|Home | About | Journals | Submit | Contact Us | Français|
The diagnosis of monoclonal B-cell lymphocytosis (MBL) is used to characterize patients with a circulating population of clonal B cells, a total B-cell count of less than 5 × 109/L, and no other features of a B-cell lymphoproliferative disorder including lymphadenopathy/organomegaly. The natural history of clinically identified MBL is unclear. The goal of this study was to explore the outcome of patients with MBL relative to that of individuals with Rai stage 0 chronic lymphocytic leukemia (CLL).
We used hematopathology records to identify a cohort of 631 patients with newly diagnosed MBL or Rai stage 0 CLL. Within this cohort, 302 patients had MBL (B-cell counts of 0.02 to 4.99 × 109/L); 94 patients had Rai stage 0 CLL with an absolute lymphocyte count (ALC) ≤ 10 × 109/L; and 219 patients had Rai stage 0 CLL with an ALC more than 10 × 109/L. Data on clinical outcome were abstracted from medical records.
The percentage of MBL patients free of treatment at 1, 2, and 5 years was 99%, 98%, and 93%, respectively. B-cell count as a continuous variable (hazard ratio [HR] = 2.9, P = .04) and CD38 status (HR = 10.8, P = .006) predicted time to treatment (TTT) among MBL patients. The likelihood of treatment for MBL patients was lower (HR = 0.32, P = .04) than that of both Rai stage 0 CLL patients with an ALC less than 10 × 109/L (n = 94) and Rai stage 0 CLL patients with an ALC more than 10 × 109/L (n = 219; P = .0003).
Individuals with MBL identified in clinical practice have a low risk for progression at 5 years. Because B-cell count seems to relate to TTT as a continuous variable, additional studies are needed to determine what B-cell count should be used to distinguish between MBL and CLL.
The diagnosis of monoclonal B-cell lymphocytosis (MBL) is currently used to characterize patients who have a circulating population of clonal B cells, a total B-cell count of less than 5 × 109/L, and no other features of a lymphoproliferative disorder including lymphadenopathy or organomegaly.1 In most patients, the immunophenotype is similar to that of chronic lymphocytic leukemia (CLL).1,2
The natural history of CLL-phenotype MBL is unclear and could differ depending on whether patients are identified through population screening using highly sensitive assays2–4 or recognized in routine clinical practice when undergoing evaluation of low-grade lymphocytosis.5–7 Because population screening for MBL is not used outside of epidemiologic research studies, the issue oncologists currently encounter is how to counsel individuals with MBL identified in routine practice.5–8 There are minimal data on the natural history of clinically identified MBL,5,9 and there is even less information regarding whether it differs from Rai stage 0 CLL.6 To develop relevant information on the latter, we used the Mayo Clinic hematopathology records to identify patients with CLL-phenotype MBL diagnosed in routine clinical practice along with a comparison group of concurrently diagnosed patients with Rai stage 0 CLL to compare clinical outcomes.
With the approval of the Mayo Clinic Institutional Review Board, we used hematopathology records to identify a cohort of 1,815 patients who had peripheral-blood flow cytometry at Mayo Clinic between January 2000 and December 2006 that detected a clonal B-cell population of CLL phenotype.10,11 After elimination of patients with more advanced-stage (Rai stages I to IV) CLL, previously treated CLL, or an established diagnosis of CLL/MBL more than 1 year before the flow cytometric study, 631 unique patients were identified.
On the basis of our prior analysis8 demonstrating that nearly all patients with an absolute lymphocyte count (ALC) more than 10 × 109/L also have a B-cell count more than 5 × 109/L, of 631 patients, we classified the 219 patients with an ALC more than 10 × 109/L as Rai 0 CLL with ALC more than 10 × 109/L but pursued additional analysis in the 396 patients who had an ALC less than 10 × 109/L and the necessary data from flow cytometry analysis to determine the B-cell count. All peripheral-blood samples were evaluated by multicolor flow cytometric immunophenotyping analysis on a FACS Calibur flow cytometer (BD Biosciences, San Jose, CA) within 24 hours of blood sampling according to previously described methods using fluorochrome-conjugated antibodies.11 The proportions of B cells, T cells, and natural killer cells were determined in each patient and normalized to 100% lymphocytes, and then an absolute count of each lymphocyte subset (B, T, and natural killer cells) was calculated using a simultaneously run (± 10 days) CBC with differential cell counts. In addition to the number of total B cells, the number of clonal B cells (CD19, CD5 coexpression) was determined among patients with an absolute B-cell count less than 1.5 × 109/L using the same methods to assess how the proportion of B cells that are clonal relates to outcome among low-count MBL patients. The threshold of less than 1.5 × 109/L was chosen for this analysis because it was the upper limit of the B-cell range in the early studies of MBL among the normal population2 (we have previously shown that as the B-cell count increases, nearly all B cells are clonal B cells12). On the basis of current criteria, patients with a total B-cell count less than 5 × 109/L were classified as having MBL.1,10 Patients with a with a total B-cell count more than 5 × 109/L but an ALC less than 10 × 109/L were classified as low-count, Rai 0 CLL (Appendix Fig A1, online only).
