DIAGNOSIS CNS spread of untreated acute myeloid leukemia associated with acute demyelination
DISCUSSION
Patients suffering from leukemias may develop CNS complications of their disease. Spread of the disease into the central nervous system (CNS) is a rare example, which is generally less frequent in adults than in children[2, 8], and particularly uncommon in untreated patients, where only single cases are reported[5]. Leukoencephalopathy is more common and potentially life-threatening. It may have several different etiologies, for example demyelination as a paraneoplastic disorder[1], progressive multifocal leukoencephalopathy after a JC-virus infection[3] or multifocal necrotizing leukoencephalopathy after chemotherapy[6]. However, as these examples illustrate, reports on leukoencephalopathy in relation to leukemia may be ambiguous because the patients typically receive treatment with potential neurotoxicity.
Recently, a CNS relapse of an AML associated with leukoencephalopathy was suggested in an adult patient that had only been treated with cytarabine, which is not thought to penetrate the CNS, making a neurotoxic side-effect unlikely[7]. MRI in that patient showed widespread leukoencephalopathy, but there was only indirect evidence for a role of a parenchymal CNS spread of the AML. Myeloblasts were found in the CSF and the leukoencephalopathy improved after chemotherapy, but no brain tissue was examined. That patient could be discharged without any neurological symptoms.
Our patient had very similar pathological features on MRI, but she had not been treated with any chemotherapeutic agents at all. In her case, though, the CNS-spread of the AML and the histopathologic pattern of the severe and extensive leukoencephalopathy could be shown in post mortem analysis. Our patient was seriously affected by a CNS-spread of the AML which may have been present for years, as suggested from her past medical history, and finally resulted in the patient's death from a blast crisis. Autopsy revealed active demyelination with complement deposits that extended over the cerebrum and myelon (spinal cord). While pronounced axonal destruction and complement-mediated demyelination can be typical for neuromyelitis optica (NMO), the patient had no clinical signs of NMO and we failed to detect any aquaporin-4 antibodies, arguing against this possibility.
In summary, we present a patient with CNS spread of an untreated acute myeloid leukemia, which is a very uncommon finding. Moreover, it was associated with acute demyelination and axonal destruction. We suggest that such demyelination may occur after and be related to the CNS spread of myeloblastic leukemic cells. Whether immunosuppression itself may trigger acute demyelination or exacerbation of a chronic multiple sclerosis or whether further mechanisms of immune system modulation related to AML are involved in the reported scenario need to be investigated in further studies and experimental settings. By all means, we believe that our case strengthens the need to strictly monitor CNS functions of AML patients.
REFERENCES
Contributed by Jan Rémi, Thomas Pfefferkorn, Fatima B. König, Hans Lassmann, Wolfgang Brück, Markus Holtmannspötter, Andreas Straube, Hans A. Kretzschmar and Ulrich Schüller