Nervous System Invasion Occurs in 4% Percent of Chronic Lymphocytic Leukemia

Introduction: Chronic Lymphocytic Leukemia (CLL) is a low-grade B-cell lymphoma that manifests as high white blood cell counts and lymphadenopathy. Current clinical paradigms for the care of CLL do not include evaluation or prophylaxis of the central nervous system. Four percent of our patients with CLL developed neurologic manifestations which approximates the 3–7% risk of CNS involvement in large series of diffuse large B cell lymphoma (DLBCL). We found that the majority of these cases had no evidence of histologic Richter’s transformation. Our experience confirms prior anecdotal reports of involvement of the central nervous system by CLL which have described: abnormal lymphocytes in the CSF or vitreous, MRI dural enhancement, enhancement within the spinal cord or cranial/peripheral nerves, orbital mass, PML-like diffuse white matter abnormalities of brain (without the presence of JC virus DNA) and/or Richter’s transformation in the nervous system.

Methods: A CLL medical records review protocol was approved by the Dana Farber Harvard Cancer Center Institutional Review Board. Using our clinical lymphoma database we collected records of 755 patients at Massachusetts General Hospital (MGH) Cancer Center who carried a diagnosis of CLL from January 1988 to August 2008. We will present the neurologic involvement identified by restricting cases to those with the following diagnostic studies: patients who had received brain, head or spine MRIs, patients who had received lumbar punctures or post-mortem examinations. We also included any patient with brain, nerve (peripheral or otherwise) and/or orbital biopsies. Within this cohort, 30 patients were found with the following sites of predominant neurological involvement of CLL: dural infiltration (2), CSF infiltration by cytology or flow cytometry (6), infiltration of peripheral nerve, optic nerve or spinal nerve roots (6), ocular involvement of the vitreous or conus (7), Richter’s transformation into primary CNS DLBCL (6), and diffuse non-enhancing white matter abnormalities without the presence of JC virus (3). Many patients presented with multiple neurological sites of disease.

Discussion: Recognition of the risk of central nervous involvement in DLBCL has led to a new paradigm of CNS prophylaxis in high risk patients. Our estimate of 3.97% involvement of the nervous system with CLL likely underestimates the true risk of this fatal complication for the following reasons. We did not directly examine patient records for neurologic symptoms. We did not re-evaluate the cytology of CLL patients whose neurologic symptoms prompted a negative lumbar puncture nor neuro-pathology specimens of autopsied CLL patients with negative gross involvement. Patients who were diagnosed with CLL at MGH but who had their neurologic studies performed outside of our electronic medical record (EMR) system are clearly not included in our numerator. Current CLL therapies have limited CNS penetration. Our risk estimate, if confirmed, will stimulate the inclusion of neur-axis penetrating agents into CLL therapy in high-risk patients.