Belinda Pope

The labelling index of primitive plasma cells determines the clinical behaviour of patients with myelomatosis

For patients with multiple myeloma the most important laboratory correlate of prognosis and disease activity is the bromodeoxyuridine (BrdUrd) plasma cell labelling index (LI). However, the traditional immunofluorescent microscope LI technique, like other manual enumeration assays, can suffer from poor precision and accuracy. In this study the LI of different subpopulations of plasma cells (CD38++) as determined by flow cytometry was correlated with disease state. The mean LI of the total CD38++ population was significantly higher (2.7±0.4%) than the LI determined by the traditional slide technique (0.6±0.1%) for 65 samples tested. Primitive plasma cells (CD38++, CD45++) had a higher labelling index than mature plasma cells (CD38++, CD45−) (7.0±1.3% v 1.8%±0.3%) and in one patient the LI of the primitive plasma cells was 46%. In addition, the LI of the mature plasma cells was lower than the total plasma cell population. As expected, there was a significant difference between the LI of patients in plateau phase and progressive disease but this difference was greatest when the LI of the primitive plasma cells was studied (9.2±2.9% v 2.2±0.7%; z=19.9, P<0.001). This study has raised some concerns about the sensitivity and accuracy of the traditional labelling index and has shown that the increased LI associated with progressive disease is almost entirely attributable to an increase in the LI of the primitive plasma cell subpopulation and that the LI of primitive plasma cells provides a more clinically significant correlation with disease status than the traditional assay.

T‐cell expansions in patients with multiple myeloma have a phenotype of cytotoxic T cells

The presence of T‐cell clones in peripheral blood has been previously shown to be associated with a survival advantage in patients with multiple myeloma and suggests that the expanded T‐cell populations may be involved in an anti‐tumour response. We studied the T‐cell receptor (TCR) repertoire of 38 patients with myeloma to identify and characterize the expanded T‐cell populations by flow cytometry. T‐cell expansions were found in 79% of the patients. The expansions occurred randomly among the 21 variable regions of the TCR β chain (Vβ) studied, representing 62% of the V‐β repertoire, and were stable during an 18‐month follow‐up. The phenotype of the expanded V‐β populations was predominantly CD8+, CD57+, CD28− and perforin+, which differed significantly from the other non‐expanded Vβ populations. The expression of the apoptosis markers Fas (CD95) and bcl‐2 were similar between the expanded and non‐expanded Vβ populations. In conclusion, expanded T‐cell populations were frequent in patients with myeloma, they remained unchanged during follow‐up and had phenotypic characteristics of cytotoxic T cells. These data add further support to the concept that the T‐cell expansions may have an immunoregulatory role in myeloma.

The Expression of T Cell Related Costimulatory Molecules in Multiple Myeloma

Presentation of tumour antigen by malignant cells not expressing costimulatory molecules is considered to be a major cause of the failure of the hosts immune response against tumours. This study has determined the expression of the B7 family of costimulatory molecules on malignant plasma cells and the expression of the counter receptor molecules, CD28 and CD152 (CTLA-4), on T cells of patients with multiple myeloma. CD28 expression was present on most CD4 cells but was lower on CD8 cells especially from those patients who also showed evidence of expanded T cell clones (median 40%. z=2.4; p<0.02). CD152 expression was increased in 50% (9/18) of patients with myeloma. CD80 (B7-1) expression was present on the plasma cells of only 1 of 27 samples but CD86 (B7-2) expression within the normal range was present on the plasma cells of 14 of 27 samples. Primitive plasma cells (CD38++ CD45++) had a higher expression of CD86 (median 78%) than mature plasma cells (CD38++ CD45-) (median 19%, z=3.7; p<0.01). Thus patients with expanded T cell clones have a downregulated T cell CD28 expression and lack B7-1 expression on their malignant plasma cells. These results are consistent with the concept that engagement of the T cell receptor by tumour antigen on B7-1 deficient malignant plasma cells would result in T cell anergy rather than productive immunity.

