High-grade Uveal B-Cell Lymphoma as the Initial Feature in Richter Syndrome

We describe a patient with a 6-year history of chronic lymphocytic leukemia who developed iris, ciliary body, and choroidal tumors associated with ocular pain and uveitis that were unresponsive to topical anti-inflammatory steroids. Because the eye was not salvageable with conservative treatment, enucleation was performed. Findings from histopathologic examination of the enucleated eye showed a high-grade B-cell lymphoma. This case represents an example of Richter syndrome (high-grade lymphoma arising in patients with chronic lymphocytic leukemia) clinically mimicking uveal melanoma.

An 81-year-old woman complained of 4 weeks of blurred vision and 10 days of pain in the right eye. The eye had been treated with topical corticosteroids and cycloplegics, prescribed by another ophthalmologist, for the 2 weeks preceding our evaluation. Her vision was 6/200 OD and 20/25 OS. Intraocular pressure was 18 mm Hg OU. There was no injection or chemosis of the conjunctiva. The right eye had keratic precipitates and a hypopyon. The iris touched the cornea temporally and was infiltrated by tumor from the 7- to 11-o'clock positions. The vitreous had 2+ cells and flare. Funduscopy showed choroidal tumors through hazy media. A total serous retinal detachment was present. Ultrasound performed a few days later showed a diffuse, irregular mass of the choroid and ciliary body with a maximum height of 9.5 mm (Figure 1, A).

The patient had a 6-year history of chronic lymphocytic leukemia, RAI stage 0 (lymphocytosis only). At an examination by her oncologist 6 months prior to onset of ocular symptoms, she was without symptoms of the disease and taking no systemic medications. Her white blood cell count at that time was 17 000/mm³ (normal, 4800-10 800/mm³) and her hematocrit was 36.9% (normal, 37%-40%). Three months before the onset of ocular symptoms, her white blood cell count had decreased to 9900/mm³, and her hematocrit had decreased to 33.7%.

Because of the hypopyon and keratic precipitates, which are rarely found in patients with uveal melanoma, the presumptive diagnosis was leukemic infiltrate of the uveal tract. However, uveal melanoma could not be ruled out. In view of the pain and no potential for vision, and because the uveal tumors were too large for radiotherapy, the right eye was enucleated 7 weeks after the onset of symptoms (Figure 1, B).

Findings from histologic evaluation showed that the uveal tumor (Figure 1, C and D) was composed of medium to large cells with scant cytoplasm and vesicular nuclei with areas of clumped chromatin and prominent nucleoli. There were more than 2 mitoses per high-power field (Figure 1, D). Scattered among these tumor cells were tingible body macrophages that produced the starry-sky pattern seen in some high-grade lymphomas, especially those of Burkitt lymphoma (Figure 1, E).

Immunoperoxidase studies showed that the tumor cells expressed bcl2 (an oncoprotein that inhibits apoptosis) and the B-lymphocyte antigen CD20. They lacked the lymphocyte subset–associated markers CD10, CD5, and TdT, as well as the melanoma markers HMB45 and MART1. Depending on the region of the tumor evaluated, 60% to 90% of the tumor cells were positive for Ki67, indicating a high fraction of proliferating cells. A few admixed nonneoplastic, small T cells (expressing the pan-T-cell antigens CD3 and CD5) were present. The diagnosis was high-grade B-cell lymphoma. The immunophenotype excluded Burkitt lymphoma (which typically expresses CD20, CD10, and, in 100% of cells, Ki67, and which lacks bcl2) and precursor B lymphoblastic lymphoma (which typically expresses CD10 and TdT but not CD20).

Six weeks after the enucleation the patient felt weak and had palpable axillary and inguinal lymph nodes and splenomegaly. The hematocrit fell to 20% and the white blood cell count rose to 146 000/mm³. Review of a peripheral blood smear revealed 27% neutrophils, 1% bands, 27% lymphocytes, 5% monocytes, 38% blasts, and 2% nucleated red blood cells. The cells counted as blasts were medium to large cells with deep-blue agranular cytoplasm and ovoid nuclei with chromatin that was usually irregularly clumped. Most lacked discrete nucleoli, but occasional cells had prominent nucleoli (Figure 1, F). The appearance was that of abnormal lymphoid cells similar to those seen in the eye, but not typical of prolymphocytes. Only palliative treatment was administered, and the patient died 4 months after the enucleation.

About 3% of patients with chronic lymphocytic leukemia develop non-Hodgkin lymphoma (usually large B-cell lymphoma); this sequence of malignancies is called Richter syndrome.^1-3 The onset of the lymphoma is usually abrupt. Clinical features include fever, weight loss, increasing lymphadenopathy, lymphocytopenia, and dysglobulinemia.³ Based on studies of immunoglobulin isotypes and gene rearrangements, the lymphoma in most cases of Richter syndrome appears to be the progression of chronic lymphocytic leukemia into a more aggressive tumor; however, in a minority of cases, the lymphoma likely represents a second independently arising tumor.⁴

We know of only 2 previously reported cases of Richter syndrome that involved the eye. In one case the malignant lymphocytes were confined to the vitreous⁵; in the other, the tumor was in the vitreous and the subretinal space.⁶ This pattern of involvement is characteristic of retinal/central nervous system lymphoma, a form of lymphoma that is not usually associated with disease outside the central nervous system. Neither case had clinically apparent uveal involvement, and neither case had the rapid downhill course typical after the development of Richter syndrome. In contrast, the lymphoma in our case predominantly involved the uveal tract, a feature that correlates with systemic lymphomas.⁷ The clinical findings mimicked uveal melanoma rather than retinal/central nervous system lymphoma. The patient had a rapid downhill course and died soon after the diagnosis was made. Our case represents what we consider to be a novel presentation of Richter syndrome and demonstrates that Richter syndrome should be included in the differential diagnosis of uveal melanoma.

None of the authors has proprietary or commercial interests related to the report.

Corresponding author: Thaddeus P. Dryja, MD, Massachusetts Eye and Ear Infirmary, 243 Charles St, Boston, MA 02114 (e-mail: dryja@helix.mgh.harvard.edu).