Prognostic Significance of ¹⁸f-FDG PET/CT in the Treatment of Primary Central Nervous System Lymphoma

Introduction Primary central nervous system lymphoma (PCNSL) is an aggressive non-Hodgkin lymphoma (NHL) which is confined to the central nervous system at diagnosis. Current standards imaging modality for PCNSL is T1-weighted (T1W) contrast-enhanced magnetic resonance imaging (MRI). However, brain MRI has some limitations of assessing the initial tumor burden or mid-treatment response to predict the clinical outcome. There are limited data exploring the value of ¹⁸F-FDG brain PET/CT in PCNSL. Therefore, this study analyzed the prognostic significance of interim ¹⁸F-FDG brain PET/CT response assessment in patients with PCNCL.

Patients and methods Fifty-six patients with PCNSL between January 2016 and July 2021 were screened for this prospective cohort study. Among the 56 patients, 3 patients did not have baseline ¹⁸F-FDG brain PET/CT scan. Among the remaining 53 patients, 9 patients who had negative finding on their initial PET/CT were excluded and 44 patients who had positive standardized uptake value (SUV) uptakes on lesions were finally included. Induction chemotherapy consisted of HD-MTX (3.5g/m²) on day 1 and cytarabine (2.0g/m²) on days 2 and 3 with or without rituximab (375mg/m²) on day 1 every 3 weeks for transplant eligible patients and HD-MTX (3.5g/m²) alone with or without rituximab (375mg/m²) every 2 weeks for elderly patients, respectively. Interim ¹⁸F-FDG brain PET/CT was tested after induction therapy, and patients who achieved more than PR after induction treatment proceeded to consolidation treatment consisted of etoposide (375mg/m²) and cytarabine (3g/m²) on days 1 and 2. The clinical prognostic factors were assessed according to International Extranodal Lymphoma Study Group (IESLG) guideline. Tumor to normal tissue ratio (T/N ratio) on ¹⁸F-FDG brain was estimated as the ratio of the SUV_max of the brain lesion to the SUV_max of contralateral normal frontal gray matter. If the patients had multiple lesions, the lesion showing the highest SUV_max was evaluated for PET response.

Results All of the patients had been diagnosed with diffuse large B-cell lymphoma. According to the IELSG risk score, 11 patients (25.0%) were high risk, 29 patients (65.9%) were intermediate risk, and 4 patients (9.1%) were low risk. Among 44 patients, 33 patients (75.0%) showed more than partial response (PR) after induction treatment, and 12 patients (36.4%) proceeded to autologous stem cell transplantation (ASCT).

Among the 44 patients, 3 patients experienced disease progression before assessing interim ¹⁸F-FDG PET/CT and these patients were excluded from interim response assessment and survival analysis. Twenty-four patients achieved complete molecular response (CMR), among whom 1 complete response (CR), 2 unconfirmed CR (CRu) and 21 PR were observed based on early brain MRI response assessment. Among 17 patients who did not achieved CMR by interim PET, 8 PR and 9 stable disease (SD)/progressive disease (PD) were assessed by early brain MRI assessment. In 29 patients with MRI-assessed PR, PET was found positive and negative in 8 and 21 cases.

Patients who achieved CMR based on interim PET assessment showed significantly longer PFS than the patients who did not achieved CMR (52.0 months vs. 3.3 months, p<0.001, Figure 1A). OS appeared superior in the patients with CMR in interim PET assessment than in the patients who did not (Not reached vs. 13.8 months, p=0.001, Figure 1B). In 29 patients who were assessed as PR by interim brain MRI, the patients with PET-negative showed longer PFS than the patients with PET-positive although the difference was not reached statistical significance (52.0 months vs. 4.7 months, p=0.07, Figure 1C). OS tended to be longer in the patients with PET-negative (Not reached vs. 13.8 months, p=0.170, Figure 1D).

Conclusion Achievement of CMR in interim ¹⁸F-FDG brain PET/CT has a prognostic significance and may provide additional clinical information of predicting survival outcome in PCNSL.

No relevant conflicts of interest to declare.