Abstract
Background Chronic lymphocytic leukemia (CLL) in most patients is diagnosed with early stage disease identified incidentally on blood counts obtained for unrelated purposes. Immunophenotyping of peripheral blood (PB) is required for the diagnosis of CLL. A scoring system that helps in the differential diagnosis between CLL and other mature B-cell neoplasms (MBN) has been described twenty years ago (Matutes et al., Leukemia 1994; modified by Moreau et al., Am J Clin Pathol 1997). CLL/SLL typically demonstrates low-intensity staining for surface immunoglobulin, low or absent expression of CD22, CD79b and FMC7 and moderate to strong expression of CD5 and CD23. However, this phenotype is not entirely specific and some overlap in immunophenotype exists between CLL and non-CLL MBN. In particular, leukemic phase of CD5 positive mantle cell lymphoma (MCL) can be misdiagnosed as CLL. Recently, it has been shown that CD200 expression may help in differential diagnosis between CLL and other MBN. The present study aimed to prove CD200 usefulness in differentiating CLL from MCL on a series of consecutive patients and to investigate whether adding CD200 could improve the utility of Matutes scoring system, especially in atypical CLL.
Methods Between January 2013 and March 2014, PB of consecutive patients with MBN was assessed in this study. Analysis was performed on a FACSCalibur flow cytometer (Becton Dickinson) and samples were stained with panels of 4-color combinations of antibodies using a standard whole-blood assay. PB specimens were incubated with antibodies purchased from eBioscience (CD200 APC, clone OX-104), Immunotech (CD23, CD79b, FMC7), BD Biosciences (CD5, CD19), and DAKO (sIg). At least 5,000 B-cells were immediately acquired on flow cytometer. Diagnosis of CLL was made according to National Cancer Institute-Working Group criteria. Furthermore, tissue biopsies of 62 (31%) CLL cases were available for histological review, including all cases of atypical CLL. Diagnosis of MCL was based on morphology and immunohistochemical detection of cyclin D1 in tissue biopsies and further confirmed by detection of t(11;14) by FISH in selected cases.
Results Table 1 provides details of the patient characteristics. In our series, CD200 was present on neoplastic B-cells of all 200 CLL cases (100%), whereas only 4 cases (8.7%) of MCL showed dim positivity of CD200. The remaining 42 cases (91.3%) of MCL were negative for CD200 expression. The revised Matutes score was calculated to classify CLL cases. All 179 cases of typical CLL (defined by a score ≥ 4) presented moderate to strong expression of CD200 (Median fluorescence intensity - MFI: median = 161). CD200 was also positive in all 21 cases of atypical CLL (defined by a score < 4), but showed lower intensity (MFI: median 128) than that observed in typical CLL (P = 0.02). Application of the Matutes scoring system to MCL cases showed that three cases scored 3 (6.5%), two cases scored 4 (4.3%) and none scored 5. Of note, CD200 was absent in two cases scoring 3 and was only dimly expressed in the remaining MCL cases scoring 3 or 4. Thus, the differential expression of CD200 in CLL and MCL retained even in those cases with otherwise indeterminate immunophenotype, therefore being particularly helpful for the distinction of atypical CLL and MCL.
Conclusions Flow cytometry is an essential tool for the diagnosis of CLL. However, a significant immunophenotypic overlapping occurs especially between CLL and MCL cells. In this study, we investigated the expression of recently identified marker CD200 in PB of consecutive CLL and MCL patients. We have confirmed previous reports that CD200 is consistently expressed in all typical CLL. Furthermore, CD200 was expressed by all immunophenotypically atypical CLL cases. On the contrary, in MCL patients CD200 showed only a dim positivity in four subjects and was absent in the remaining 42. The inclusion of CD200 in the MBN routine flow cytometry panels facilitates the differential diagnosis between CLL and MCL and has a great impact on accurate diagnosis in cases with immunophenotypic aberrancies.
This work was supported by grant RVO VFN64165 and PRVOUK P27/LF1/1
Table 1 MCL (46 pts.) CLL (200 pts.) Age (median, range) 66.7; 47.8-82.4 67.6; 32.2-90.7 Sex (F/M) 19/27 74/126 WBC x109/L (median, range) 10; 2.1-285.4 21.9; 2.8-375.2 % neoplastic B-cells of WBC (median, range) 17.1; 1.3-90.5 54; 1.7-94.7 CD200 MFI (median, range) 2.16; 1-53.2 147.5; 20.6-637
Disclosures
No relevant conflicts of interest to declare.
A laboratory-based scoring system predicts early treatment in Rai 0 chronic lymphocytic leukemia
