IMiD-based monotherapy for maintenance of remission in primary CNS lymphoma: a single-center experience

TO THE EDITOR:

Primary central nervous system lymphoma (PCNSL) is an aggressive extranodal lymphoma with increasing incidence among older adults.¹ Although the median survival of PCNSL patients improved from 12.5 months in the 1970s to 26 months in the current era, outcomes in patients older than 70 years remain poor.^2,3 Standard therapy for PCNSL consists of methotrexate-based induction followed by intensive chemotherapy consolidation.^4,5 A significant proportion of patients with PCNSL are not candidates for consolidation due to age, comorbidities, or functional status. Retrospective studies have demonstrated high rates of morbidity and nonrelapse mortality (up to 22%) among older patients who undergo autologous hematopoietic stem cell transplant or dose-intensive etoposide plus cytarabine (EA) consolidation.^6,7 Consolidative whole brain radiation therapy is often avoided due to neurotoxicity.^8,9

There is no standard of care for patients unfit for consolidation. Median progression-free survival (PFS) is as low as 7 to 12 months from the time of induction in newly diagnosed patients who do not receive consolidation.^10,11 Patients with relapsed PCNSL also have dismal outcomes and limited therapeutic options, with a 1-year median survival of ∼50%.¹²

Lenalidomide is an immunomodulatory drug (IMiD) that is highly active in relapsed and refractory PCNSL.^13,14 We previously described our experience with 13 patients with PCNSL receiving low-dose lenalidomide maintenance after induction and 10 patients with PCNSL treated with lenalidomide after relapse.^13,15 Here, we report long-term follow-up data.

We evaluated the outcomes of IMiD monotherapy as maintenance in 3 categories of patients with PCNSL: (1) low-dose lenalidomide maintenance at a starting dose of 5 mg per day (maximum 10 mg/day) for 21 days/28-day cycle in patients in complete remission after methotrexate-based induction¹⁶; (2) low-dose lenalidomide in patients in complete or partial remission after first salvage therapy; and (3) higher-dose lenalidomide or pomalidomide in patients who progressed on low-dose lenalidomide. IMiD maintenance was continued until progression or intolerance. Patients with PCNSL treated with methotrexate-based induction with subsequent maintenance lenalidomide at the University of California, San Francisco between January 2011 and June 2025 were eligible for inclusion. Retrospective analysis was performed in accordance with a protocol approved by the University of California, San Francisco Institutional Review Board. This study was conducted in accordance with the Declaration of Helsinki. Survival outcomes were estimated using the Kaplan-Meier method and measured from initiation of lenalidomide. A total of 89 patients were identified.

We first evaluated outcomes for 39 patients (36 with PCNSL and 3 with primary vitreoretinal lymphoma [PVRL]) at diagnosis who did not receive dose-intensive consolidation but instead received low-dose lenalidomide maintenance (5 mg/d in 38 patients and 10 mg/d in 1 patient). Their induction therapies and aggregate outcomes are documented in supplemental Table 1. Sixty-four percent were ineligible for dose-intensive consolidation due to age (>73 years), 18% due to performance status/comorbidities, and 18% declined consolidation (supplemental Table 2). Baseline characteristics are listed in supplemental Table 3. With median follow up of 73.9 months, median PFS was 38.5 months (Figure 1A) from start of lenalidomide maintenance and median overall survival was 64.5 months (Figure 1B). Median duration of lenalidomide maintenance was 17.3 months, with duration up to 109.6 months. Notably, 59% (23/39) of these patients did not relapse.

Figure 1.

Outcomes with low-dose lenalidomide maintenance in 2 cohorts of patients with PCNSL. Postinduction (A-B) and postsalvage therapy after relapse (D-E). (A) Kaplan-Meier curve depicting PFS among patients who received low-dose lenalidomide maintenance after induction therapy. (B) Kaplan-Meier curve depicting overall survival for patients who received low-dose lenalidomide maintenance after induction therapy. In the postinduction cohort 1, 38 patients started lenalidomide at a dose of 5 mg/d and 1 patient started at a dose of 10 mg/d. Three of the 38 patients who started at a dose of 5 mg/d were escalated to a dose of 10 mg/d. (C) Distribution of types of salvage therapies received before maintenance low-dose lenalidomide among patients with relapsed/refractory disease. The distribution of salvage therapies included methotrexate plus rituximab (52%), focal radiotherapy (20%), intraocular therapy (14%), EA (8%), ASCT (2%), and WBRT (2%). Of 50 patients, 23 (46%) achieved a complete response with salvage therapy before lenalidomide, and 27 (54%) achieved a partial response. Median PFS from the start of lenalidomide maintenance was 48.8 months for those who achieved CR with salvage vs 10.0 months for those who achieved PR with salvage. (D) Kaplan-Meier curve depicting PFS among patients receiving low-dose lenalidomide maintenance after salvage therapy. (E) Kaplan-Meier curve depicting overall survival for patients who received low-dose lenalidomide maintenance after salvage therapy. In the postsalvage cohort 2, 34 patients started at a lenalidomide dose of 5 mg/d and 16 patients started at a dose of 10 mg/d. Six of the 34 patients who started at a dose of 5 mg/d were escalated to a dose of 10 mg/d. Restaging magnetic resonance imaging scans of the brain were performed with a median frequency of every 3.75 months; restaging ophthalmologic examinations were performed with a similar frequency in patients at high risk for ocular relapse. ASCT, autologous hematopoietic stem cell transplant; WBRT, whole brain radiation therapy; XRT, radiation therapy.

