Neural cell adhesion molecule (NCAM)/CD56 plays a crucial role in supporting normal hematopoiesis by mesenchymal stem cells, albeit it is currently suggested that its expression is largely restricted to natural killer cells and a subset of T cells among normal hematopoietic cells. Expression of this molecule by clonal plasma cells (PC) is found in around 50-80% of multiple myeloma (MM) patients and might be associated with a favorable prognosis. The biological function of CD56 expression by clonal - and potentially also normal - PC is still unknown.

We studied the immunophenotype of normal (i.e. polyclonal) PC originating from tonsil (n=11), peripheral blood (PB, n=24) and bone marrow (BM, n=24) with 8-color multiparameter flow cytometry. Hereby, a large panel of 41 different monoclonal antibodies directed against antigens involved in differentiation (e.g. CD45, CD138), survival (e.g. CD95, bcl-2) and adhesion/homing (e.g. CXCR4, CD31, CD49d, CD56) was used. This also included analysis of CD56- and CD56+ normal PC populations - the latter was detected in a small PC fraction of all analyzed normal BM samples -. Clinical and biological characteristics of MM with CD56-vs. CD56+ clonal PC were also studied in a cohort of 401 patients enrolled in two prospective trials of the Spanish Multiple Myeloma group.

Normal PC originating from tonsil, PB and BM showed a gradual increase in CD44, CD49d and CXCR4 expression, whereas the expression intensities of other markers (e.g. CD45, CD81, CD95, HLADR, CXCR5) were gradually decreased. Normal PC from tonsil and PB were systematically CD56-. In turn, a small fraction of CD56+ PC (median 9% of all PC) was detected in all investigated normal BM samples. CD56+vs. CD56- normal BM PC further showed a more mature phenotype with decreased CD19, CD45 and CD95 expression, but increased CD31, CD44, CD49d, CD138 and bcl-2 expression intensities. These results suggest a more mature immunophenotypic profile, less prone to undergo apoptosis of CD56+vs. CD56- normal BM PC.

In MM, CD56 positivity of clonal PC was significantly associated with absence of CD19, CD27 and CD28 expression. Hierarchical cluster analyses using median fluorescence intensities further showed that CD56 expression had the most important impact on clustering of clonal PC by their overall immunophenotypic profile. In addition, we found that cases carrying t(4;14) were virtually always CD56+ (98%) (P < .001). Conversely, presence of t(14;16) and t(11;14) were significantly (P < .001) associated with infrequent CD56 expression (13% and 42%, respectively). CD56 positivity of clonal PC was also associated with other important disease characteristics such as DNA hyperdiploidy by flow cytometry (65% vs. 36%, P < .001) and a higher frequency of major bone lesions (39% vs. 24%, P < .05). Cytogenetic abnormalities and the expression status of CD27 and CD81 - but not of CD56 - were of prognostic value.

In summary, we found that there systematically exists a small subset of CD56+ PC in normal BM; such cells displaying a mature immunophenotype vs. CD56- BM PC. These results suggest that CD56-vs. CD56+ MM could mainly represent accumulation of clonal PC at different maturation stages, but also with slightly different cytogenetic patterns.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution