The Location of Multiple Myeloma Adhesion Variants in Diverse Niches within the Bone Marrow Affects Plasma Cells Enumeration.

Background - The diagnosis of multiple myeloma (MM) is based on clinical and laboratory criteria combined with bone marrow (BM) plasmocytosis, estimated by inspection of bone marrow aspirates. Recent advances in flow-cytometry (FCM) have provided an additional tool for the diagnosis of MM and for monitoring response to therapy. However, significant discrepancy has been reported regarding the enumeration of plasma cells in marrow samples of MM patients using these two methods.

Aims - In this study we compared the bone marrow plasmocytosis by microscopic examination of BM aspirates, to the flow cytometry results in samples obtained form MM patients. We tested whether the noted discrepancy between these two methods applies only to MM, or represents a trend in other hematopoietic malignancies as well. We defined this discrepancy and explained it.

Methods - The number of plasma cells or blasts from BM aspirates of 41 MM or seven acute myeloid leukemia (AML) patients respectively were analyzed simultaneously by morphological evaluation and by FCM. Each sample was assessed independently by two qualified laboratory specialists and/or hemato-pathologist. In MM we found plasma cell fractions that were characterized by FCM and gene expression profile.

Results - In MM it was evident that FCM under-estimated the number of BM plasma cells samples by an average of 60%, compared with conventional morphological evaluation. On the other hand in AML there was a good correlation between the morphological and FCM assessments of the blast cell population, indicating that the discrepancy observed in the MM BM samples may be related to unique characteristics of the malignant plasma cells.

Since flow cytometry is performed on the bone marrow fluid which is depleted of fat tissue-adhesive plasma cells, we disrupted spicules from MM BM samples (by repeated passages through 21g needle) and found a 40% increase in plasma cell compared with the fluid of the same BM samples.

In order to determine the FCM profile of the cells in these two fractions, we isolated BM derived spicules from aspirates of MM patients and treated them with extracellular matrix (ECM) degrading enzymes followed by mechanical shearing. This combination released the highly adhesive plasma cells from the spicules. The released myeloma cells displayed a different FCM profile and in particular had a higher level of CD138 expression. Gene expression profile, which was performed on similar adhesion variants of cultured MM cells, demonstrated distinct oncogenic and transcriptional programs.

Summary - We have shown a major discrepancy between the percentage of MM cells obtained by routine BM morphology and flow cytometry counts. It is possible that this discrepancy is partially attributable to the two distinct microenvironmental components occupied by MM cells in the BM sample - the lipid spicules, and the fluid phase. MM cells located in different niches of the BM also differ in their FCM and gene expression profile. This study indicates that multiple myeloma patients contain heterogeneous populations of malignant plasma cells. These sub-populations may play distinct roles in the biological and clinical manifestations of the disease and differ in their response to anti-myeloma therapy.