Background/Purpose:  The occurrence of lymphoma, especially Hodgkin’s lymphoma (HL), in patients with primary central nervous system vasculitis (PCNSV) has been reported but is considered uncommon. The aim of this study was to determine the frequency of lymphoma in a large cohort of patients with PCNSV and compare the presenting clinical, laboratory, and imaging features in those with both lymphoma and PCNSV to those without lymphoma.

Methods:  We reviewed all patients seen at the Mayo Clinic, Rochester, MN over the 32- year period of 1982-2014, who were diagnosed with vasculitis and lymphoma. 10 patients associated PCNSV and lymphoma Cerebral biopsy specimens were reviewed by one pathologist (CG) without knowledge of clinical information. We also used as comparator our updated cohort of 158 consecutive patients with PCNSV without lymphoma who were examined at Mayo Clinic over a 29-year period from 1983 to 2011. The diagnosis of PCNSV was based on brain/spinal cord biopsy, or cerebral angiography, or both. Clinical data were collected.

Results:  10/168 (5.9%) patients were found to have both lymphoma and PCNSV: 6 of these 10 had HL and 4 non-HL. In 8 patients PCNSV diagnosis was established by cerebral biopsy and in 2 by cerebral angiography. Two of the 10 patients had brain and spinal cord vasculitis involvement. A granulomatous inflammatory histologic pattern was found in all 8 patients with cerebral biopsies, accompanied by vascular deposits of β-amyloid peptide in 2. In 7 patients (5 HL and 2 NHL) medical diagnostic workup for PCNSV revealed lymphoma. In one other patient PCNSV symptoms appeared during a recurrence of lymphoma, in another 6 months after an allogenic bone marrow transplant for a recurrence of lymphoma, and in one lymphoma occurred 26 years before PCNSV diagnosis. Cerebrospinal fluid (CSF) was negative for herpes virus when assayed by PCR. The 10 patients with lymphoma were compared with the 158 patients with PCNSV without lymphoma. The patients with lymphoma were more frequently males (80% vs 44%, p = 0.04). No other significant differences in the clinical manifestations at presentation and CSF findings were observed in the two groups, although cognitive dysfunction was more frequent in patients with lymphoma (80% vs 53%), while visual field defects and intracranial hemorrhage were less frequent (0% vs 19%, and 0% vs 10%, respectively). Systemic manifestations were infrequent in both groups (10% vs 9.5%). More patients with lymphoma showed meningeal gadolinium enhancing lesions on MRI (50% vs 19%, p = 0.03). The frequency of PCNSV relapse (20% vs 28.4%) and patients not requiring therapy at last follow-up (30% vs 25.5%) were similar in both groups. Patients with lymphoma had a higher frequency of poor outcomes (modified Rankin disability score > 4) at last followup ( 60% vs 21,5%, p = 0.01). Considering all 168 patients, univariate Cox proportional hazards modeling showed an increased mortality rate in those with increasing age (hazard ratio, HR, 1.3), lymphoma (HR 3.9), cerebral infarction on initial MRI (HR 3.4), and angiographic large vessel involvement (HR 3.9), while mortality rate was lower in those with gadolinium–enhancing lesions on MRI (HR 0.3).

Conclusion: Lymphoma was found in 5.9% of of 168 patients with PCNSV. Lymphoma may occur simultaneously with PCNSV, before or after the diagnosis of PCNSV. Most clinical characteristics of PCNSV were similar in those with or without lymphoma, however, patients with lymphoma had a more severe cerebral vasculitis with increased neurological disability and mortality.

« Back to 2016 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/primary-central-nervous-system-vasculitis-associated-with-lymphomas/