Background: Several classifications have been used for pseudomyxoma peritonei (PMP), and among these, the Ronnett classification is the most commonly used. However, a new consensual Peritoneal Surface Oncology Group International (PSOGI) classification has recently been proposed. Nonetheless, to date, the ability of the PSOGI classification to predict survival based on its different disease histologic categories has not been validated.
Methods: This study enrolled 117 patients with PMP who had undergone cytoreductive surgery and hyperthermic intraperitoneal chemotherapy (HIPEC) between 1997 and 2020. Cox proportional hazards regression models and time-dependent curve receiver operating characteristic (ROC) analyses were used to assess the predictive capacity of both classification systems for the overall survival (OS) and disease-free survival (DFS) of these patients.
Results: Significant differences in the 5-year OS rate were found for the different histologic grades according to each of the classifications. The completeness of cytoreduction score (CCS) was identified as a factor that predicted patient OS prognosis (p = 0.006). According to the time-dependent ROC curves at the "100" time point, adjusted by the CCS and DFS, the capacity to predict OS was optimal and achieved an area under the curve (AUC) of about 69% for OS and approximately 62% for DFS.
Conclusions: Both the Ronnett and PSOGI classifications were able to predict survival optimally for this patient cohort. However, when the classifications were adjusted by the CCS, the predictive availability for OS was better with the PSOGI classification than with the Ronnett classification.