|Values are valid only on day of printing.|
Published: October 2008Print Record of Viewing
Dr. Hanson will be discussing recent changes in the detection and diagnosis of chronic lymphocytic leukemia (CLL). This is the first of two presentations. Part 2 reviews the laboratory’s role in risk stratification for CLL.
Presenter: Curtis A. Hanson, MD
Welcome to Mayo Medical Laboratories' Hot Topics. These presentations provide short discussions of current topics and may be helpful to you in your practice.
Our speaker for this program is Dr. Curtis Hanson, from the Division of Hematopathology at Mayo Clinic. Dr. Hanson will be discussing recent changes in the detection and diagnosis of chronic lymphocytic leukemia (CLL). This is the first of two presentations. Part 2 reviews the laboratory’s role in risk stratification for CLL.
Our goals today will be to distinguish CLL from other B-cell chronic lymphoproliferative disorders, to understand the definition and the use of the term monoclonal B-cell lymphocytosis or MBL, and to realize the issues associated with minimal residual disease or MRD detection in CLL.
As a side note I will be presenting a future Hot Topics on CLL entitled “Risk stratification in CLL – The Role of the Clinical Laboratory.”
To start off with I want to point out that there are three terms that have been used to describe an overlapping group of diseases. CLL, MBL and also Small Lymphocytic Lymphoma (SLL). All three of these have many common features and are closely related from a biological perspective. We know that these diseases have a similar morphologic appearance in peripheral blood, bone marrow and lymph nodes. We know that they have a similar immunophenotype and that they have common and similar genetic factors that can be detected. The separation of these entities is rather arbitrary and is in part related to patients’ sensitivities and how the term leukemia is applied to them. Regardless we must be sure to remember that these are not three separate and unrelated processes.
The diagnostic criteria of CLL have evolved over the last 30 years. In 1975, Rai staging for CLL emerged which also defined CLL as those cases having greater than a 15,000 absolute lymphocyte count, or ALC, in the peripheral blood. By the late 1980’s and reaffirmed in 1996 the NCI working group redefined CLL as those cases having a greater than 5,000 ALC and pointing out the necessity of having a precise immunophenotype.
Just recently the International working group of CLL, or the IWCLL, published new criteria which now defines CLL as being those cases having greater than 5,000 B-cells of at least 3 months’ duration. Please note the change from a lymphocyte to a B-cell count.
Flow cytometry remains important in determining both clonality as well as the CLL immunophenotype. In addition the presence of cytopenias caused by a typical marrow infiltrate defines CLL regardless of the number of B-cells or the amount of nodal involvement.
This table highlights the prototypic immunophenotypes that are found in the various B-cell lymphoproliferative disorders. Now I don’t want to go through this entire table with you but I do want to point out two things.
First of all, in CLL and SLL the classic immunophenotype is that of dim surface immunoglobulin, dim CD20, and expression of CD5 and CD23. Although this precise phenotype is not found in classic cases of other lymphoproliferative disorders, please note that similar phenotypes can be identified in some cases of lymphoplasmacytic leukemia and occasional cases of Splenic marginal zone lymphoma. As well, you should note that exceptions to the prototypic CLL phenotype can be found in some cases of CLL.
This slide demonstrates the prototypic flow sedimentric histograms that one finds from patients with CLL demonstrating the dim surface kappa immunoglobulin, dim CD20, and coexpression of CD23 and CD5 with CD20.
This slide points out the caution with which one needs to interpret flow findings. In this case of a mantle cell lymphoma, one clearly sees the bright CD20 expression with dual expression of CD5. It might also be easy for one to assume that this case is CD23 positive but note that the expression of CD23 is what I would call partial and is not characteristic of what one sees in a typical case of CLL but indeed is actually a common finding in mantel cell lymphoma.
This slide demonstrates a case of lymphoplasmacytic lymphoma that has an immunophenotype that is closer to that of a mantel cell lymphoma but it does have dim CD20 and dual expression of CD5 and CD20.
In addition to redefining CLL the IWCLL also redefined the diagnosis of MBL as well as SLL. MBL is now recognized as a disease entity and is defined as those cases that have a B-cell count of less than 5,000 and having the absence of lymphadenopathy or organomegaly which would be characteristic of SLL. It is important to note that there are similar expressions of genetic risk factors in MBL cases as compared to early stage CLL. SLL is defined as those cases with lymphadenopathy or organomegaly and having a B-cell count of less than 5,000.
