Defining MRI-based follow-up protocol for primary central nervous system lymphoma

Abstract

Background: The optimal follow-up protocol for primary central nervous system lymphoma (PCNSL) is unclear. This is the first study to evaluate the benefit of structured follow-up imaging of PCNSL with respect to the timing of relapse. Methods: A total of 198 PCNSL patients (57.1% males) diagnosed between 2003 and 2020 were included. The data were collected retrospectively from 8 hospitals in Finland. Relapse detection details included structured follow-up imaging (sMRI) studies, additional imaging (aMRI) studies based on patients’ new symptoms, and all outpatient and emergency visits. Overall survival (OS) with respect to the relapse detection method, sMRI versus aMRI, was also evaluated. Results: Relapse was diagnosed in 71 patients (35.9%), 66.1% of whom experienced relapse during the first 2 years after diagnosis. During the first year, 48.3% (14/29) of the relapses were detected via sMRI, and 51.7% (15/29) via aMRI. During the second year, the percentages were 33.3% and 66.7%, respectively. More than 5 years after the diagnosis, all the relapses were detected via aMRI. To observe one relapse during the first year, 9.4 sMRI studies were needed. Overall survival after relapse (OS2) was 4.0 months for the patients whose first relapse was detected via sMRI and 3.0 months for those whose first relapse was detected via aMRI (P = 0.203). Conclusions: We found that structured imaging was beneficial for relapse detection during the first year after PCNSL diagnosis. A minor trend towards better survival after relapse was observed for patients who experienced relapse according to structured imaging.