Either interleukin-12 or interferon-γ can correct the dendritic cell defect induced by transforming growth factor β1 in patients with myeloma

The poor response to immunotherapy in patients with multiple myeloma (MM) indicates that a better understanding of any defects in the immune response in these patients is required before effective therapeutic strategies can be developed. Recently we reported that high potency (CMRF44+) dendritic cells (DC) in the peripheral blood of patients with MM failed to significantly up‐regulate the expression of the B7 co‐stimulatory molecules, CD80 and CD86, in response to an appropriate signal from soluble trimeric human CD40 ligand. This defect was caused by transforming growth factor β1 (TGFβ1) and interleukin (IL)‐10, produced by malignant plasma cells, and the defect was neutralized in vitro with anti‐TGFβ1. As this defect could impact on immunotherapeutic strategies and may be a major cause of the failure of recent trials, it was important to identify a more clinically useful agent that could correct the defect in vivo. In this study of 59 MM patients, the relative and absolute numbers of blood DC were only significantly decreased in patients with stage III disease and CD80 up‐regulation was reduced in both stage I and stage III. It was demonstrated that both IL‐12 and interferon‐γ neutralized the failure to stimulate CD80 up‐regulation by huCD40LT in vitro. IL‐12 did not cause a change in the distribution of DC subsets that were predominantly myeloid (CD11c+ and CDw123−) suggesting that there would be a predominantly T‐helper cell type response. The addition of IL‐12 or interferon‐γ to future immunotherapy trials involving these patients should be considered.

Nucleoside transporters, bcl-2 and apoptosis in CLL cells exposed to nucleoside analogues in vitro

The purine nucleoside analogues fludarabine (Fl) and chlorodeoxyadenosine (2‐CdA) are considered to be cell cycle specific agents which require DNA synthesis for cytotoxicity. However, their efficacy in the treatment of CLL, an indolent lymphoid malignancy suggests additional mechanisms of action. Like cytosine arabinoside (AraC), Fl and 2‐CdA gain access to the cell via a specific nucleoside transporter (NST) protein. To investigate the mode of action of these drugs in CLL, we used a fluorescent ligand for the NST (5‘‐(SAENTA‐ x 8)‐fluorescein) and 3‐colour flow cytometry to determine NST expression on CD5+/CD19+ B‐cells from the peripheral blood (PB) of patients with CLL. NST levels on these cells was found to be not significantly different from normal control lymphocytes (mean = 485±425) vs. (mean = 553±178). Exposure to varying concentrations (0, 3 μM and 30 μM) of Fl and 2‐CdA, however, resulted in an upregulation of NST (mean = 1552±775 with 30 μM FL; mean = 3392±2197 with 30 μM 2‐CdA) after 48 h. “Large” lymphoid cells (not present in normal PB) were found to express significantly more NST (mean = 2540±2861) and have a higher proliferative capacity than “small” cells (mean = 357±517 NST/cell). Incubation of CLL cells with Fl (n = 6) and 2‐CdA (n = 8) in vitro over 48 h also resulted in an increase in the proportion of cells in S‐phase (0 μM = 0.2±0.1; 30 μM FL = 2.4±2.0; 30 μM 2‐CdA = 3.3±1.3) and a significant increase in morphologically identifiable apoptosis. Apoptosis was confirmed by flow cytometric DNA analysis (0 μM = 13±8%; 30 μM FL = 40±20%; 30 μM 2‐CdA = 48±11%). In situ hybridization using a biotinylated cDNA bcl‐2 probe demonstrated that bcl‐2 mRNA expression was markedly decreased in treated cells after 24 h. These studies have demonstrated that: (1) NST expression on CLL lymphocytes is low; (2) in vitro exposure to the analogues increases both the level of NST expression and the % cells in S‐phase; (3) exposure to the analogues downregulates bcl‐2 expression and increases apoptosis.

