Discussion
Waldenstrom macroglobulinemia (WM) is defined by the World Health Organization (WHO) as an indolent lymphoplasmacytic lymphoma (LPL) belonging to the category of non-Hodgkin B-cell lymphomas (NHL) [4]. Although there remains a lack of specific chromosomal or oncogene abnormalities in LPL, 90% of cases share MYD88 L265P mutations that are most commonly associated with IgM monoclonal gammopathies such as WM [4]. While WM remains incurable, median survival rates have increased from 5 to 8 years due to increased awareness of the disease and advancements in therapy over the last decade [2]. Essential aspects of the initial evaluation for WM include a detailed history and physical examination prior to pursuing diagnostic workup. Although constitutional symptoms of this disease are shared amongst various other lymphoid malignancies, thorough history taking may reveal clinical manifestations specific to hyperviscosity syndrome such as spontaneous epistaxis, recurrent headaches, visual changes, and blurred vision [1,2]. Peripherally circulating antibodies not only cause detrimental effects through changes in blood viscosity, but also by their deposition in end-organs and through immune system autoreactivity [5]. However, the multiple complications associated with WM are not simply related to the effects of a monoclonal gammopathy. For example, peripheral neuropathy which is observed in approximately 20% of patients who suffer from this disease process may be a result of direct lymphoplasmacytic infiltration or IgM deposition of nerve fibers, amyloidosis from excess light chain production, or development of autoimmunity [1,2].
In approximately 1% of WM cases, patients develop Bing-Neel syndrome (BNS) in which there is infiltration of the CNS by lymphoplasmacytic cells. The absolute incidence of BNS is unknown, but in a retrospective cohort study of 1,523 WM patients, only 13 patients (0.8%) were diagnosed with BNS, suggesting a very low prevalence [3,6]. A review of published literature shows primarily case reports and few small retrospective surveys, demonstrating the rarity of this disease manifestation. Retrospective analysis has demonstrated a significantly lower median survival time (4 months) between symptom onset and diagnosis of BNS as compared to WM [3]. To further complicate matters, BNS can present at any time during the active treatment course for WM, even when the patient is in apparent remission from disease [1-3]. Rarely, BNS precedes the diagnosis of WM and appears as a primary CNS LPL, having only been described in twenty-four cases [7]. Prognosis appears to be better in these cases as compared to those with a prior history of WM whose disease progresses to BNS [3].
The evaluation for BNS begins with diagnostic testing including gadolinium-enhanced magnetic resonance imaging (MRI) of the brain and whole spine as well as cerebrospinal fluid (CSF) sampling for cytology, flow cytometry, and mutational analyses [1-3]. Findings on MRI include contrast-enhanced infiltrations with or without thickening of the meningeal sheaths depending on leptomeningeal involvement, in addition to accentuated diffusion weight imaging (DWI) with elevated or normal apparent diffusion coefficient (ADC) values suggestive of vasogenic edema due to perivascular space invasion by malignant cells [9]. Of note, there are two categories of BNS that can be distinguished on MRI: a diffuse form, in which there is leptomeningeal and perivascular infiltration versus a tumoral form which is unifocal or multifocal and usually involves the deep subcortical hemispheric regions [3, 9]. Infrequently, infraorbital or periorbital involvement can also be seen [20]. In regards to our patient, she presented as the tumoral form of BNS with evidence of multiple enhancing lesions within the subcortical white matter. Although these different forms of BNS from MRI findings are described extensively in literature, actual invasion of malignant lymphoplasmacytic cells may be more extensive than what is evident on imaging. Autopsy reports are scarce given the rarity of this disease, but there appears to be prominent perivascular infiltration by malignant cells without deposition of IgM in BNS that is not readily apparent on MRI imaging [10].
Despite these sensitive imaging techniques, MRI cannot distinguish between other forms of CNS lymphoma thus necessitating further testing with both CSF analysis and tissue biopsy. CSF should be collected and sent for cytology, flow cytometry, in addition to electrophoresis and immunofixation in order to determine whether M-protein and specific immunoglobulin elevations are present within the CNS [1-3]. Yet, results from immunofixation may be skewed by other form of lymphoplasmacytic lymphomas. Thus, mutational analysis of CSF plays a prominent role in the workup of BNS. As is seen in approximately 90-97% of WM cases, MYD88 L265P gene mutation identified by either next generation sequencing (NGS) or real-time quantitative PCR (qPCR) aids in diagnosis of BNS and the absence of such may be associated with poorer prognosis [3,8,11,12]. As demonstrated by Hiemcke-Jiwa et al, when clinical suspicion remains high for BNS and there is concern for low tumor DNA concentration within the CSF, highly sensitive double droplet PCR (ddPCR) techniques may be used as an alternative for identifying MYD88 L265P mutations [11]. In addition, testing for immunoglobulin gene rearrangements can help establish whether or not there are monoclonal heavy and light chain gene rearrangements in lymphoplasmacytic cells identified from the CSF, further augmenting the diagnostic workup of BNS [3]. Of note, the presence of these genetic mutations in either biopsy or CSF analysis is not necessarily specific for BNS as it can be found in other forms of primary CNS lymphoma [3,8,11]. Regardless of CSF results, biopsy of cerebral or meningeal lesions remains the gold stand for diagnosis of BNS with pathological evidence of lymphoplasmacytic lymphoma and immunochemical demonstration of monotypic B cells expressing antigens such as CD19, CD20, CD79a, and CD79b [3].
