This study was performed using 105 leftover EDTA-peripheral blood samples stored at 4°C up to seven days. They were collected from 105 healthy individuals (67 males and 38 females) among the 73,727 healthy individuals who were ≥40 years old and visited the Health Promotion Center, Samsung Medical Center, Seoul, Korea, from June 2018 to August 2019. The inclusion criteria were as follows: (1) no history of malignancy, autoimmune disease or infectious disease and (2) the samples revealed idiopathic lymphocytosis (lymphocyte count >4.0×10⁹/L). In contrast to other studies that enrolled healthy individuals with normal blood (lymphocyte) count [5, 7], the present study enrolled individuals with lymphocytosis (>4.0×10⁹/L) because the results of our pilot study showed an extremely low prevalence of MBL in healthy Korean individuals older than 40 years and with normal blood counts (data not shown). We recorded the age, sex, and complete blood count (CBC) information, for the selected samples meeting the inclusion criteria. The CBC was determined on a Sysmex XN-9000 analyzer (Sysmex, Kobe, Japan) at the time of samples collection. The median age of the 105 healthy individuals was 56 years (range 40–81 years). The median absolute lymphocyte count was 4.3×10⁹/L (range 4.0–6.7×10⁹/L). The study was approved by the Institutional Review Board of Samsung Medical Center (IRB No. SMC 2018-05-083-001).

Identification and subtyping of MBL were performed by eight-color flow cytometry (FC) using the following monoclonal antibodies: CD5-fluorescein isothiocyanate (FITC) (Beckman Coulter Inc., Miami, FL, USA), kappa (κ) light chain-allophycocyanin (APC) (Becton Dickinson, San Jose, CA, USA), lambda (λ) light chain-phycoerythrin (PE) (Becton Dickinson), CD45-AmCyan (Becton Dickinson), CD19-peridinin-chlorophyll protein-Cyanine5.5 (Becton Dickinson), CD10-PE-Cyanine7 (Becton Dickinson), CD20-Pacific blue (Becton Dickinson), and CD23-APC-Cyanine7 (APC-Cy7; eBioscience, San Jose, CA, USA). The final concentration was adjusted to 100,000 cells per tube.

After incubation with the monoclonal antibodies for 15 minutes at room temperature (approximately 20 to 25°C), the samples were lysed by adding 2 mL of FACS Brand Lysing Solution (Becton Dickinson) and incubated for 10 minutes at room temperature. The samples were centrifuged at 540×g for 5 minutes and washed twice with 2 mL of phosphate-buffered saline containing 0.5% bovine serum albumin (Sigma-Aldrich, St. Louis, MO, USA) and 0.09% sodium azide (Sigma-Aldrich). Data were acquired on a three-laser FACSCanto II flow cytometer (Becton Dickinson) and analyzed with Kaluza software (Becton Dickinson). The presence of a monoclonal B-cell subset was determined by a κ/λ ratio of >3:1 or κ/λ <0.3:1) or more than 25% of the B cells lacking surface immunoglobulin [5]. For suspected MBL cases with a monoclonal B cell subset, extended phenotyping was performed to determine the immunophenotypic characteristics of the clonal B cells using the following monoclonal antibodies: CD5-PE-Cyanine7 (Beckman Coulter), CD19-Pacific blue (eBioscience), CD38-APC-Cy7 (eBioscience), CD79b-APC (Becton Dickinson), FMC7-FITC (eBioscience), and CD45-AmCyan (Becton Dickinson). The prevalence of MBL was estimated using 95% confidence interval (CI) for the one-sample proportion. The estimation was performed with SPSS for Windows, version 11 (SPSS Inc., Chicago, IL, USA).

The overall prevalence of MBL was 2.9% (3/105, 95% CI, 0.9–8.1%) in healthy Koreans with idiopathic lymphocytosis. To the best of our knowledge, this is the first study on the MBL prevalence in an East Asian population, and the prevalence demonstrated here (3/105, 2.9%) is lower than that reported in other studies conducted using a similar sensitive method and age group (5.7–14.3%) [5–8]. Our calculations suggested that 105 samples would be required to obtain a 95% CI with a±3% margin of error. A summary of the prevalence of MBL across various geographical regions is provided in Table 1.

For three MBL cases, detailed laboratory findings and immunophenotypic characteristics are shown in Table 2. Using the diagnostic criteria proposed by Marti, et al. [4], we classified Case 2 as atypical CLL-like MBL and Case 3 as CD5−MBL. Based on the cell count (cut-off value of 0.5×10⁹/L), Case 1 and Case 3 can be categorized as high-count MBL (2.22×10⁹/L and 4.26×10⁹/L, respectively), which is also known as clinical MBL [9]. Parikh, et al. [10] conducted a large cohort study to assess the clinical progression of high-count MBL using the Mayo Clinic CLL database. They found that 7% of the patients in the cohort were treated for progression to CLL, and 0.4% required therapy for high-grade lymphoma. For Case 1, regular annual check-up showed monoclonal B cells with the same phenotype. However, since the absolute count of monoclonal B cells increased to 3.35×10⁹/L, we recommended close follow-up to quickly identify progression to CLL and take adequate action as needed. In particular, genetic variants in SF3B1 and NOTCH may be important for predicting the prognoses of CLL patients, because approximately 10–15% of CLL cases and approximately 1–3% of high-count MBL cases harbor these genetic variants [11–13].

We classified Case 2 as atypical CLL-like MBL based on the immunophenotyping results (expression of CD5 with strong expression of CD20), which revealed CD5⁺ MBL featuring monoclonal B cells that were CD23-negative and CD79b-positive. However, FMC7, which is an important marker for distinguishing CLL and mantle cell lymphoma, shows an intermediate expression pattern in the majority of atypical CLL-like MBL and CD5⁻MBL cases, in contrast to CLL-like MBL as in our Cases 2 and 3 [14]. In addition, several studies have reported that CLL cases with atypical features such as a positive reaction to FMC7 antibodies and/or a negative reaction to CD23 antibodies are more common in Asia [15, 16]. Although we did not perform molecular tests to confirm a translocation between 11q13 and 14q32, which is the hallmark of mantle cell lymphoma, Case 2 can be classified as atypical MBL given the lack of a specific finding during the regular check-up, including physical examination.

The phenotypic characteristics of CD5⁻ MBL differ substantially from those of the other two MBL types. Unlike CLL-like MBL, the monoclonal B cells of CD5⁻ MBL are characterized by expression of CD20 and CD79b, with strong surface expression of immunoglobulin and no expression of CD5 and CD23, as in Case 3 [17, 18]. Parker, et al. [19] reported that CD5⁻MBL might be a precursor stage of splenic marginal zone lymphoma owing to a common genetic basis. With the exception of a cross-sectional study conducted in rural Uganda, which reported a high prevalence of CD5⁻ MBL (13.6%), most population-based studies have found a relatively lower prevalence of CD5⁻MBL (0.5–2.3%) compared with that of CLL-like MBL (Table 1) [3, 5–8].

The main limitation of this study is that it is a relatively small-scale, single-center study attempting to represent the Korean population, which did not account for the age and gender distribution among the MBL subtypes. Moreover, although several studies have shown that certain genetic variants are associated with CLL and high-count MBL outcomes, we could not confirm the presence of a chromosome abnormality or genetic variation in the three identified MBL cases. Therefore, a detailed follow-up study should be conducted to determine the potential for progression to CLL or other lymphoproliferative diseases in these MBL cases.

In conclusion, this is the first study on the MBL prevalence in an East Asian population and reveals a relatively low prevalence of MBL in healthy Korean individuals with lymphocytosis.