We next evaluated outcomes of 50 patients who had relapsed (44 with PCNSL and 6 with PVRL) treated with low-dose lenalidomide maintenance after salvage therapy (5 mg/d in 34 patients and 10 mg/d in 16 patients). Baseline characteristics are listed in supplemental Table 4 (Figure 1C). With median follow up of 91.4 months, median PFS was 13.4 months from start of lenalidomide maintenance (Figure 1D), and median overall survival was 43.9 months (Figure 1E). Forty-six percent (23/50) achieved complete response (CR) before lenalidomide, and 54% (27/50) were in partial response (PR). Among patients with CR to salvage therapy, PFS with lenalidomide was 48.8 months. Among patients with PR to salvage therapy, PFS with lenalidomide was 10.0 months. The overall median duration of maintenance was 12.2 months, with duration up to 130 months. Thirty percent (15/50) of these patients have not relapsed.

Finally, we evaluated the outcomes of 42 patients from both cohorts (11 from cohort 1 and 31 from cohort 2) who ultimately relapsed on low-dose lenalidomide maintenance and received subsequent therapy. Sixty percent achieved complete response to the next line of therapy, 26% achieved partial response, whereas 14% had refractory disease (Figure 2A). Sixty-two percent of patients went on to receive subsequent IMiD therapy; 27% received lenalidomide at doses exceeding 5 mg (median 10 mg; range 10-15 mg); 73% received pomalidomide (2 mg/d). The median PFS from IMiD re-challenge was 13.3 months (Figure 2B). PFS was 5.95 months for patients treated with higher-dose lenalidomide (supplemental Figure 1A; supplemental Table 5). PFS was 14.3 months for patients treated with pomalidomide (supplemental Figure 1B; supplemental Table 5). Eleven patients re-challenged with an IMiD were in remission at 24 months; 9 of them received pomalidomide.

Figure 2.

Outcomes of patients with PCNSL treated with pomalidomide or higher-dose lenalidomide after lymphoma progression on low-dose lenalidomide maintenance. (A) Distribution of types of salvage therapies received after progression on low-dose lenalidomide maintenance. Of 42 patients who progressed on low-dose lenalidomide, 60% achieved complete response to the next line salvage therapy, 26% achieved partial response, and 14% had refractory disease. Twenty-six out of 42 patients who progressed on low-dose lenalidomide went on to receive IMiD re-challenge after salvage therapy. (B) Kaplan-Meier curve depicting PFS for patients re-challenged with IMiD monotherapy (pomalidomide, N = 19 or higher-dose lenalidomide 10-15 mg, N = 7) after progression on low-dose lenalidomide maintenance. PFS was superior for patients in CR before IMiD re-challenge (15.6 months) compared with those in PR (5.2 months) or PD (4.7 months). XRT, radiation therapy.

Low-dose lenalidomide was well tolerated, with adverse events consistent with its established toxicity profile. Common toxicities are detailed in supplemental Table 6. Discontinuation was uncommon and most often attributable to fatigue and gastrointestinal symptoms. One patient experienced reversible bradykinesia, and there was 1 pulmonary embolism. There was 1 episode of febrile neutropenia. No patient treated with lenalidomide has yet developed a myeloid neoplasm. There was 1 grade 5 adverse event in a patient who had relapsed disease, due to West Nile virus encephalitis, not clearly attributable to lenalidomide.

To our knowledge, our study represents the largest cohort to date receiving lenalidomide maintenance in PCNSL and confirms that its long-term administration is an effective strategy in lieu of dose-intensive consolidation. Lenalidomide maintenance after salvage therapy likewise demonstrates impressive efficacy, with survival outcomes surpassing published historical values. The ANOCEF-GOELAMS study in older patients with PCNSL evaluated distinct chemotherapy approaches without consolidation in a comparable population, reporting a median PFS between 6.1 and 9.5 months, indicating the need for effective postinduction strategies in this population.¹⁰ Although retrospective analyses of high-dose (HD) methotrexate (MTX) salvage in relapsed PCNSL have been described,^17,18 significant limitations preclude comparison with our series: in one study, patients were treated with HD-MTX maintenance monotherapy until progression,^17, and in another, post–HD-MTX treatment strategies were heterogeneous, including autologous hematopoietic stem cell transplant.¹⁸

Importantly, most patients who progressed on lenalidomide maintenance achieve remission with salvage therapy and a significant fraction experience extended response duration with IMiD re-challenge. Notably, to our knowledge, our study also provides the first evidence for significant efficacy of pomalidomide in PCNSL after progression on lenalidomide. Outcomes with lenalidomide maintenance, in particular, a PFS that exceeds 3 years in consolidation-ineligible patients, are encouraging and support further investigations of lenalidomide in PCNSL, including the ongoing Alliance A51901 trial, focused on transplant-ineligible patients.¹⁹ Several unanswered questions remain, including the role of frailty models to guide patient selection and integration of tumor genetic risk biomarkers into maintenance decisions.²⁰ Patients with higher-risk tumor mutational profiles may be prioritized for lenalidomide maintenance.

Limitations of our study include its sample size and retrospective single-center design. Many variables determine eligibility for dose-intensive consolidation, including patient preference.^10,21 Although our study excluded patients with secondary CNSL, the development of novel maintenance strategies in this population is warranted. In addition, the role of lenalidomide maintenance after chimeric antigen receptor T cell in CNS lymphoma remains undefined.²²

In summary, this data set provides further evidence that low-dose lenalidomide maintenance significantly extends response duration and is well tolerated in patients with PCNSL. These findings support the integration of lenalidomide maintenance into clinical practice in PCNSL, particularly for consolidation-ineligible patients, as well as further investigations of IMiDs in this disease.

Conflict-of-interest disclosure: J.L.R. receives research funding from Gilead, Incyte and NURIX. The remaining authors declare no competing financial interests.