Spatial distribution of tumour immune infiltrate predicts outcomes of patients with high-risk soft tissue sarcomas after neoadjuvant chemotherapy

Sandro Pasquali,Viviana Vallacchi,Luca Lalli,Paola Collini,Marta Barisella,Cleofe Romagosa,Silvia Bague,Jean Michel Coindre,Angelo Paolo Dei Tos,Emanuela Palmerini,Vittorio Quagliuolo,Javier Martin-Broto,Antonio Lopez-Pousa,Giovanni Grignani,Jean-Yves Blay,Robert Diaz Beveridge,Elena Casiraghi,Silvia Brich,Salvatore Lorenzo Renne,Laura Bergamaschi,Barbara Vergani,Marta Sbaraglia,Paolo Giovanni Casali,Licia Rivoltini,Silvia Stacchiotti,Alessandro Gronchi
DOI: https://doi.org/10.1016/j.ebiom.2024.105220
IF: 11.205
EBioMedicine
Abstract:Background: Anthracycline-based neoadjuvant chemotherapy (NAC) may modify tumour immune infiltrate. This study characterized immune infiltrate spatial distribution after NAC in primary high-risk soft tissue sarcomas (STS) and investigate association with prognosis. Methods: The ISG-STS 1001 trial randomized STS patients to anthracycline plus ifosfamide (AI) or a histology-tailored (HT) NAC. Four areas of tumour specimens were sampled: the area showing the highest lymphocyte infiltrate (HI) at H&E; the area with lack of post-treatment changes (highest grade, HG); the area with post-treatment changes (lowest grade, LG); and the tumour edge (TE). CD3, CD8, PD-1, CD20, FOXP3, and CD163 were analyzed at immunohistochemistry and digital pathology. A machine learning method was used to generate sarcoma immune index scores (SIS) that predict patient disease-free and overall survival (DFS and OS). Findings: Tumour infiltrating lymphocytes and PD-1+ cells together with CD163+ cells were more represented in STS histologies with complex compared to simple karyotype, while CD20+ B-cells were detected in both these histology groups. PD-1+ cells exerted a negative prognostic value irrespectively of their spatial distribution. Enrichment in CD20+ B-cells at HI and TE areas was associated with better patient outcomes. We generated a prognostic SIS for each tumour area, having the HI-SIS the best performance. Such prognostic value was driven by treatment with AI. Interpretation: The different spatial distribution of immune populations and their different association with prognosis support NAC as a modifier of tumour immune infiltrate in STS. Funding: Pharmamar; Italian Ministry of Health [RF-2019-12370923; GR-2016-02362609]; 5 × 1000 Funds-2016, Italian Ministry of Health; AIRC Grant [ID#28546].
What problem does this paper attempt to address?