Clinical information regarding date of diagnosis, physical examination, prognostic parameters, treatment history, and disease-related complications were abstracted from clinical records. Prognostic testing performed as part of clinical or research studies, including evaluation of IgHV gene mutation analysis, ZAP-70 status, CD38 status, and cytogenetics abnormalities by interphase fluorescent in situ hybridization (FISH), was performed as previously described by our group.13–15
Time to treatment (TTT) was defined as the time between date of diagnosis and date of initiation of first treatment or date of last follow-up at which patient was known to be untreated. Overall survival (OS) was defined as the time between date of diagnosis and date of death or date of last follow-up. The accepted indications to initiate treatment for CLL were based on the National Cancer Institute Working Group 1996 criteria during the study interval.16 Estimates of TTT and OS were calculated using the Kaplan-Meier method. Cox proportional hazards models were used to model the relationship of B-cell count or other prognostic parameters with TTT and OS. Likelihood ratio tests were used to test the effects of each individual factor. Hazard ratios (HRs) and 95% CIs were calculated from the Cox models. P < .05 was considered significant.
We identified 631 patients seen at Mayo Clinic between January 2000 and December 2006 who had a circulating clonal population of CLL phenotype on flow cytometry and no other characteristics of a lymphoproliferative disorder (eg, palpable lymphadenopathy, organomegaly, cytopenia caused by marrow infiltration, and so on). Of these, 219 patients had an ALC more than 10 × 109/L and were classified as Rai 0 CLL with ALC more than 10 × 109/L. Among the 396 patients with an ALC less than 10 × 109/L, 302 patients (76%) had a B-cell count less than 5 × 109/L at the time of flow cytometry analysis and met the current criteria for the diagnosis of MBL.1 The remaining 94 patients (24%) had a B-cell count more than 5 × 109/L (with an ALC < 10 × 109/L) and were considered low-count, Rai 0 CLL.10 Lymphocytosis (ALC > 3 × 109/L) was present in 263 (87%) of the 302 patients with MBL, and the median B-cell count was (2.76 × 109/L), which is slightly lower than that reported in the only other large MBL series (median, 3.3 × 109/L)5 reporting these data. Median age at the time of MBL diagnosis was 69 years (range, 34 to 93 years). The age, sex distribution, and hemoglobin levels of patients with MBL were similar to those of patients with low-count, Rai 0 CLL, although the median platelet count was slightly higher among MBL patients (Table 1).
Because CD38, ZAP-70, IgHV gene mutation status, and FISH analysis were not routinely performed during the entirety of the study period, results were not available for all patients. In MBL patients, CD38 status was available for 274 patients (91%), ZAP-70 status was available for 120 patients (40%), IgHV gene mutation status was available for 109 patients (36%), and cytogenetic analysis by FISH was available for 126 patients (42%). Results of prognostic testing are listed in Table 1. No significant differences in prognostic test parameters were observed between patients with MBL and those with low-count, Rai 0 CLL.
A follow-up lymphocyte count more than 3 months (range, 3 to 95 months) after the baseline count was available for 210 MBL patients (70%). Although a persistent absolute lymphocytosis (> 3 × 109/L) was present in 182 (95%) of 192 patients with lymphocytosis at baseline, only 66 patients (36%) experienced a doubling in lymphocyte count during follow-up.
