Nomograms for predicting the overall and cause-specific survival in patients with malignant peripheral nerve sheath tumor: a population-based study.

Abstract

Malignant peripheral nerve sheath tumor (MPNST) is extremely rare in soft tissue sarcoma, with a high rate of recurrence and metastasis. Due to its rarity, the epidemiological features and prognostic factors are still uncertain. Moreover, nomograms for patients with MPNST have not been constructed and validated until now. Patients diagnosed with MPNST between 1973 and 2014 were selected from the Surveillance, Epidemiology, and End Results (SEER) database. Survival analysis, machine learning and Lasso regression were used to identify the prognostic factors for overall survival (OS) and cause-specific survival (CSS). Significant prognostic factors were integrated to construct nomograms and then the nomograms were validated externally with a separate cohort from our own institution. A total of 689 patients were included in the training set and 42 patients in the validation set. Multivariate analysis suggested that age, histology, historic stage and chemotherapy were independent prognostic factors for OS and primary site, surgery, historic stage and chemotherapy for CSS. The nomograms based on multivariate models were developed and validated for predicting 3- and 5-year OS and CSS, with a C-index of 0.686 and 0.707, respectively. In the external validation set, the C-index was 0.700 for OS and 0.722 for CSS. ICD-O-3 histology, historic stage and chemotherapy were independent prognostic factors for OS and primary site, surgery, historic stage and chemotherapy for CSS. The constructed nomograms could provide individual prediction for MPNST patients and assist oncologists in making accurate survival evaluation.