The Oncoprotein Phenotype of Plasma Cells from Patients with Multiple Myeloma

The expression of 6 different oncoproteins and 2 tumour suppressor gene products in the plasma cells of 63 bone marrow samples was used to determine a profile of the oncogenic phenotype of patients with multiple myeloma. Dual label flow cytometry after periodatelysine paraformaldehyde fixation was used to detect cell surface phenotype and intracellular protein expression simultaneously. The normal range for both the incidence and intensity of expression was determined for each protein by analysing plasma cells (high CD38 intensity) in 22 normal bone marrow samples. The percentage of myeloma patients with a greater than normal incidence of plasma cells expressing these proteins was 53% for c-myc, 28% for Rb, 28% for bcl-2, 27% for c-fos, 24% for p53 wild, 22% for p53 mutant, 13% for c-neu and 13% for pan-ras. When a panel of 8 antibodies was used, 82% of the samples (n = 28) had an increased incidence of expression by at least one oncoprotein or tumour suppressor gene product. The 5 patients with a normal incidence of expression of all 8 proteins were in plateau stage and 4 had not received chemotherapy for more than 12 months. The number of patients with an increased incidence of expression by 2 or more oncoproteins was significantly greater (X2 = 9.0; p < 0.005) in progressive disease (55%) than in stable disease (14%) but there was no specific phenotype pattern associated with progressive disease. All 6 oncoproteins and both tumour suppressor gene products had a greater incidence and intensity of expression in progressive than in stable disease. The expression of c-myc oncoprotein correlated with c-myc mRNA expression in the same samples (n = 10) but c-myc did not correlate with either the plasma cell labelling index (r = -0.15) nor serum thymidine kinase (r = 0.10). Our results suggest that there is a heterogeneous, non-systematic but almost universal presence of activated oncogenes and tumour suppressor genes in the plasma cells of patients with multiple myeloma and that disease progression is associated with the accumulation of a variety of secondary genetic changes which confer increased malignant behaviour.

The bone marrow plasma cell labeling index by flow cytometry.

The bone marrow plasma cell labeling index is the most important prognostic indicator for patients with multiple myeloma. Traditionally, this test has been performed as a two color immunofluorescent microscope technique which is time consuming and requires a degree of subjectivity in its interpretation. We have assessed various adaptations of this method to flow cytometry. A bromodeoxyuridine method has been compared with a propidium iodide DNA method to detect cells in S phase and CD38-FITC has been compared with CD38-FITC + CD138-FITC and CD38-biotin + streptavidin FITC to identify plasma cells. The mean channel fluorescent intensity of the plasma cell peaks for each of these markers was 12. 7, 17.4 and 35.3 respectively demonstrating the superiority of CD38-biotin + streptavidin FITC. Analysis after propidium iodide staining provided a good correlation with the slide technique (r = 0. 71; P < 0.0001) but the bromodeoxyuridine method did not correlate with the slide method (r = 0.09; P = NS). The labeling index values obtained from either of the flow methods were greater than the microscopic method. Thus a labeling index of >4% will replace the traditional >1% threshold for identifying patients with a significantly increased labeling index. The advantages of the new method are that it takes less time to perform, is more objective and provides additional data on ploidy and cell cycle status.

Plasma cells in peripheral blood stem cell harvests from patients with multiple myeloma are predominantly polyclonal.

