Masquerade syndromes are known in ophthalmology to mimic eye infections. One of these diseases, vitreoretinal lymphoma (VRL), a subtype of B-cell lymphomas, affects the retina and vitreous body and represents an example of clinical mimicry as there are developed similarities between the features of this lymphoma and infectious retinitis. Common symptoms of VRL include blurred vision and floaters. The characteristic features of VRL are retinal white lesions, necrosis, vitreitis and punctate yellow-white infiltrates in subretinal pigment epithelium.1 However, VRL may be associated with several symptoms, which mimic viral retinitis or retinal vasculitis, and is therefore recognized as a masquerade syndrome.
As a consequence, the diagnosis of lymphoma is often substantially delayed, leading to unnecessary exposure to anti-viral or anti-inflammatory drugs and lymphoma dissemination, instead of a timely start of chemotherapy. Why is it so important to promptly verify the etiology of retinitis? The fast destruction of retinal tissue demands immediate diagnosis in order to commence the appropriate therapy and impede the progression of retinitis. An effective approach to distinguish features and patterns of the lesions observed in VRL from the features of viral and protozoal retinitis has recently been found in a novel study by Marchese et al., published in a recent issue of Ocular Immunology and Inflammation.2 The leading authors of the paper, Dr. Alessandro Marchese from the University Vita- Salute San Raffaele, Milano, Italy, and Dr. Aniruddha Agarwal from the Postgraduate Institute of Medical Education and Research (PGIMER), Chandigarh, India, were invited by PIE magazine to comment on their paper.
Since no central database for VRL is established, it is difficult to measure the incidence of this rare form of lymphoma. Nevertheless, about 380 cases of VRL are diagnosed in the United States annually.3 Dr. Agarwal believes that “the total incidence of vitreoretinal lymphoma is increasing”. This might be explained by improved diagnostic methods, such as vitreous biopsies, serum interleukins, flow cytometry, targeted next generation sequencing, MYD88 oncogene detection and retinochoroidal biopsies. Moreover, according to Dr. Agarwal, “there is also an increase in central nervous system (CNS) lymphomas all over the world. The involvement of the eyes is approximately 20% of CNS lymphoma”. He also added that in their settings of more than 600 uveitis patients per month, they diagnose VRL in at least two new patients every month.
How often can retinitis-like lesions be detected in patients with VRL? The authors indicated that these lesions were observed in 6 eyes out of 76 eyes of 38 patients, involved the peripheral retina, and masqueraded as infectious retinitis. Earlier studies have reported retinal infiltrates formed by atypical B-lymphocytes.4 Furthermore, a clinical case of VRL manifested with blurry vision in the right eye and accompanied by retinal hemorrhages has also been published. Although infectious retinitis was originally suspected, PCR testing disproved infectious genesis of the eye disorder.5
What are the key imprints specific to VRL revealed by Marchese et al.? The most prominent feature was massive retinal thickening in the areas resembling retinitis. Furthermore, the infiltrates in the subretinal space and retinal pigment epithelium also distinguished VRL from other types of retinitis. In contrast to the VRL-initiated lesions, anterior uveitis and iris atrophy were found in patients with viral retinitis. Moreover, the authors observed more intraretinal cysts in viral retinitis than in VRL lesions. Also, toxoplasmic lesions showed less hemorrhages in the retina and were associated with chorioretinal scars.
Why does VRL mimic viral retinitis? Could viruses be the triggers, which induce VRL development? The authors of the article stated that VRL has been associated with several infectious agents, such as herpes viruses, as metagenomic deep sequencing detected viral pathogens, which may potentially be involved in the development of VRL.6 In particular, the Epstein-Barr virus has been recognized as a risk factor for the development of VRL. Besides, VRL has been strongly associated with immunodeficiency and immunosuppression.
There are several methods for VRL characterization with non-imaging techniques, such as optical coherence tomography (OCT).7 Typical imaging features include hyper-reflective infiltrates in inner retinal and subretinal layers, deposits in pigment epithelium and agglomerates of vitreous cells. In particular, the structure of hyper-reflective infiltrates has been suggested to be highly specific for VRL. Besides, vitreous cells in sheets and the absence of cystoid macular edema also indicate towards VRL. Moreover, OCT analysis helps to distinguish between VRL and choroidal lymphoma, since the cellular infiltrates are detected between the RPE and Bruch’s membrane in VRL patients, while they are located deep to Bruch’s membrane in the cases of choroidal lymphoma.
In the current paper, the authors strongly advocate the use of OCT, a highly sensitive and non-invasive technique, which allows high-resolution analysis of retinal tissue and detection of retinal structural and ultrastructural alterations and intraretinal microangiopathic lesions. The OCT scanning, which was described in the current paper, has not been performed earlier for distinguishing VRL from microbial retinitis. Furthermore, VRL lesions, which resemble retinitis, have never been compared to viral retinitis. Notably, all patients with VRL included in the study showed negative for viral and Toxoplasma PCR.
