Discussion:
To the best of our knowledge, this is the first reported case of a CNS involvement similar to BNS associated with a generalized CSH in the setting of a MZL.
Here we present a generalized form of CSH which represents 42% of clinical presentation of this disease, whereas the localized pattern is more frequent and often localized in the head and neck region [2]. CSH do not usually impair CNS: Dogan S. et al. report only 3% of dura and pia mater involvement in patients with generalized CSH, and Flanagan ME et al. report one case of CSH involving cerebellum and caudal brain stem [2, 7].
In our case, neurological manifestations and MRI findings were highly suggestive of BNS with signs of parenchymal and leptomeningeal involvement. The lumbar puncture cellularity was consistent with a type B BNS according to the Fintelmann’s proposed classification [8]. Type B BNS represents 25% of total BNS and is defined by a very low cells rate (<5 cells/mm3) in the cerebrospinal fluid. This suggested neurological manifestations being associated with IgM deposition with an intrathecal secretion, as the pathological CSF IgM index indicates, rather than lymphoplasmacytic infiltration [8].
There is no clear causal link between CSH and BNS. We consider that they are just two manifestations of an abnormal circulation of immunoglobulins in quantity and quality.
In our case, definitive classification of the underlying small B-cell lymphoma, which shows a striking plasmacytic differentiation, remains difficult. Lymphoplasmacytic lymphoma (LPL) usually involves bone marrow and sometimes lymph nodes and spleen. WM is defined as a LPL with bone marrow involvement and an IgM monoclonal gammopathy. MYD88 L265P mutation has been described in almost all, if not all (this is a point of debate) LPL/WM, associated with a CXCR4 mutation in approximately 30% of cases. Conversely, this MYD88 L265P mutation is rarely (~5% of cases) present in MZL. So, in our case, polyadenopathy with minimal bone marrow involvement and absence of MYD88 L265P mutation would favor MZL with plasmacytic differentiation rather than LPL/WM. However, because this neurological presentation similar to BNS, which is, to our knowledge, exclusively described as a complication of WM in literature, our patient was considered as having WM complicated by a BNS for therapeutic decision. Nevertheless, one can legitimately wonder why BNS could not also complicate a MZL with plasmacytic differentiation and IgM paraprotein, knowing that neurological manifestations in BNS type B are probably related to IgM rather than to lymphoma cells infiltration.
The patient was treated as a BNS with R-Bendamustine associated and intrathecal treatment allowing a clinical and radiological improvement. Ibrutinib appears to be a good alternative as an oral treatment for this disease [9]. CSH treatment depends of the causal pathology. There is little information concerning specific response of CSH after chimiotherapy but the histologic lesions persist in available observations [10].