Isolated T-cell acute lymphoblastic leukemic optic disc infiltration.

Abstract

A 56-year-old female presented for acute onset headache, cloudy vision, peripheral vision loss, and floaters in the left eye. Eight months prior, she was diagnosed with T-cell acute lymphoblastic leukemia (ALL) (balanced 6;20 and 7;16 translocations and fluorescence in situ hybridization (FISH) positive for heterozygous CDK2NA deletion) and treated with hyper-CVAD (cyclophosphamide, vincristine, doxorubicin, and dexamethasone) alternating with high-dose cytarabine and methotrexate and a matched, unrelated donor allogeneic peripheral blood stem cell transplant. Visual acuity was 20/20 right eye and 20/40 left eye with 13/13 Ishihara pseudoisochromatic color plates right eye and 5/13 left eye. There was no relative afferent pupillary defect in either eye. Funduscopy showed grade 5 optic disc edema, peripapillary hemorrhages, surrounding retinal pigmentary changes, and posterior vitreous cells in the left eye (Figure 1A). Examination of the right eye was normal. Fundus autofluorescence of the left eye showed a stippled pattern of hyperautofluorescence and hypoautofluorescence consistent with leopard spotting throughout the peripapillary retina and macula (Figure 1B). Optical coherence tomography (OCT) of the left eye demonstrated optic disc edema with epipapillary hyperreflective material and vitreous cell (Figure 1C). Two months prior, her bone marrow biopsy was negative for measurable residual T-ALL. In addition, recent brain magnetic resonance imaging without and with contrast and a diagnostic lumbar puncture (LP) demonstrated no evidence of leukemic recurrence or optic nerve enhancement. A broad infectious, inflammatory, and paraneoplastic evaluation, including serology for Treponema pallidum, Borrelia burgdorferi, rheumatoid factor, antinuclear antigen, anti-cyclic citrullinated peptide, double-stranded DNA, anti-Smith, proteinase-3, myeloperoxidase, and anti-collapsin response mediating protein-5, was negative. Given that her presentation was strongly suggestive of leukemic infiltration, we recommended diagnostic pars plana vitrectomy to establish a diagnosis prior to treatment initiation. The patient elected to proceed after a thorough discussion of risks, including vision loss related to optic atrophy due to iatrogenic injury, acceleration of cataract development, and endophthalmitis, among others. Pars plana vitrectomy with biopsy of the epipapillary material was performed. The dense cellular material overlying the optic nerve was slowly aspirated with a vitreous cutter and expeditiously sent to the pathology laboratory. Histopathology revealed a hypercellular specimen composed of CD3 and TdT-positive cells consistent with optic disc infiltration from the patient's known T-cell ALL (Figures 1D,F). The patient was promptly treated with craniospinal proton irradiation therapy (2400 centigray in 12 fractions) followed by systemic cytarabine and methotrexate to prevent additional central nervous system (CNS) involvement. One month after radiotherapy, visual acuity improved to 20/25 with grade 1 disc edema (Figure 1G). Fundus autofluorescence demonstrated a leopard spotting pattern (Figure 1H). Optic nerve OCT showed excrescences of the retinal pigment epithelium with disruption of the ellipsoid zone and resolution of epipapillary hyperreflective material (Figure 1I). Physicians should consider a broad differential for any patient with ALL presenting with optic disc edema, including infectious, inflammatory, paraneoplastic, and iatrogenic etiologies, given possible immunocompromised state, history of chemotherapy, and radiation therapy. However, leukemic infiltration should be the working diagnosis until proven otherwise. Isolated leukemic optic nerve infiltration has rarely been reported as the initial presentation of disease recurrence.1, 2 Because the optic nerve is a pharmacologic sanctuary site, systemic chemotherapy penetration may be inadequate, making the optic nerve susceptible to recurrence.3 In this case, optic disc infiltration was evident only on ophthalmic examination with no evidence of CNS disease or optic nerve enhancement on MRI. Prior studies have demonstrated MRI sensitivity of 20% for leukemia detection, and bone marrow can remain in morphological remission for months to years after the development of extramedullary relapse,3 demonstrating the importance of a dilated fundus examination.4 Our patient also underwent LP with cytology, which was negative for CNS involvement, defined as ≥ 5 leukemic cells per microliter and/or leukemic blasts in cerebrospinal fluid (CSF). Prior reports estimate < 50% sensitivity of CSF cytology for leukemia.5 Prompt treatment of leukemic optic nerve infiltration can restore visual acuity and prevent the development of additional CNS disease. The authors declare no conflict of interest. Data sharing not applicable—no new data generated.