Primary vitreoretinal lymphoma (PVRL) is a rare type of non-Hodgkin lymphoma (NHL) of an immune privileged site, defined as a lymphoma arising in the retina and/or vitreous without brain involvement 1. It is estimated that there are 50 new PVRL cases per year in the United States 2. There is no standard of care treatment for PVRL 3 and survival data is limited. Recent data from a retrospective study of 59 patients treated heterogeneously from the French network database showed a median progression-free survival (PFS) and a median overall survival (OS) of 18 and 75 months, respectively 4. Starting in February 2005, we adopted a bilateral external beam ocular radiation therapy (BORT) and systemic high-dose methotrexate (HDMTX)-based combination chemotherapy in all eligible patients with newly diagnosed PVRL that can tolerate this treatment. Patients diagnosed with PVRL were systematically staged with brain magnetic resonance imaging (MRI), lumbar puncture (LP) with flow cytometry and cytology, computerized tomography of the chest, abdomen and pelvis and/or FDG-PET-CT, bone marrow biopsy, serum lactate dehydrogenase, and comprehensive ophthalmologic examination. Patients with isolated vitreoretinal involvement were included in the analysis, that was approved by the Institutional Review Board at the University of Miami. Patients were typically treated with BORT to a dose of 30 to 40 Gray (Gy), given at 1.8 to 3 Gy per fraction, including the optic nerves and conus. Opposed lateral beams of 6 MV were used for bilateral orbits with either half-beam blocks or posterior divergence matching to ensure no excess radiation to the normal tissues including brain, optic chiasm, and brainstem. Prior to 2020, patients received systemic chemotherapy (ChT) after 2/3 of the planned dose was delivered. This was later changed to start ChT within 1 week after BORT. ChT consisted of HDMTX 3.5 g/m2 intravenously (i.v.) over 2 h, followed by leucovorin rescue (15 mg i.v. every 8 h), vincristine 2 mg i.v. push and oral procarbazine 100 mg/m2/day for 7 days (given on odd cycles only), with (12 patients) or without (8 patients) rituximab 500 mg/m2 i.v. ChT was given every 2 weeks for a total of 6 cycles. Patients treated as described comprised the treated per-protocol (TPP) cohort. At the end of treatment, response assessment was performed by ophthalmic examination, brain MRI and LP, if previously involved. Complete response (CR) was defined as the disappearance of lymphomatous infiltrates on post-treatment ophthalmic examination and normal brain MRI. Patients were followed every 3 to 6 months with semi- or annual surveillance brain MRI. Preserved visual function was defined as vision acuity of 20/50 or better. OS was calculated from the date of diagnosis to the date of death or date of last follow-up. Disease-free survival (DFS) was calculated from the time of initial CR to lymphoma progression or death from lymphoma and progression free survival (PFS) as a time from diagnosis to lymphoma progression or death. Patients were censored either when lost to follow-up or after death due to a non-lymphoma cause. The Kaplan–Meier method was used to estimate OS, DFS and PFS. Twelve of the current patients have been previously reported, and updated follow-up of these patients is provided in the current report 5. Twenty-seven patients were diagnosed with PVRL, and 20 comprised the TPP cohort. No patient had extra-ocular neurologic or systemic involvement at initial presentation. In the TPP cohort, the median age was 68.0 years (range: 38–73 years), 6 patients (30%) were Hispanic, and ECOG performance status was 0 in 18 patients and 1 in 2 patients (Table S1). The most common symptom at presentation was eye floaters in eleven patients (55%) with bilateral intraocular involvement present in 17 (85%) patients. Eighteen patients had vitreous involvement, 1 had choroidal and epibulbar mass, and 1 had panuveitis and retinal involvement. Of the 18 patients with vitreous involvement, 5 had concomitant choroidal involvement, one with concomitant choroidal and orbital involvement, 3 with concomitant retinal involvement. Representative eye examination findings before and after treatment are shown in Figure 1. The most common pathology on cytologic and flow cytometry analyses was diffuse large B-cell lymphoma in 19 patients. One patient had marginal zone lymphoma (MZL). TPP led to CR in 19 patients (95%), and partial response in 1 with overall response of 100% (Figure 2A). Details of presentation, treatment and survival are shown in Supplementary Table S1. With a median follow-up time of 44 months (range: 6–215), 8 patients had lymphoma recurrence, including in the brain parenchyma with/without leptomeningeal involvement in 7 and isolated vitreous recurrence in one. The median time to disease recurrence was 29 months (range: 15–158 months). At the last follow-up, 7 of the 8 patients who relapsed had died of lymphoma, and one patient treated with systemic cytarabine, intrathecal methotrexate and autologous stem-cell transplantation (ASCT), achieved and remains in CR for 169 months. Additional 2 patients in the TPP cohort died of non-lymphoma related causes. Regarding the one patient with ocular MZL, he had intracranial relapse 28 months after initial treatment for which biopsy confirmed transformation to DLBCL. This patient died 1 year later due to complications of therapy and neurological decline. In the TPP cohort, the median OS was 15 years (95% CI 5.73–24.27), while median DFS was 13.1 years (95% CI 0.0–32.55). The 5-year DFS, OS, and PFS were 56.9% (95% CI: 30.6%–83.2%), 71.8% (95% CI: 48.28%–95.32%), and 53.2% (95% CI: 27.5%–78.9%), respectively (Figure 2B–D). Sixteen (80.0%) patients developed post-treatment ocular complications with median latency from BORT to complication of 14 months (range: 1–52). At the end of treatment, visual acuity was preserved in both eyes in 10 patients and in 1 eye in 4 patients. Details of