This study provides new safety and preliminary efficacy data for immunotherapy in recurrent glioblastoma, suggesting that dual immune checkpoint blockade targeting LAG-3 and PD-1 may enhance anti-tumor immune responses within specific tumor microenvironments. It offers critical guidance for the design of subsequent biomarker-driven clinical trials.
Literature Overview
The article titled 'Anti-LAG-3 with or without anti-PD-1 in recurrent glioblastoma: a phase 1 trial,' published in Nature Medicine, systematically investigates the safety, tolerability, and preliminary immunological effects of the anti-LAG-3 antibody relatlimab as monotherapy or in combination with the anti-PD-1 antibody nivolumab in patients with recurrent glioblastoma (GBM). This multicenter, open-label, phase 1 trial enrolled 46 patients, evaluated dose-escalation regimens, and leveraged a neoadjuvant treatment window to explore the drugs' immunomodulatory effects on the tumor microenvironment (TME). The results indicate that both treatment regimens are tolerable; although the combination therapy led to some grade 3–4 adverse events, no dose-limiting toxicities were observed with monotherapy. Additionally, neoadjuvant treatment was found to promote intratumoral CD8+ T-cell infiltration, and long-term survivors exhibited stronger interferon signaling and T-cell clonality in their tumors, suggesting potential predictive biomarkers of response.Background Knowledge
Glioblastoma (GBM) is a highly aggressive primary brain tumor. Despite maximal safe resection, radiotherapy, and temozolomide chemotherapy, median overall survival remains approximately 14 months. Recurrent GBM after standard therapy is highly resistant to current treatments and generally unresponsive to immune checkpoint inhibitors targeting PD-1 and CTLA-4. The major therapeutic barrier lies in the profoundly immunosuppressive tumor microenvironment (TME), characterized by T-cell exhaustion, enrichment of myeloid-derived suppressor cells (MDSCs), and co-expression of multiple immune checkpoint molecules such as LAG-3, TIM-3, and TIGIT. LAG-3 (CD223), an inhibitory receptor highly expressed on exhausted T cells, promotes immune escape by suppressing CD8+ T-cell function and enhancing regulatory T-cell (Treg) activity. Preclinical studies have shown that combined blockade of LAG-3 and PD-1 can synergistically restore T-cell function, with demonstrated progression-free survival benefits over monotherapy in melanoma. Therefore, researchers hypothesized that targeting LAG-3 in GBM, particularly in combination with PD-1 blockade, could overcome resistance to immunotherapy. This study represents the first systematic evaluation of relatlimab’s safety in GBM and uses a neoadjuvant design to directly analyze the dynamic remodeling of the TME, integrating multi-omics approaches to identify response-associated biomarkers, thereby providing mechanistic rationale for future precision immunotherapy strategies.
Research Methods and Experiments
The study employed a multicenter, open-label, phase 1 clinical trial design (ABTC 1501, NCT02658981), enrolling 46 patients with histologically confirmed recurrent GBM, randomized into a relatlimab monotherapy group (n=23) and a relatlimab plus nivolumab combination group (n=23). The maximum tolerated dose (MTD) was determined through dose escalation: 800 mg for monotherapy and 160 mg relatlimab + 240 mg nivolumab for combination therapy. A subset of patients received neoadjuvant treatment (single preoperative dose) to enable immunological analysis of post-treatment surgical specimens. Safety was assessed using NCI CTCAE v5.0 criteria, and efficacy was evaluated by mRANO criteria. Immunological analyses included: multiplex immunofluorescence (mIF) to assess intratumoral T-cell infiltration; NanoString nCounter platform for gene expression profiling of FFPE samples; T-cell receptor (TCR) sequencing to evaluate T-cell clonality; and MIBI-TOF for spatial proteomics. These assays collectively established a multilayered evidence chain from systemic toxicity to TME remodeling, supporting findings on drug safety, immune activation, and potential response biomarkers.Key Conclusions and Perspectives
Research Significance and Prospects
This study opens a new pathway for immunotherapy in GBM, demonstrating the feasibility of targeting LAG-3 in recurrent disease. Its greatest strength lies in the neoadjuvant design, which enabled paired pre- and post-treatment sample analysis, directly revealing the drugs’ remodeling effects on the TME and providing valuable data for understanding immune checkpoint inhibitor mechanisms in cold tumors. The study emphasizes that T-cell infiltration alone is insufficient to predict response; instead, predictive models integrating baseline interferon signaling, T-cell clonality, and myeloid cell states are essential for future precision patient selection.
From a drug development perspective, this study supports advancing anti-LAG-3 plus anti-PD-1 combination therapy into phase 2 trials in GBM (e.g., Alliance A077201), with recommendations to enrich for patients using interferon gene signatures and T-cell receptor repertoire. Furthermore, the critical role of myeloid cells in treatment response suggests that future combinations targeting myeloid-suppressive pathways (e.g., CSF-1R, CD47) could further enhance efficacy. For clinical monitoring, dynamic assessment of T-cell clonality in peripheral blood may serve as a non-invasive predictive tool.
Conclusion
This study is the first systematic clinical evaluation of anti-LAG-3 therapy in recurrent glioblastoma, establishing the safety of relatlimab as monotherapy and in combination with nivolumab, while providing preliminary immunological and survival signals. Through an innovative neoadjuvant design, it reveals that LAG-3 blockade promotes CD8+ T-cell infiltration and identifies baseline interferon signaling and T-cell clonality as potential response biomarkers, offering key insights into overcoming immunotherapy resistance in GBM. Although not randomized, the 52.2% 12-month overall survival rate in the combination group is markedly higher than historical data, strongly supporting further validation. This work not only advances LAG-3-targeted drug development in neuro-oncology but also highlights the central role of multi-omics analysis in guiding precision immunotherapy. From bench to bedside, these findings lay the foundation for designing biomarker-driven phase II/III trials, potentially improving outcomes for GBM patients and representing a significant step toward reshaping the immunotherapy landscape for GBM. The study also suggests that future success may depend on combination strategies targeting both T-cell exhaustion and myeloid suppression pathways to achieve more durable anti-tumor immune responses.