This bar graph highlights the key numeric cup points that one might encounter in the clinical laboratory and how they fit into the new IWCLL diagnostic criteria. An ALC count of 3 times 10 to the ninth per liter, or 3,000, is the upper limit of normal for a peripheral blood lymphocyte count. The 5 times 10 to the ninth per liter or 5,000 ALC mark indicates the previous cup point defining CLL while the 5 times 10 to the ninth per liter, or 5,000, B-cell count defines the new IWCLL cut off for diagnosing CLL versus MBL. Please note that from left to right one would encounter those cases that would be called MBL with no increase in lymphocytes. Those cases with an ALC between 3,000 and 5,000 would be MBL associated with a lymphocytosis, and those between a 5,000 ALC and a 5,000 B-cell count would represent those cases previously called CLL which would now be reclassified as MBL.
One should also note that SLL may be diagnosed in any case with less than 5,000 B-cells depending on the presence and the prominence of lymphadenopathy and organomegaly.
This slide shows the same bar graph with three areas highlighted. These highlighted areas really point out where the questions will arise within the clinical laboratory as you attempt to distinguish between CLL, MBL, and SLL.
The identification of MBL can really occur within three settings. First of all, one may find MBL as part of the general screening of the population, and we’ll talk about that a little bit more in a later slide. We will also talk in a later slide about the concept of familial CLL. From a practical perspective, however, we need to be aware of when we will be making a diagnosis of MBL within the context of routine clinical practice.
In clinical practice, MBL will most commonly be identified when a lymphocytosis is found via CBC screening. I frequently hear questions as to what limits actually define a lymphocytosis. I feel very strongly that a lymphocytosis is defined as those cases having an ALC above3, 000. I think that this has been confirmed by various laboratories who have done thorough normal value studies.
In addition, if one looks at the quantitative lymphocyte subsets that are broadly published, the sum of T-cells, B-cells, and NK-cells would certainly suggest that a lymphocyte value of 3,000 is close to being accurate. The historic number of 5,000 being the upper limit for an ALC should not be used in today’s clinical environment.
In addition to MBL being recognized within the context of a lymphocytosis evaluation, MBL will also be identified as an incidental finding when a flow analysis of peripheral blood or bone marrow are being done for unrelated reasons when a bone marrow biopsy with a lymphoid infiltrate, followed by a peripheral blood flow study in a case without lymphocytosis, or a small lymphocytic lymphoma is identified at the time of surgery without associated adenopathy, or peripheral blood lymphocytosis.
Thus if we go back to our bar graph again, one can see the left hand circle of MBL without an increase in lymphocytes being recognized in those cases identified through population screening or as an incidental finding within the clinical practice.
In the right hand circle being those cases identified following a lymphocytosis evaluation.
Various studies have now shown that MBL is a common disorder with up to 3.5% of the normal population having an identifiable clonal population with the CLL phenotype. It is also important to note that the prevalence of MBL appears to increase with age, going from a 2.1% incident rate in people between 40 and 60 years old up to an incident rate of almost 8% in those folks over the age of 70. Low-risk genetic factors such as mutated IgVH and 13q- are frequently identified. The progression rate of MBL to CLL is uncertain and remains to be defined. Studies by Rostern and colleagues have shown that 1-3% of MBLs defined via population studies will progress to CLL, whereas studies done here at the Mayo Clinic have shown that 40% of MBLs identified through routine clinical practice, i.e., identification and evaluation of cases with lymphocytosis will progress to CLL within one year.
To summarize, MBL may be identified through population screening or through routine clinical practice. These latter cases likely will exhibit a different behavior than those cases that are identified through population screening. It is important to note those differences when evaluating and reporting these cases.
The impact of this diagnostic criteria change is not small. We have shown that approximately 40% of what previously would have been called Rai Stage 0 CLL will now be reclassified as MBL.
In addition, no standard methodology for measuring B-cell counts in MBL and CLL was established with the new IWCLL definitions. This lack of standardization will undoubtedly lead to questions and confusion in individual patient cases.