Given the paucity of BNS cases, there remains a lack of universal guidelines pertaining to the management of this disease process. Although there is no standardized approach for treatment, therapy should begin with identifying individuals who are symptomatic from their disease and requires antineoplastic agents that have the ability to penetrate the blood-brain barrier or agents available for intrathecal administration [3,13,21]. Treating those who are asymptomatic is felt to be contraindicated since the aim of treatment is not curative and tailored more towards reversing symptoms and inducing long term progression free survival [3,13]. Previously, systemic chemotherapy with purine nucleoside analogs such as cytarabine, fludarabine, and bendamustine, known to be effective at treating other lymphoproliferative disorders such as WM, showed promise as therapy for the treatment of BNS [14-16]. High-dose methotrexate and rituximab have been effective for the treatment of BNS as well, but only in combination with one another, as monotherapy with either of these agents has proven to be ineffective [13,14]. Rituximab itself is believed to have poor penetrance of the blood-brain barrier, thus contributing to its modest effect. In recent years, the oral Bruton tyrosine kinase (BTK) inhibitor ibrutinib has demonstrated promising results for the treatment of BNS, with it already having proven to be effective as therapy for WM [17-19]. The CNS penetrance of ibrutinib has been well-established in mouse models in addition to individual cases assessing the concentrations of active metabolite, PCI-45227, between synchronous measurements of plasma and CSF [18]. In a multicenter study conducted by Castillo et al. involving 28 patients with BNS treated with ibrutinib, it was found that 86% of patients had improvement of their associated symptoms with 83% of cases demonstrating tumor response on brain MRI [17].
Additional potentially targetable mutations have been identified throughout the past decade and thus offer further therapeutic alternatives for the treatment of WM and BNS. WHIM (warts, hypogammaglobulinemia, infections, and myelokathexis) syndrome caused by heterozygous mutations within the CXCR4 gene is observed in the pediatric population and characterized by chronic noncyclic neutropenia [24]. Although MYD88 L265P mutation is the most common somatic mutation in WM, this is followed closely by CXCR4 WHIM-like frameshift and nonsense mutations [22,23]. Plerixafor is an FDA-approved CXCR4 partial agonist and allosteric antagonist of CXCR7 which has been studied for clinical efficacy and safety in treatment of WHIM syndrome [26,27]. In vitro studies have demonstrated that ibrutinib resistance can be potentially reversed by CXCR4 inhibition, yet MYD88 inhibition superseded the survival benefits provided by CXCR4 frameshift mutations [25]. Therefore, it is theorized that plerixafor may have potential use in reversing ibrutinib failure in patients with WM and its efficacy is currently being evaluated within clinical trials. CXCL12 is another molecule that has been identified as an activator of AKT 1 and MPK1 pathways which helps confer the ability of malignant WM cells to resist ibrutinib-triggered apoptosis through mechanisms that increase frameshift and nonsense mutations [27]. Through continued research of molecular alterations leading to the development of WM and BNS, as well as genetic pathways contributing to resistance of targeted therapy, we can expect to see further improvement in both progression free and overall survival in this patient population.
As illustrated in our patient’s case, switching from ibrutinib to rituximab had short-lived efficacy and soon led to worsening serum viscosity and IgM gammopathy. It is possible that she had already developed BNS at the time of disease progression which resulted in an inadequate response to rituximab monotherapy because of its poor blood-brain barrier permeability. Resuming therapy with ibrutinib resulted in significant improvements of her lab values and symptoms. However, despite continued treatment and relatively stable disease the patient developed BNS, raising concern given ibrutinib’s proven efficacy for the treatment of BNS. Once systemic therapy was combined with WBRT there was noted improvement in both the patient’s neurological symptoms as well as extent of disease evident on repeat imaging. Our case illustrates the importance of recognizing the early signs and symptoms of BNS within those who have a history of WM, regardless of their current and past treatment regimens. Diagnostic evaluation should be thorough and include biopsy of intracranial lesions when the suspicion remains high for BNS, as this remains the gold standard of practice. Early recognition of BNS can ultimately lead to more prompt treatment and rapid improvement of symptom burden, extended progression free survival, and overall improved quality of life in this scarce patient population.