Patients underwent sequential monitoring with a median follow-up time of 18 months (range, 0 to 97 months). For individuals with MBL, the proportion of patients free of treatment was 99% (95% CI, 98% to 100%) at 1 year, 98% (95% CI, 96% to 100%) at 2 years, and 93% (95% CI, 86% to 100%) at 5 years (Fig 1A). During the first 5 years of follow-up, the rate of progression requiring treatment for the MBL patients in our series was approximately 1.4% per year. Primary indications for treatment among MBL patients were progressive lymphadenopathy (n = 2), marrow failure/cytopenia (n = 3), progressive lymphocytosis (n = 1), and autoimmune hemolytic anemia (n = 1).
We next evaluated how total B-cell count and clonal B-cell count related to clinical outcome. Total B-cell count as a continuous variable (HR = 2.9; 95% CI, 1.1 to 8.2; P = .04) was associated with TTT. Although we have previously shown that nearly all B cells among individuals with MBL are clonal as the B-cell count increases,12 clonal B cells may represent a small component of the total B-cell count in some MBL patients with lower B-cell counts. To evaluate this aspect, we measured the clonal B-cell count among the 64 individuals with MBL whose total B-cell count was less than 1.5 × 109/L and who had treatment information available. The clonal B-cell percentage ranged from 51.9% to 99.7% (median, 93.9%) in these individuals (absolute B-cell counts of 0.015 to 1.46 × 109/L; median, 0.76 × 109/L). Despite this variability in the proportion of clonal B cells, only one (1.5%) of 64 individuals whose total B-cell count was less than 1.5 × 109/L required treatment. Similarly, no difference in survival was observed based on the percentage of clonal B cells among individuals whose total B-cell count was less than 1.5 × 109/L (P = .71). These results indicate that having a high proportion of clonal B cells does not necessarily indicate a higher risk for progression among individuals with a total B-cell count less than 1.5 × 109/L.
With respect to other demographic and prognostic parameters, CD38 status (HR = 10.8; 95% CI, 2.0 to 59.5; P = .006) was associated with TTT (Fig 2), but age (P = .62) and sex (P = .99) were not. Because of an insufficient number of patients with ZAP-70, IgHV gene mutation status, or FISH analysis, we could not accurately evaluate the correlation of these variables with TTT or OS.
Likelihood of treatment for MBL patients was substantially lower (HR = 0.32, P = .04) than that of low-count, Rai 0 CLL patients with B-cell counts between 5.0 and 10.0 × 109/L (n = 94) as well as patients with Rai stage 0 CLL who had an ALC more than 10.0 × 109/L (n = 219; P = .0003; Fig 3). To date, no difference in OS has been observed between these three patient groups (P = .45).
Because it is only a recently defined diagnostic entity,1 there is little clinical information on the natural history of individuals with MBL. We observed a low rate of progression among individuals with CLL-phenotype MBL identified in our routine clinical practice. Our study is only the third series of MBL patients to include more than 50 individuals5,9 and, to our knowledge, is the largest reported series of molecular/biologic prognostic parameters in MBL patients. Notably, the rate of progressive lymphocytosis and need for treatment in our series are similar to those reported by Rawstron et al5 in a cohort of 185 patients with clinically identified MBL who underwent sequential monitoring.
Importantly, our study is the first, to our knowledge, to compare the outcome of a large series of clinically identified MBL patients to a contemporaneous cohort of patients with newly diagnosed, Rai stage 0 CLL with comparable clinical follow-up. The MBL patients in our cohort had longer TTT than not only individuals with Rai stage 0 CLL in general, but also patients with low-count, Rai stage 0 disease as defined by having a B-cell count more than 5 × 109/L but an ALC less than 10 × 109/L. The only previous studies that compared the outcome of individuals with clinically identified MBL with that of patients with Rai stage 0 CLL were two small analyses (our own pilot study [n = 47]8 and a series from British Columbia [n = 46]6), both of which had little statistical power and neither of which found a difference in TTT between MBL and Rai stage 0 CLL. The findings of the current study provide the first clinical evidence to support the recently proposed revisions to the CLL diagnostic criteria, which recommend basing the diagnosis of CLL on a B-cell count of more than 5 × 109/L rather than an ALC of more than 5 × 109/L.