Are there other imaging techniques that may help to diagnose VRL in addition to OCT? Earlier clinical studies have applied fluorescein to detect punctate hyperfluorescent defects, hypofluorescent loci and granularity. In addition, indocyanine green angiography detects hypofluorescent lesions. As Dr. Marchese pointed out, the authors “have not analyzed other multimodal techniques to differentiate retinitis lesions of VRL from other causes. However, on fundus photographs, sharp edges of retinitis were significantly associated with viral forms and not VRL”. Also, Dr. Agarwal noted that in addition to OCT, “fluorescein angiography is a traditional tool that demonstrates the important features of VRL such as leopard skin appearance. On the other hand, viral retinitis appears highly hyperfluorescent with associated arteritis on fluorescein angiography. Toxoplasma chorioretinitis also appears hyperfluorescent with central hypofluorescence, and has associated phlebitis, such as Kyrieleis arteritis, which can also be evaluated on indocyanine green angiography”.
Another study, recently published by Marchese et al., has revealed several features of vitreous haze using ultrawide-field imaging, ophthalmic ultrasonography and slit-lamp photography.8 The observations in eyes of patients with VRL included an aurora borealis pattern, which was represented by cell conglomerates aligned along the vitreous fibrils, and a string of pearls pattern. Furthermore, medical scientists from Northwestern University, Chicago, have reported hyperreflective lesions in all retinal layers in VRL patients.9 The observed lesions would usually appear prior to pigment deposits and surrounded retinal vessels and vanished mostly without scarring after the start of treatment.
In addition to diagnostic applications, the authors of the current study followed the healing patterns in both VRL lesions and infectious retinitis using CT analysis. According to Dr. Marchese, the most common treatments for VRL are intravitreal rituximab and methotrexate. Other treatments include radiotherapy, intravitreal melphalan, systemic chemotherapy with high-dose methotrexate or cytarabine. Additional or alternative treatments are performed in the case of the involvement of central nervous system. Intravitreal chemotherapy led to the decrease of sub-RPE lesions and partial regeneration of the retina. In contrast to VRL, patients with viral and toxoplasmic retinitis revealed retinal destruction and necrotic lesions during the healing process.
In conclusion, the authors suggest that attentive analysis of the OCT scans would provide clear benefits for patients, especially in cases of suspected VRL versus infectious retinitis. This diagnostic method helps to distinguish VRL from viral or toxoplasmic retinitis. However, clinicians should beware of unspecific OCT features detectable in both VRL and infected retina. Thus, early diagnosis of VRL and appropriate treatment strategy would help to prevent retinal deterioration.
References:
1 Aronow ME, Singh AD. The use of imaging in the diagnosis and management of intraocular lymphoma. Int Ophthalmol Clin. 2012 Fall;52(4):199-208.
2 Marchese A, Agarwal A, Miserocchi E, et al. Features
of Retinitis-like Lesions in Vitreoretinal Lymphoma. Ocul Immunol Inflamm. 2019;30:1-8. [Epub ahead of print]
3 Chan CC, Rubenstein JL, Coupland SE, et al. Primary vitreoretinal lymphoma: a report from an International Primary Central Nervous System Lymphoma Collaborative Group symposium. Oncologist. 2011;16(11):1589-1599.
4 Albadri ST, Pulido JS, Macon WR, et al. HISTOLOGIC FINDINGS IN VITREORETINAL LYMPHOMA: Learning From Enucleation Specimens. Retina. 2019: [Epub ahead of print]
5 Reddy V, Winslow R, Cao JH, et al. Vitreoretinal lymphoma, secondary to non-CNS systemic lymphoma, masquerading as an infectious retinitis. Am J Ophthalmol Case Rep. 2019;16:100545.
6 Gonzales J, Doan T, Shantha JG, et al. Metagenomic deep sequencing of aqueous fluid detects intraocular lymphomas. Br J Ophthalmol. 2018;102(1):6-8.
7 Barry RJ, Tasiopoulou A, Murray PI, et al. Characteristic optical coherence tomography findings in patients with primary vitreoretinal lymphoma: a novel aid to early diagnosis. Br J Ophthalmol. 2018;102(10):1362-1366.
8 Marchese A, Miserocchi E, Giuffre C, et al. Aurora borealis and string of pearls in vitreoretinal lymphoma: patterns of vitreous haze. Br J Ophthalmol. 2019 Nov;103(11):1656-1659.9 Deak GG, Goldstein DA, Zhou M, et al. Vertical Hyperreflective Lesions on Optical Coherence Tomography in Vitreoretinal Lymphoma. JAMA Ophthalmol. 2019;137(2):194-198.
