Focal Symptomatic Bone Involvement with Waldenström’s Macroglobulinaemia and Non-IgM Lymphoplasmacytic Lymphoma

Background
Waldenstrom’s macroglobulinaemia (WM) and non-IgM lymphoplasmacytic lymphoma (LPL) are low grade lymphoproliferative disorders and distinct clinicopathological entities. Although a predominantly bone marrow-based disease with varying degree of plasmacytic differentiation, unlike multiple myeloma focal bone involvement or bony sequalae are not typically described. We describe a cohort of patients treated at our specialist centre with focal skeletal involvement or bone complications from WM/LPL.

Methods
All patients with a diagnosis of WM/LPL in the University College London Hospital WM registry were evaluated. All patients with focal or abnormal bone infiltration on computer tomography (CT) and/or magnetic resonance imaging (MRI) were included in this retrospective analysis.

Results
A total of 570 patients with a diagnosis of WM/LPL were identified, of whom 22 (4%) had symptomatic focal bony involvement: 4 patients were diagnosed with transformation to high-grade B cell lymphoma and another plasmablastic lymphoma, and excluded from analysis. Infiltration with biopsy-proven or presumed LPL occurred in 17 (3%) patients.

The median age at WM diagnosis was 59 years and 9/17 (53%) were male. All patients experienced skeletal pain, prompting further investigation. Median time from WM/LPL diagnosis to confirmed bone involvement was 53 months (range 0-158). Four patients had image-guided biopsies to make the diagnosis and rule out transformation. A presumptive diagnosis of infiltration was made in the remainder due to imaging characteristics as assessed by musculoskeletal radiologists. Limb involvement was present in 10/17 (59%) patients, axial involvement 8/17 (47%) and skull involvement in 2/17 (12%). Distal symmetrical limb involvement was seen in three of these cases. Bone infiltration was apparent on CT in 8/11, MRI 14/14 and FDG-avid on CT PET in 7/10 cases.

An IgM paraprotein was present in 11/17, IgG in 3/17, and 3/17 had both IgM and IgG paraproteins. MYD88 mutation analysis was undertaken in 13 cases; 11 were of MYD88^L265P and two MYD88^WT.

Prior lines of treatment (median 2, range1-7) had been received in 8/17 cases. Three cases occurred in patients already established on Bruton’s Tyrosine Kinase inhibitors (BTKis;Ibrutinib 2, Zanubrutinib 1) of whom two were managed symptomatically and one stopped BTK. Bone involvement occurred in treatment-naïve patients in 10/17 cases.

Treatment was given in 14/17 cases. One patient died prior to starting therapy. Therapy included rituximab-bendamustine (n=5), dexamethasone-rituximab-cyclophosphamide (n=3,) rituximab-cyclophopshamide-doxorubicin-vincristine-prednisolone (R-CHOP) (n=3), rituximab-prednisolone (n=1), bortezomib-dexamethasone (n=1), ibrutinib (n=1), and rituximab-etoposide-cytarabine-methylpredisolone-cisplatin (R-ESHAP) and autologous transplant consolidation (n=1). Radiotherapy was incorporated in three cases. Symptomatic improvement was seen in all evaluable (n= 14) patients, with resolution in 10/14 (71%). One patient had discontinued BTKi due progressive bone disease, and experienced complete resolution with subsequent R-Bendamustine. The patient treated with Ibrutinib for bone disease experienced partial symptomatic and imaging resolution.

Two patients subsequently relapsed with progressive bone disease, requiring further therapy with rituximab-prednisolone (n=1) and bortezomib-dexamethasone-rituximab (n=1) .

Fracture secondary to bone involvement was seen in only 1/17 patients, who had multilevel vertebral collapse and >90% LPL infiltration on bone biopsy with osteopaenia. Concurrent Bing-Neel syndrome was present in 2/17 patients (12%), both of whom had extensive skeletal involvement.

Median follow up duration from diagnosis of bone disease was 102.5 months (2-218m) and 16/17 patients were alive at the time of data collection.

Conclusion
Although relatively rare, focal and symptomatic bone involvement with WM/LPL occurs. High-grade transformation is responsible for a minority of cases of bone involvement in WM/LPL. IgG LPL was overrepresented in the cohort, at 1/3 of patients. Symptomatic symmetrical distal limb involvement was seen in several patients. Fractures are rare. All patients in this series who received treatment experienced symptomatic improvement, with complete resolution in the majority. MRI is the most sensitive imaging modality.