|Home | About | Journals | Submit | Contact Us | Français|
Plasma cell dyscrasias are a heterogeneous group of disorders characterised by the expansion of a single clone of immunoglobulin secreting, terminally differentiated, end‐stage B cells.1 Many of their clinical and morphological features are the result of the production and accumulation of excessive amounts of monoclonal immunoglobulin. Intracellular inclusions, both intracytoplasmic (Russell bodies) and intranuclear (Dutcher bodies) are occasionally seen in bone marrow biopsy specimens from patients with plasma cell neoplasms, and less frequently in reactive states.2,3 Extracellular protein deposition also occurs, although less frequently. “Myeloma kidney” with tubular casts composed of a heterogeneous mixture of electron dense homogeneous material and Tamm–Horsfall protein, together with elongated crystals, some with a fibrillar or lattice substructure, is an important cause of myeloma‐associated morbidity and mortality.4,15 The formation of extracellular crystalline structures within the cornea, skin, thyroid gland, intravascular space and bone marrow along with other sites has been described.5,6,7,8,9 However, this phenomenon is unusual and is limited to isolated reports. In our department, we see approximately 1200 bone marrow trephines per year, including about 300 cases of multiple myeloma, and they have not previously come to our notice.
A 72‐year‐old man presented with back pain. A spinal x ray carried out six months later showed collapse of the first lumbar vertebra. Further investigations included haemoglobin 9.3 g/dl, creatinine 129 μmol/l, calcium 2.5 mmol/l, β2‐microglobulin 3.9 mg/l, IgG λ paraprotein 65 g/l and bone marrow aspirate showing 80% plasma cells.
The paraprotein decreased to 25 g/l after chemotherapy. Peripheral stem cell mobilisation and collection was carried out. A bone marrow trephine biopsy and aspirate were taken prior to peripheral blood stem cell transfer (PBSCT) with melphalan preconditioning. Six weeks after PBSCT, the paraprotein level was 16 g/l. Bone marrow trephine was repeated five months later.
The bone marrow trephines from before and after PBSCT and the corresponding bone marrow aspirates are the subject of this report.
The trephine biopsy specimens had been fixed in aceto‐zinc formalin and decalcified in 10% formic acid, then routinely processed and paraffin‐embedded. Sections (1 μm) were cut and stained with H&E, reticulin, Perls, Giemsa, toluidine blue and congo red stains. In the initial trephine, the cellularity was approximately 40%, with marked suppression of haematopoietic cell lines and replacement by sheets of plasma cells and plasmacytoid cells, which comprised approximately 80% of the marrow cells. There were numerous refractile rhomboidal and hexagonal eosinophilic crystals, predominantly in the interstitium. They were of variable length and had a diameter of 10–100 μm (fig 1A1A).). There was not an associated giant cell reaction. Occasional intracellular inclusions, including Russell bodies, were also noted. There was no evidence of metachromasia or of congo red positivity.
Immunohistochemical staining with CD79a (1:10; Dako, Cambridgeshire, UK), CD138 (clone BB4, 1:100; Serotec, Oxford, UK) and CD56 (1:100; Novocastra, Newcastle, UK) antisera was positive in the plasma cells. There was monotypic λ light chain expression. Staining with CD20 (1:200; Novocastra), EMA (1:100; Dako), cyclin D1 (1:50; Lab Vision, Newmarket, UK) and CD3 (1:50; Novocastra) was negative. The crystals showed weak, peripheral staining for λ light chain (1:10000, Dako). Staining for κ light chain (1:10000, Dako) was negative (fig 1B,C1B,C).
Unfortunately, tissue was not appropriately preserved for electron microscopy, which precluded ultrastructural examination.
The aspirate smears stained with May–Grunwald–Giemsa showed 23% of plasma cells, with easily identifiable extracellular crystals of similar appearance (fig 1D1D).
The second trephine contained 25% plasma cells with extracellular crystals, the proportion of which was similar with respect to the previous biopsy.
The interstitial deposition of extensive extracellular material in the form of eosinophilic crystals is a highly unusual and interesting feature of this case, the precise aetiology of which is unclear. A previous description of a case of monoclonal gammopathy of undetermined significance suggested that the crystalline appearance could be artefactual, due to precipitation during tissue fixation and processing.10 The observation that these structures were found in both the bone marrow aspirate and the trephine does not support this hypothesis, despite the lack of tissue reaction. Although immunohistochemistry for λ light chain is weak, these crystals are most likely to be composed of precipitated light chains. Keller et al have previously noted absence of staining for light chains in a case with renal crystalloid deposits.17 They have attributed this to conformational change of the protein with masking of epitopes during the process of crystallisation.
Colourless, refractile oxalate crystal deposition in the bone marrow interstitium occurs in oxalosis.16 Macrophages laden with cholesterol and cystine crystals may be identified in hyperlipidaemic states and in cystinosis respectively.16 All of these appear different to the eosinophilic crystals described in the current case; none has an association with myeloma.
Animal experiments on mice receiving intraperitoneal injections of myeloma‐associated monoclonal protein have shown tubular casts, basement membrane precipitates or crystals in mice kidneys, histology comparable to that seen in patients from whom the protein was derived.4 Molecular analysis of the light chain sequences in patients with crystal storing histiocytosis and Fanconi's syndrome secondary to light chain gammopathy have shown unusual amino acid substitutions, altering light chain hydrophobicity.11,12 These data indicate that the physiochemical properties of the monoclonal protein determine its tendency to crystallise, in addition to the local concentration of the protein. Other local factors such as pH, temperature and concentration of other solutes may determine the precise location and pattern of deposition in each individual case. Of note, the potential of crystallisation does not appear to depend on the serum concentration of paraprotein, as seen in the current case.
Unfortunately we have been unable to examine the trephine biopsy taken prior to the institution of chemotherapy. The proportion of crystals did not change following PBSCT, which suggests that therapy has not exerted a significant effect. Mullen and Chalvardjian have previously considered that this may be a therapy‐related phenomenon without proposing a potential mechanism.13 Crystals disappeared post‐therapy in cases of cutaneous crystal deposition due to multiple myeloma and in cryoglobulinaemia.6,14
The chief importance of these extracellular crystals is their morphological pointer towards the diagnosis of a clonal plasma cell proliferation. Renal extracellular crystal deposition has been associated with rapidly progressive disease18; whether these deposits in the bone marrow also herald clinically significant extracellular immunoglobulin deposition is as yet unknown.
The authors are grateful for the expert technical assistance of Ms Donna Horncastle and Ms Kay Elderfield.
Competing interests: None declared.