To the Editor:
We have read with interest the paper of Witzyg et al1 about the prognostic value of peripheral blood clonal plasma cells (PBPC) in patients with multiple myeloma (MM). We have recently evaluated the presence of PBPC at diagnosis in 49 MM patients. Plasma cell leukemia cases were excluded (>20% or >2,000/μL plasma cells in PB).2
We have also studied the relationship with other clinical and biologic variables, including age and stage (Durie and Salmon system), performance status according to the Eastern Cooperative Oncology Group scale, the presence of bone lesions, bone marrow plasmocytosis, and serum concentrations of creatinine and β2 microglobulin. The expression of Ki67, a monoclonal antibody that detects cells in cycle, was evaluated in both bone marrow and PBPC. In addition, patients were evaluated for response to therapy and survival.
Mononuclear cells were isolated from PB by ficoll-hypaque density gradient centrifugation and cytospins were stored at −20°C until immunostaining was performed. Sequential double immunoenzymatic staining was applied as described previously,3 using an immunoperoxidase sandwich technique with Ki67 and the alkaline phosphatase/antialkaline phosphatase method to detect κ and λ Ig light chains. The proportion of PBPC that expressed the same light chain isotype as the patient M protein was evaluated by two independent observers examining 2,000 mononuclear cells from cytocentrifuge slides. The Ki67 proliferative index was estimated using bone marrow plasma cells. In patients with PBPC we found that the plasma cells of PB can be used to determine the Ki67 proliferative index with results equivalent to the bone marrow plasma cells.
. | LPC Group . | HPC Group . | P Value . |
---|---|---|---|
. | (BPC <1%) . | (BPC ≥1%) . | . |
. | No. (%) . | No. (%) . | . |
. | (n = 31) . | (n = 20) . | . |
Mean age (range) | 69 (42-87) | 71 (51-91) | NS* |
M-component type | 14 (70%) | NS† | |
IgG type | 22 (71%) | ||
Other | 9 (29%) | 6 (30%) | |
ECOG | P < .01† | ||
0-2 | 22 (71%) | 5 (25%) | |
3-4 | 9 (29%) | 15 (75%) | |
Stage | P < .01† | ||
I or II | 18 (58%) | 4 (20%) | |
III | 13 (41%) | 16 (80%) | |
Lytic bone lesions | NS† | ||
≤3 or osteoporosis | 22 (71%) | 14 (70%) | |
>3 | 9 (29%) | 6 (30%) | |
Creatinin (mg/dL) | 1.13 (0.6-6.4) | 1.55 (0.6-7.1) | NS* |
CD56− | 13 (42%) | 8 (40%) | NS† |
CD56+ | 18 (58%) | 12 (60%) | |
% Infiltration | 31.07 ± 17.35 | 53.44 ± 22.45 | P < .01* |
Ki67 | 2.16 ± 2.7 | 6.53 ± 5.27 | P < .01* |
β2 microglobulin | 4.34 ± 2.85 | 8.88 ± 8.66 | P < .01* |
Response | 14/21 (67%) | 2/17 (12%) | P < .05† |
. | LPC Group . | HPC Group . | P Value . |
---|---|---|---|
. | (BPC <1%) . | (BPC ≥1%) . | . |
. | No. (%) . | No. (%) . | . |
. | (n = 31) . | (n = 20) . | . |
Mean age (range) | 69 (42-87) | 71 (51-91) | NS* |
M-component type | 14 (70%) | NS† | |
IgG type | 22 (71%) | ||
Other | 9 (29%) | 6 (30%) | |
ECOG | P < .01† | ||
0-2 | 22 (71%) | 5 (25%) | |
3-4 | 9 (29%) | 15 (75%) | |
Stage | P < .01† | ||
I or II | 18 (58%) | 4 (20%) | |
III | 13 (41%) | 16 (80%) | |
Lytic bone lesions | NS† | ||
≤3 or osteoporosis | 22 (71%) | 14 (70%) | |
>3 | 9 (29%) | 6 (30%) | |
Creatinin (mg/dL) | 1.13 (0.6-6.4) | 1.55 (0.6-7.1) | NS* |
CD56− | 13 (42%) | 8 (40%) | NS† |
CD56+ | 18 (58%) | 12 (60%) | |
% Infiltration | 31.07 ± 17.35 | 53.44 ± 22.45 | P < .01* |
Ki67 | 2.16 ± 2.7 | 6.53 ± 5.27 | P < .01* |
β2 microglobulin | 4.34 ± 2.85 | 8.88 ± 8.66 | P < .01* |
Response | 14/21 (67%) | 2/17 (12%) | P < .05† |
Comparisons between groups were made by the non-parametric Mann-Whitney test (*) for quantitative variables and the χ2 test for qualitative variables (†).
Abbreviation: NS, nonsignificance (P value >.05).
The mean percentage of PBPC was 0.87% of the mononuclear fraction (range, 0% to 14%). A cut-off point of 1% was chosen because it was as predictive as the mean value and easier to use.
The patients were classified into two groups: with high number of PBPC (group HPC, ≥1%) and a lower number (group LPC, <1%). Table 1 presents the distribution of the biologic and clinical variables according to the number of PBPC.
The method of Kaplan and Meier4 was used to compute the survival curves and to estimate median survival. The differences in survival curves were tested with the log rank statistic method. An evaluation of survival using the cut-off value is summarized in Fig 1A. The patients in the HPC group presented a median survival of 6 months, which was significantly shorter than in the LPC group (>18 months).
The interaction between the number of PBPC and other single parameters was estimated. We found statistically significant correlations between HPC and advanced Salmon-Durie stage and poor performance status (χ2 test, P < .01). Moreover, the β2 microglobulin levels, percentage of bone marrow plasma cells, and bone marrow or peripheral blood plasma cell proliferative index were increased in the group with HPC, with a significant difference being observed with respect to the group with LPC (U-Mann Whitney, P < .01). There were no significant differences between patient groups comparing other clinical characteristics, including age, advanced lytic bone lesions, creatinin levels, and type of treatment. Loss of CD56 antigen, which was postulated to contribute to the leukemic expression in MM,5 was also similar.
Because there was a strong correlation between HPC and stage III of MM, a separate analysis of survival was made including only patients in stage III. Statistically significant differences in survival were observed as complete series (Fig 1B). In addition, HPC before treatment was associated to a low response rate (χ2 test, P < .05).
In conclusion, the methodology described allows us to easily identify and quantify circulating plasmatic cells in an important proportion of MM patients, as well as to establish the proliferative index of disease. We confirm the adverse prognostic implications of PBPC in MM observed by Witzyg et al.1 In our series, the proportion of PBPC identified a subgroup of MM characterized by advanced stage, poor status performance, high proliferative index, and short survival.
This feature is available to Subscribers Only
Sign In or Create an Account Close Modal