The 29th Annual Meeting of the European Hematology Association (EHA) was grandly held from June 13 to 16, 2024, in Madrid, Spain. As the largest international conference in the field of hematology in Europe, it attracts experts and scholars from around the world every year to share and discuss innovative ideas and the latest scientific and clinical research results in hematology. At this year's conference, a study by Professor Liang Wang's team from Beijing Tongren Hospital, Capital Medical University, was selected for poster presentation (P1187). The study, based on a PRISMA-compliant meta-analysis and systematic review, explored the best treatment options for vitreoretinal lymphoma (VRL). To provide a better understanding of the research findings, "Oncology Frontier - Hematology Frontier" invited the first author of the study, Dr.  Jing Gao, to introduce the research in detail and Professor Liang Wang to provide an in-depth commentary.

Background and Purpose

Vitreoretinal lymphoma (VRL), a subtype of central nervous system lymphoma (CNSL), generally has a poor long-term prognosis, with approximately 65% to 85% of patients eventually developing CNSL. Currently, there is no unified international treatment standard for VRL. Treatment methods vary and include methotrexate-based local or systemic therapy, local or whole-brain radiotherapy, and chemotherapy combined with hematopoietic stem cell transplantation (IC+ASCT). Additionally, BTK inhibitors, temozolomide, and rituximab are being explored as potential treatment options, but their efficacy and safety in VRL treatment remain unclear. This study aims to provide scientific evidence for the best treatment options for VRL through a systematic review and meta-analysis.

Research Methods

This study retrospectively collected all literature related to VRL treatment published on PubMed, Embase, and Scopus up to November 2023. The primary endpoints were objective response rate (ORR) and complete response rate (CRR), while secondary endpoints included overall survival (OS) and progression-free survival (PFS). OR, CR, OS, and PFS data extracted from the literature were combined using forest plots. Survival data were analyzed using the Kaplan-Meier method and log-rank test, and survival curves were plotted for comparison.

Research Results

A total of 37 studies were included in this systematic review, of which 10 were prospective studies and 27 were retrospective studies. The follow-up period ranged from 0.2 to 246 months, with a median follow-up time of 29.5 months. A total of 801 patients were involved, aged 16 to 90 years, with a median age of 64 years. Among the 660 patients with available gender information, 259 were male and 401 were female. Treatment measures included intravitreal injections (such as methotrexate, rituximab), high-dose systemic methotrexate (HD-MTX), targeted therapy (such as Bruton’s tyrosine kinase (BTK) inhibitors), radiotherapy, and other methods (such as intrathecal methotrexate, autologous hematopoietic stem cell transplantation).

When evaluating the efficacy of the treatment methods, we found that the complete response rate (CRR) and objective response rate (ORR) reached 84.8% (95% CI: 0.758 to 0.920) and 93.3% (95% CI: 0.867 to 0.978), respectively. Notably, the combination of intravitreal low-dose methotrexate (IV-MTX) and systemic high-dose methotrexate (sysMTX), as well as radiotherapy (RT) combined with methotrexate-based chemotherapy, both showed high response rates exceeding 90%. In contrast, the response rates for BTK inhibitors (BTKi) alone or methotrexate-based chemotherapy alone were relatively lower but still exceeded 70%. The detailed results are shown in Figure 1. These findings indicate that traditional treatment methods such as radiotherapy, IV-MTX, and sysMTX are highly effective in treating VRL. The study also explored the potential of targeted drugs like BTK inhibitors as monotherapy. Combining systemic and local treatment strategies appears to achieve higher response rates.

Regarding the duration of efficacy, the median progression-free survival (mPFS) and median overall survival (mOS) for all treatment methods were 22.871 months (95% CI: 17.345 to 30.159 months) and 52.519 months (95% CI: 40.903 to 64.077 months), respectively. The study also found significant differences in PFS and OS among patients receiving different numbers of intervention treatments (P-values were 0.0081 and 0.007, respectively), indicating that increased interventions prolonged PFS and OS. Patients receiving systemic treatment showed significant improvement in PFS and OS compared to those who did not (P-values were 0.00098 and 0.0091, respectively). Additionally, age, gender, and whether both eyes were affected did not significantly impact postoperative survival time (P-values all greater than 0.05).

Research Conclusions

Our meta-analysis of existing studies shows that current treatment strategies yield high response rates and prolonged OS in VRL patients. However, PFS remains suboptimal, with early relapse and progression noted in most studies. Therefore, combining multiple treatment modalities and systemic treatment may be the primary options for extending relapse-free and overall survival time.

Expert Commentary

Professor Liang Wang: Vitreoretinal lymphoma (VRL) is a rare disease, and most patients experience delays in diagnosis or misdiagnosis. Diagnostic vitrectomy remains the gold standard for diagnosing VRL. However, due to the small sample size of vitrectomy specimens, it is often insufficient for completing morphological, flow cytometric, gene rearrangement, and cytokine tests, resulting in inconsistent positive rates of vitrectomy worldwide. For patients with highly suspected VRL based on clinical symptoms and ophthalmic imaging, a clinical diagnosis of VRL can be made if the IL-10/IL-6 ratio in aqueous humor or vitreous fluid is greater than 1, with or without MYD88 mutation. However, we strongly recommend completing diagnostic vitrectomy before medication to maximize the pathological diagnosis as the gold standard.

Currently, there are no guidelines recommended for VRL treatment both domestically and internationally. Due to the extremely low incidence, high-evidence RCT studies are difficult to conduct, and previous studies have mostly been small-sample phase II IIT studies or retrospective studies. Despite this, more and more ophthalmology and hematology experts recognize that VRL is a special subtype of CNSL, and without active CNS prevention, most VRL patients will progress to CNSL within 1-2 years, significantly affecting their survival. Therefore, the limited consensus among VRL treatment experts suggests that CNS prevention treatment should be conducted for VRL patients, but the best prevention strategy is unclear. In our team’s earlier studies on VRL patients, oral BTK inhibitors showed high local control rates and some degree of CNS prevention effect, but long-term results were still unsatisfactory. Through this meta-analysis and systematic review of VRL, we clearly see that systemic high-dose MTX combined with local treatment (intravitreal low-dose MTX or RT) can achieve better response rates and long-term survival. Additionally, whether the lesion is in one eye or both eyes, early CNSL recurrence and progression can occur. Therefore, we cannot decide on CNS prevention based solely on whether the lesion is in one or both eyes. The diagnosis and treatment of VRL require close multidisciplinary collaboration (MDT), and the combined use of multiple administration modes can bring better prognosis to VRL patients.