ocular baseline exam, end of treatment exam and last exam visual acuity and other complications are shown in Table S2. Additional 7 patients not eligible for the TPP due to advanced age, contraindications for HDMTX, and comorbidities were seen during the same period. Two were treated with BORT alone, two with BORT and intravitreal chemotherapy, one with intravitreal methotrexate and rituximab, and one with intravitreal methotrexate and HDMTX monotherapy at 1.5 mg/m2 due to chronic kidney disease. Progression and relapses were observed in 3 patients (Table S1). Overall, 3 patients died due to lymphoma progression, metastatic brain disease from secondary solid tumor, and dementia. The median DFS was significantly longer in TPP cohort (13.17 years (95% CI, 0.00–32.55)) compared to non-TPP cohort (0.86 years (95% CI, 0.00–3.88)) (log-rank p = 0.049). OS difference was not significant (Figure 3A,B). This cohort represents the largest, uniformly treated cohort of PVRL patients. Here we combined BORT and MPV-based chemotherapy aiming to eradicate intraocular disease and microscopic brain lymphoma that may be a source of frequent brain recurrences, without consolidation ASCT. Due to the rarity of this disease and lack of prospective trials, treatment is largely based on retrospective data. Response to initial therapy signifies the highest possibility of cure in these patients. Reported CR rates in the literature are high, ranging between 70% and 100% 4, 6. Despite a high initial response, 56% and 90% of patients die secondary to central nervous system (CNS) progression 2, 4, 7. In our 20 TPP patients, CR was high at 95%, consistent with what was previously reported 4, 6. Only 8 (40%) had disease progression and 7 died of lymphoma. Two patients died of natural causes while 10 patients (50%) remain disease free. This is notable as these results were achieved without consolidation ASCT or whole brain radiation, as used in recent studies 6, 8. A study by Castellino et al. included 33 patients with PVRL, 15 of which received systemic therapy, but this included two different regimens. In addition, 29% of the overall cohort, which included secondary VRL, received ASCT after first CR. The median failure-free survival was reported at 2.6 years 8. Another study by Lam et al. provides the largest retrospective cohort of PVRL 4. However, 6 patients included in this study had meningeal disease at presentation. The treatment was heterogeneous but included intravenous methotrexate in 57 of 59 patients. Only six patients received ocular radiation at 30 Gy, and 2 patients received intraocular therapy. CR was achieved in 70%. Forty two patients (71%) relapsed, including 17 patients (29%) who had only ocular relapse, suggesting inadequate local control with systemic therapy alone 4. In contrast, only one ocular relapse occurred in our TPP patients. This suggests that our combined modality achieves superior local control. Chemotherapy-free approaches are also being explored. In one study, 44 patients were treated with either Zanubrutinib plus rituximab (ZR) or lenalidomide plus R (R2) achieving a 93.2% CR rate. All patients treated with R2, and 28.6% treated with ZR relapsed 9. The median OS in our cohort was longer (15 years) and is more favorable compared to the French study with median OS of 75 months 4. Overall, this combination treatment represents a highly effective treatment of PVRL. While the treatment was done prospectively, the data collection was done retrospectively, limiting unequivocal conclusion. However, herein we report on a homogenous, combined modality approach that achieves superior local and systemic control without consolidation ASCT. The rarity of this disease precludes prospective trials dedicated to PVRL and hence these data should help to shape guidelines for the treatment of PVRL. A.T. was responsible for the conceptualization, methodology, conducting the research, screening potentially eligible patients, extracting and analyzing data, interpreting results, updating reference lists, writing the manuscript, and creating tables and figures; J.K. was responsible for extracting, writing the manuscript, and creating tables and figures; D.I. was responsible for writing the manuscript, and creating tables and figures; M.C. was responsible for screening potentially eligible patients, and writing the manuscript; S.H. was responsible for analyzing the data; B.K.W. was responsible for writing the manuscript, and creating tables and figures; C.Z. was responsible for collecting the data; J.L.D. was responsible for writing the manuscript, and creating tables and figures; M.D.N. was responsible for writing the manuscript, and creating tables and figures; T.A.A. was responsible for writing the manuscript, and creating tables and figures; I.S.L. was responsible for the conceptualization, methodology, conducting the research, interpreting results, and writing the manuscript. The authors have nothing to report. This work was approved by an Institutional Review Board (IRB) and complied with the Declaration of Helsinki, ensuring ethical standards and participant safety. This is a retrospective study and used no patient identifying information and was approved by the IRB. Individual informed consent was waived and not required. A.T., D.I., M.C., J.L.D., M.D.N., T.A.A., S.H., C.Z., and J.K. disclose no conflicts of interest. I.S.L. Research support from Genentech and ADC therapeutics. B.K.W.—Consults for AbbVie, Alcon, Alimera, Astellas, DORC International, EyePoint, Genentech/Roche, Immunocore, Ocular Therapeutix, and Regeneron and has stock options in Lumata Health. Deidentified datasets generated during and/or analyzed during the current study are available from the corresponding author. Table S1: Characteristics of patients with primary vitreoretinal lymphoma. Table S2: Radiation-induced ocular toxicity. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.
Trabolsi et al. (Fri,) studied this question.