Abstract
Background
In plasma cell neoplasms, the percentage of plasma cells in bone marrow (PPCBM) is important for diagnosis and assessment of the treatment. This quantification is performed using bone marrow multiparameter flow cytometry (FC), aspirate smears (BMA) and biopsy. Differences among the percentage of infiltration detected by these techniques have been reported, which can be related to a heterogeneous pattern of infiltration of multiple myeloma (MM) or sample quality. However, a simultaneous evaluation of these three techniques has not been reported nor their ability to detect bone marrow involvement with high or low infiltration, associated with different stages of this disease.
Purpose
The aim of this study was to compare the results of PPCBM obtained by FC, BMA, and biopsy with CD138 immunohistochemistry (BMB) in different stages of the disease with high and low infiltration, and the ability of these techniques to correctly classify the disease.
Methods
Pathological studies of patients referred to Hospital Fundación Santa Fe, Colombia were reviewed between January 2015 and June 2018. The selection of patients was based on both, a diagnosis of plasma cell neoplasms, classified according to the International Myeloma Working Group criteria and the simultaneous use of FC by FACSCanto II flow cytometer (Infinicyt 2.0 software program), wright-stained aspirate smears and biopsy with CD138 immunohistochemistry in the detection.
Descriptive analysis was performed and PPCBM was expressed as the mean± standard error of the mean (SEM). The Kruskal-Wallis test was used to compare the mean of PPCBM among the three methods. Lineal regression and Spearman´s coefficient were used to correlate the variables. Statistical association was considered significant for p values <0.05.
Results
A total of 130 patients with plasma cells neoplasms were included in this study, 57 with newly diagnosed MM (N), 58 following therapy MM (F) and 15 MGUS patients. In the N patients, the mean of infiltration was 12.3% ±1.8, 30.3% ±3.4 and 52.2% ±3.74 detected by FC, BMA and BMB, respectively. The PPCBM detected by this three analysis were significantly different (p<0.001). In F patients, the mean of infiltration was 0.37% ±0.05, 1.88 % ±0.16 and 2.61% ±0.21 while in MGUS-patients was 0.49% ±0.09, 2.067 % ±0.37 and 2.4% ±0.28 detected by FC, BMA and BMB, respectively. For F and MGUS patients, significant statistic differences between BMB and BMA versus FC (p<0.01) were observed, but not between BMB and BMA (Figure 1).
In the comparative analysis of the three methods in all patients, the highest infiltration was always detected by BMB, followed by BMA and finally by FC (p<0.01) (Figure 2). Linear regression established that for each 1% of infiltration detected by FC or BMA, the BMB identified 2.11% and 1.4% of infiltration, respectively (Figure 2). However, each univariate model only explained 55% and 67% of the observed results. Notably, there was a high correlation among these three techniques (Spearman´s coefficient > 0.8).
Finally, given that there are important PPCBM such as ≥10 that allows establishing the diagnosis of MM or ≥60%, considered an MM definitive event, we found that 21% and 52% of cases assessed by BMA and FC had PPCBM <10%, but all of them were reclassified as MM with BMB. None case had PCBM <10% using BMB. Finally, among all cases diagnosed as MM, there were identified 26 by FC and 35 cases by BMA with a PCBM percentage between 10 and 59%, of which 84% and 54% respectively had ≥60% of involvement detected by BMB
Conclusion
In the comparison of the PCBM was observed a high linear correlation between MFC, BMA, and BMB, although we found a differential behavior of these methods, depending on the level of tumor infiltration.
High percentages of infiltration, such as newly diagnosed MM patients, the BMB detected significantly more PPCBM than the others two methods (1.7 to 4-fold compared to BMA and FC). This supports the systematic incorporation of BMB into the analysis of patients with suspected MM for allowing a proper classification of disease.
With low percentages of infiltration, such as MGUS and following therapy patients, the difference of PPCBM between BMA and BMB was not significant. Although, FC detected
the lowest percentage of infiltration, it known its high specificity to discriminate tumor and normal plasma cells, being in one of the most important analysis to monitor minimal residual.
Disclosures
No relevant conflicts of interest to declare.
A Limited Plasma Cell Flow Cytometry Panel With Reflex CD138 Immunohistochemistry Is an Optimal Workflow Process for Evaluating Plasma Cell Neoplasms in Bone Marrow Specimens