Evolving of therapeutic strategies for CNS-PNET

Abstract

A protocol for the intensive treatment of non-cerebellar PNET (CNS-PNET) combining chemotherapy and radiotherapy was launched in 2000. Efforts were subsequently made to improve the prognosis and to de-escalate the treatment for selected patient groups. Twenty-eight consecutive patients were enrolled for a high-dose drug schedule (methotrexate, etoposide, cyclophosphamide, and carboplatin ± vincristine), followed by hyperfractionated accelerated CSI (HART-CSI) at total doses of 31-39 Gy, depending on the patient's age, with two high-dose thiotepa courses following CSI. After the first 15 patients had been treated, craniospinal irradiation (CSI) was replaced with focal radiotherapy (RT) for selected cases (non-metastatic and not progressing during induction chemotherapy). Eight of the 28 children received the same chemotherapy but conventionally fractionated focal RT at 54 Gy. The 5-year progression-free survival (PFS), event-free survival (EFS), and overall survival (OS) rates were 62%, 53%, and 52%, respectively, for the whole series, and 70%, 70%, and 87% for the eight focally irradiated children. Residual disease and metastases were not prognostically significant. In children with residual disease, response to RT was significant (5-year PFS 59% vs. 20%, P = 0.01), while the total dose of CSI was not. There were three treatment-related toxic events. Relapses were local in seven cases (including two of the eight focally irradiated patients), and both local and disseminated in 2. This intensive schedule enabled treatment stratification for the purposes of radiation, thereby sparing some children full-dose CSI. Local control is the main goal of treatment for CNS-PNET.