Immunoprofiling in alveolar soft part sarcoma.

Abstract

11059 Background: Alveolar Soft Part Sarcoma (ASPS) is a distinctive tumor characterized by a canonical ASPL-TFE3 fusion. Treatment options are limited. We assessed tumor immune cell infiltrates, and correlated this with patients receiving PD-1 blockade. Methods: A retrospective institutional review was performed for 18 cases of ASPS. Immunohistochemistry was performed on paraffin-embedded tissue (PET) for T-lymphocyte markers (CD3/CD4/CD8), and PD-1/PD-L1 (Ventana). Expression was quantified by standard methods: (total cells per high power field: score; 0:0; 1-10:1; 11-50:2; 50-99:3; 100:4). Select cases underwent DNA sequencing analysis using whole exome (WES, > 80X, n = 4) and genome (WGS, > 30X, n = 1) sequencing. Indel analysis was conducted via mutect2 ( > 5% variant allele frequency) and mutational signature was performed using deconstructSigs. Results: The median age was 27 (15-54). Disease status at diagnosis was: 44% localized; 56% metastatic. The median overall survival was 17 yrs (2.9-31). Four patients (pts) received immunotherapy with PD-1 blockade with 1 complete response (CR), 2 durable partial responses (PR) and 1 stable disease (SD). PET was available in 12 cases. PD-1/PD-L1 expression (≥1) was seen in 50% and 17%, respectively. Composite CD3, CD4 and CD8 infiltration were 2, 1, and 1, respectively. Patients with CR/PR to PD-1 blockade (n = 3) had no clear correlation with PD-1, PD-L1 or lymphocyte markers. Exomic characterization (n = 4) demonstrated no clear excess mutation burden compared to Ewing sarcoma (5.7 vs 6.4 mut/MB). Mutational signatures via COSMIC were identified in the mismatch repair (MMR) pathway in 2 of 4 cases (Pt A: Signature (S) 26; pt B: S6 and S15). Pt B also underwent WGS which confirmed a COSMIC signature in the MMR pathway. Indel analysis did not confirm aberrations in standard MMR or polymerase genes. Conclusions: Preliminary findings suggest activity with PD-1 blockade in ASPS; however, this does not appear to correlate with tumour-infiltrating T lymphocytes. Genomic analysis suggests an MMR signature may account for these responses, but standard MMR aberrations were not identified. Further validation is underway.