High-risk soft tissue sarcomas treated with perioperative chemotherapy: Improving prognostic classification in a randomised clinical trial

BACKGROUND: Patients with extremity and trunk wall soft tissue sarcoma (STS) with high malignancy grade and size >5 cm are at high-risk of death. This risk varies depending also on other patient and tumour features, including histologic subtype. This study investigated whether a prognostic nomogram can improve risk assessment of these patients. METHODS: Data from high-risk STS patients enrolled in a randomised controlled trial investigating different perioperative chemotherapy regimens were analysed. Ten-year probability of overall survival (OS) and incidence of distant metastasis (DM) were computed using the prognostic nomogram Sarculator (pr-OS and inc-DM, respectively). Tumour response according to RECIST and Choi criteria was also investigated. FINDINGS: Variation in pr-OS and inc-DM were observed and patients stratified in three prognostic groups. The 10-year OS in the low, intermediate, and high pr-OS categories were 0·42 (95%CI 0·32-0·52), 0·63 (95%CI 0·53-0·72), and 0·78 (95%CI 0·68-0·85), respectively. Patients in the intermediate (HR 0·51, P = 0·002) and high (HR 0·28, P < 0·001) pr-OS categories were at statistically significant lower risk of death compared with those in the low pr-OS category. Higher rate of Choi partial tumour responses were detected in intermediate pr-OS category. Tumour response according to Choi but not to RECIST criteria stratified patient survival of pr-OS categories, particularly for patients with intermediate to low pr-OS. Analyses conducted for 10-year inc-DM were consistent with results for pr-OS for prognostic value of Sarculator predictions and Choi tumour response. INTERPRETATION: Sarculator identifies variations in outcomes of high-risk STS treated with perioperative chemotherapy and improve prognostic classification, which is also associated with different patterns of tumour response, an outcome that further stratifies survival particularly for patients predicted at higher risk. Future trials investigating neoadjuvant chemotherapy should consider prognostic tool for selecting patients to be enrolled.