Despite the fact that a B-cell threshold of 5 × 109/L does stratify patients with different rates of progression requiring treatment, B-cell count seems to relate to TTT as a continuous variable. Additional studies are needed to determine what B-cell threshold optimally distinguishes the clinical outcome between MBL and Rai stage 0 CLL. In a separate recent study of 459 patients with an ALC more than 5 × 109/L who met the 1996 criteria for a diagnosis of CLL16 and who were referred to a hematologist, we found that a B-cell count of 11 × 109/L may be the optimal threshold for stratifying a patient's risk of treatment and death.12 Although 146 individuals in the previous report are included in the present study as a result of overlap between the 1996 CLL diagnostic criteria16 and the 2005 MBL diagnostic criteria,1 the present study includes 156 additional individuals who were not seen by a hematologist at our center and/or had an ALC less than 5 × 109/L (ie, could not be classified as CLL by either the 200810 or 199616 criteria). The results of the two studies are complementary, with both suggesting that a higher B-cell count may better stratify patients' risk of adverse clinical outcome. It is also worthy of emphasis that, in the present study, having a higher proportion of clonal B cells did not predict a shorter TTT among individuals with MBL whose total B-cell count was less than 1.5 × 109/L. This finding suggests the total B-cell count is an adequate measure of risk among individuals with MBL and that having a large proportion of clonal B cells does not increase the risk of progression among those whose total B-cell count is less than 1.5 × 109/L.
Although numerous studies have demonstrated that biologic features of CLL B cells, such as ZAP-70, CD38, IgHV gene mutation status, and cytogenetic abnormalities on FISH, are powerful predictors of TTT and OS in patients with Rai stage 0 CLL,15,17,18 there is limited information on the prognostic value of these characteristics in individuals with MBL. However, among the individuals with MBL in our series, we found that CD38 status was strongly related to TTT.
It is important to note that our study is limited by its retrospective nature, and longer follow-up will be necessary to more precisely characterize patients' long-term outcome. Additional studies are also needed to determine what biologic characteristics best predict outcome among individuals with MBL and to identify the biologic events that contribute to disease progression in the hope that such events can be prevented or reversed.
In summary, MBL seems to be a distinct clinical entity with a clinical behavior that differs from that of Rai stage 0 CLL. In this regard, the relationship between MBL and CLL may be akin to that of monoclonal gammopathy of undetermined significance and myeloma. Individuals with MBL identified in clinical practice have a low risk of requiring treatment in the next 5 years. B-cell count and CD38 status seem to predict TTT among MBL patients. Additional studies are needed to determine the optimal criteria to distinguish between MBL and CLL. Differentiating between CLL and MBL based on a patient's risk of adverse clinical outcome could reduce the psychological distress caused by classifying asymptomatic individuals at low risk for adverse clinical consequences as having CLL.
Supported by National Cancer Institute Grant No. CA 113408, Gabrielle's Angel Research Foundation, and the Mayo Clinic Cancer Center.
Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
The author(s) indicated no potential conflicts of interest.
Conception and design: Tait D. Shanafelt, Neil E. Kay, Kari G. Rabe, Timothy G. Call, Clive S. Zent, Kami Maddocks, Greg Jenkins, Susan L. Slager, Curtis A. Hanson
Financial support: Tait D. Shanafelt, Neil E. Kay
Administrative support: Tait D. Shanafelt, Curtis A. Hanson
Provision of study materials or patients: Tait D. Shanafelt, Neil E. Kay, Timothy G. Call, Clive S. Zent, Susan Schwager, Deborah Bowen, Curtis A. Hanson
Collection and assembly of data: Tait D. Shanafelt, Kari G. Rabe, Kami Maddocks, Greg Jenkins, Diane F. Jelinek, William G. Morice, Justin Boysen, Susan Schwager, Curtis A. Hanson
Data analysis and interpretation: Tait D. Shanafelt, Neil E. Kay, Kari G. Rabe, Timothy G. Call, Clive S. Zent, Kami Maddocks, Greg Jenkins, Diane F. Jelinek, William G. Morice, Deborah Bowen, Susan L. Slager, Curtis A. Hanson
Manuscript writing: Tait D. Shanafelt
Final approval of manuscript: Tait D. Shanafelt, Neil E. Kay, Kari G. Rabe, Timothy G. Call, Clive S. Zent, Kami Maddocks, Greg Jenkins, Diane F. Jelinek, William G. Morice, Justin Boysen, Deborah Bowen, Susan L. Slager, Curtis A. Hanson