worse OS. For example, 20 TARGET-NBL cases with the HLA-DQB1*04:02 and TRAJ29 usage combination did not reach the median compared to the 1319-day OS median for all remaining cases (log-rank p = 0.009). Among the cases with at least one HLA allele, TCR V- or J-gene segment usage combination with improved OS, we found that the RNAseq values for the immune markers CD4, CD22, CD38, RPH1 , TNFRSF17 , and TNFRSF13B were upregulated, as assessed via a Mann-Whitney U analysis. Conclusions: Identifying specific HLA allele, TCR V- and J- gene segment usage combinations associated with survival may further indicate patients who could benefit from immunologic-boosting treatments. Studies employing functional assays, immunogenomic profiling, and targeted immune pathway analyses may advance immunotherapeutic strategies and predictive biomarkers for neuroblastoma, particularly in high-risk patients.