Further to my post above, following are additional letters to the editor in the same issue of NEJM dealing with Tysabri and PML. No resolutions to be found here – just an interesting exchange:
To the Editor: Kleinschmidt-DeMasters and Tyler described a patient with multiple sclerosis and PML that developed after she received natalizumab treatment. However, some of the findings in this patient are uncommon in multiple sclerosis. The absence of contrast enhancement in the MRI studies that were performed — almost a rule in patients with PML — is uncommon in multiple sclerosis.1 The lack of intrathecal synthesis of IgG is unusual in multiple sclerosis, although it is frequent in PML.2 Moreover, an autopsy study found no multiple sclerosis plaques but only PML lesions. Therefore, the pathological findings did not support a diagnosis of multiple sclerosis, which raises some questions. Can initial PML mimic multiple sclerosis? Was PML the only disease in this patient? To our knowledge, PML with a relapsing–remitting course has not been described. To clarify whether PML may have a relapsing–remitting course, it is necessary to study the cerebrospinal fluid in patients with relapsing neurologic disorders to rule out this disease. This seems particularly important for patients with suspected multiple sclerosis with atypical findings, such as early mental symptoms or an absence of oligoclonal bands. If it is demonstrated that PML can have an initial relapsing–remitting course, the risks of natalizumab treatment3 should be reevaluated.
José C. Álvarez-Cermeño, M.D.
Jaime Masjuan, M.D.
Luisa M. Villar, Ph.D.
Hospital Ramón y Cajal
28034 Madrid, Spain
"Drs. Kleinschmidt-Demasters and Tyler reply: We appreciate the comments of Drs. Berger and Deisenhammer and Dr. Álvarez-Cermeño and colleagues. Our involvement with this case began only post mortem; we were not involved in this patient's initial clinical evaluation, diagnosis, and treatment. The clinical and laboratory data presented were supplied by the patient's treating physician and supplemented by the Biogen company representative. Unfortunately, in some instances only the original MRI reports, not the actual scans, were available for independent review.
A variety of diagnostic criteria for multiple sclerosis have been proposed, including the widely accepted criteria of McDonald et al.1 These criteria do allow for the diagnosis of multiple sclerosis to be made solely on clinical grounds in a patient with two or more attacks and objective clinical evidence of two or more lesions. Criteria specific to MRI can provide supportive evidence of the dissemination of lesions over time; this criterion can be satisfied by the appearance of new T2-weighted lesions on appropriately spaced sequential scans. Oligoclonal bands on evaluation of the cerebrospinal fluid and enhancing lesions as shown on MRI are found in the majority of patients with definite multiple sclerosis; however, their presence is not absolutely required for diagnosis, nor does their absence completely rule out a diagnosis of multiple sclerosis.
We cannot resolve whether the absence of multiple sclerosis lesions at autopsy reflects obliteration of a limited number of multiple sclerosis plaques by the widespread and extremely destructive PML lesions, sampling error, or the presence of an alternative disease process. With regard to this last possibility, we believe it is unlikely that a chronic indolent or relapsing–remitting form of PML was the sole disease that this patient had or that it was responsible for this patient's symptoms early in her clinical course.
Our report, in combination with the others published in the same issue of the Journal,2,3 clearly indicates that natalizumab is associated with an increased risk of PML in patients without other recognized risk factors for this disease. The common denominator in all three reported cases, regardless of the underlying disease, was treatment with natalizumab.2,3
We hope that ongoing studies will clarify the pathogenesis of this disorder, including the viral factors and specific host factors responsible, and also the mechanisms by which natalizumab facilitates induction of PML.
B.K. Kleinschmidt-DeMasters, M.D.
Kenneth L. Tyler, M.D.
University of Colorado Health Sciences Center
Denver, CO 80262 "
"Dr. Langer-Gould and colleagues reply: The clinical history of our patient does not support the notion that subclinical cases of PML are masquerading as multiple sclerosis. To clarify, our patient had typical relapsing–remitting multiple sclerosis with clinical episodes of neurologic dysfunction for more than 20 years. He repeatedly had enhancing lesions in his brain and spinal cord during the 10 years before he was randomly assigned to the natalizumab trial, as well as a typical periventricular enhancing lesion concomitant with the first PML lesion. Finally, PML rarely affects the spinal cord.
We agree that the case in Colorado that was reported by Kleinschmidt-DeMasters and Tyler is troubling. The clinical criteria for multiple sclerosis were recently revised to improve their sensitivity, with unknown effects on specificity. The revised criteria rely heavily on the presence or absence of lesions on T2-weighted imaging, which frequently occur in patients with migraine,1 such as this patient. The natalizumab trial included patients with normal neurologic examinations (such as this patient), despite the fact that the risk of permanent disability in such patients is very low.2 These problems with the diagnostic criteria for multiple sclerosis and the entry criteria for clinical trials need to be addressed.
Annette Langer-Gould, M.D.
Scott W. Atlas, M.D.
Stanford, CA 94305-5405
Daniel Pelletier, M.D.
University of California, San Francisco
San Francisco, CA 94143 "