A flow cytometric technique has been developed to detect individual plasma cells in PBSC harvests and to establish light chain restriction as a surrogate marker of their clonality. Plasma cells were identified by high intensity CD38 (CD38++) and cytoplasmic immunoglobulin (cIg) expression. The ratio of cytoplasmic kappa to lambda expression was used to detect light chain restriction. All 25 PBSC harvests studied contained CD38++/cIg plasma cells (mean 0.7%, range 0.03–2%). Harvests from non-myeloma patients also contained plasma cells (mean 0.4%, range 0.01–1.5%). Most of the plasma cells detected in the harvests from myeloma patients were immature (CD45+/CD45++) rather than mature (CD45−). When the total plasma cell population was studied, definite isotype restriction could be detected in only 16% of harvests. Light chain restriction was found in 53% of harvests when the mature plasma cells (CD45−) were analysed but only in 9% of harvests when immature (CD45+/CD45++) plasma cells were analysed. Five percent of patients with myeloma had detectable light chain restriction in peripheral blood CD19+ cells. There was concordance between the ratio of malignant (CD19−/CD56+) to normal (CD19+/ CD56−) plasma cells and light chain expression in 86% of patients studied. This study has demonstrated that the majority of plasma cells in PBSC harvests from patients with myeloma are not only immature but are also predominantly polyclonal and that monoclonality is best detected in mature plasma cells.

THE FUNCTIONAL PHENOTYPE OF THE PRIMITIVE PLASMA CELL IN PATIENTS WITH MULTIPLE MYELOMA CORRELATES WITH THE CLINICAL STATE

The malignant plasma cells from patients with multiple myeloma display considerable phenotypic heterogeneity. All plasma cells express high intensity CD38 (CD38++), cytoplasmic immunoglobulin and either kappa or lambda light chains. Subpopulations of mature (CD45-), immature (CD45+) and primitive (CD45++, CD19+) plasma cells can be defined but little is known about the functional differences and clinical significance of these subpopulations. Three colour flow cytometry and permeabilisation was used to determine the expression of functionally important antigens in plasma cell subpopulations. These antigens included the labelling index (LI, bromodeoxyuridine), number of nucleoside transporter per cell, p-glycoprotein (JSB-1), and oncoprotein expression (c-myc, c-fos, c-neu, bcl-2, p-ras, p53m, p-53w, and Rb). In progressive disease there was an increase in the absolute number but not the percentage of CD45++ plasma cells. There was a significant difference in the mean LI of the CD38++, CD45++ population in progressive disease compared with stable disease (9.2% vs 2.2%; z = 19.9, p < 0.001). The LI of CD45++ cells ranged up to 45% and provided a better correlation with disease status than the LI of the total cell population. Any increase in nucleoside transporters or p-glycoprotein expression was almost entirely attributable to an increase in the primitive plasma cell population. In 96% (n = 28) of samples from patients in progressive disease there was at least one abnormality in the functional phenotype of the primitive plasma cells. This is in contrast with 44% of samples from patients in stable disease (n = 58). These studies suggest that the functional phenotype of the primitive plasma cell determines the clinical phenotype of patients with myeloma.

Multiple Myeloma: Expression of Nucleoside Transporters on Malignant Plasma Cells and Their Relationship to Cellular Proliferation

The expression of nucleoside transporters is a limiting factor in the pharmacology of the nucleoside analogue, cytosine arabinoside (AraC) and is associated with cellular proliferation. We investigated the expression of nucleoside transporters on plasma cells from the bone marrow of 51 patients with multiple myeloma by 2-colour immunofluorescence flow cytometry, utilising 5-(SAENTA-x8)-fluorescein, a fluorescent ligand for the nucleoside transporter and anti-CD38 conjugated to phycoerythrin, as CD38 expression has unique characteristics on plasma cells. Mean nucleoside transporter expression on bone marrow plasma cells from patients with myeloma (1777 +/- 2181 transporters/plasma cell) was not significantly different from expression on plasma cells from normal bone marrow (997 +/- 1096 transporters/plasma cell). However, analysis of disease subgroups revealed a significant trend towards increased transporter expression in patients with progressive disease compared to those with stable disease (chi 2 = 4.0, p < 0.05). Nucleoside transporter expression correlated significantly with the plasma cell labeling index (LI) (r = 0.45, p < 0.01) and serum thymidine kinase levels (r = 0.66, p < 0.01), both markers of cellular proliferation but not with c-myc oncoprotein expression. These findings suggest that flow cytometric measurement of nucleoside transporter expression on plasma cells provides a rapid and convenient measurement of disease activity or quiescence in myeloma.