Although we will talk about prognostic factors a future Hot Topic session, molecular prognostic factors will likely contribute to the risk of disease progression better than any arbitrary lymphocyte or B-cell count.
The concept of familial CLL has recently emerged. It is now recognized that families with known CLL patients have an increased risk of having MBL or CLL identified within their own kindred. Anywhere from 12%-18% of CLL patients will have an extended family member with CLL or some other type of lymphoproliferative disorder. The precise genetic factors within familial CLL remain undefined and are uncertain.
I do want to share with you a case of MBL. This was a 66 year old female who presented with a normal CBC and a normal lymphocyte count. The blood smear was normal, and no adenopathy or organomegaly was found in the physical exam. A flow study was performed on the peripheral blood.
After reviewing the chart, I still have no idea why the flow study was originally ordered. After our findings, the patient was evaluated in clinical hematology and a bone marrow study performed.
This slide shows the immunophenotype of a small monoclonal kappa B-cell population identified in this patient that accounted for approximately ½ of 1% of the total peripheral blood count.
The bone marrow on the left hand side stained with Pax-5, a pan-B-cell antibody, shows a slightly increased number of scattered B-cells. On the right hand side, one can identify a rare small aggregate of Pax-5-positive B-cells. This case exemplifies a minimal disease burden that one can find with an early MBL.
It should be noted that this patient, after a couple years of follow-up, has shown no progression of disease.
This cartoon outlines the sequence of events that one follows in reaching a diagnosis in these patients. First of all in the peripheral blood, the CBC and differential count, together with the immunophenotype, are the principal components needed to reach a diagnostic decision. Occasionally, a bone marrow specimen may be helpful but is certainly not required in examination of these patients. Once a diagnosis is reached, classification to CLL, MBL, or other B-cell lymphoproliferative disorders may then be accomplished.
I want to finish this session by briefly talking about CLL and MRD. MRD eradication is goal of current therapies. But questions certainly exist:
Does absence of MRD improve overall survival? Does MRD detection predict early relapse?
These questions and others have not been definitively answered. But without a doubt, laboratories are now being asked to detect MRD in CLL patients.
More questions regarding CLL and MRD.
Which specimen type should flow immunophenotyping studies be done on, peripheral blood or bone marrow? I think the consensus in the community now is that blood is the preferred specimen as it is the easiest specimen to obtain. No studies have shown consistent and significant differences between blood and bone marrow.
To what detection level is MRD required in these cases? I would say the current expected level of MRD detection required is to 0.01%. To get to that level, we need to probably collect anywhere from 200,000 to 500,000 events by flow cytometry to be able to have enough B-cells to be able to analyze and interpret it with confidence.
Different Immunophenotypic approaches have been used including just simply CD5 versus CD19 versus 4-color vs. 6-color? Multicolor adds specificity, but not necessarily sensitivity. Sensitivity is dependent on the cell mix and how many polyclonal B-cells are present. If there are very few polyclonal B-cells, a simple CD5 versus CD19 will work as several studies in the literature have shown. Challenges, however, arise when there is a mixture of monoclonal and polyclonal B-cells. In those situations, multi-color flow clearly provides diagnostic advantage.
Does immunohistochemistry (IHC) have a role in bone marrow specimens? I would say that stains are often complementary to flow studies, but they are often hard to interpret in isolation away from flow studies. One also needs to be aware that T-cell nodules depleted of B-cells may be identified in patients who have been treated with Rituxan and clearly can be confused with CLL on a strict morphology basis.
What antibodies should be used for IHC? There is no specific and easy answer. A variety of B-cell antibodies have been used, I would say that CD20 however has a minimal role because of the Rituxan effect; T-cell markers are also frequently done. However, CD5 may be hard to interpret in these cases.
I want to close my talk by sharing a brief case of a CLL with minimal residual disease. This was a 58 year-old lady who approximately 9 years ago was diagnosed with CLL on the basis of her peripheral blood lymphocytosis and a diagnostic flow study. No organomegaly and no cytopenias were present. She was called a Rai Stage 0 CLL and was not treated.
Over the course of the next nine years, she showed steady progression of disease, and in early 2007 went on chemotherapy with the standard Pentostatin, Cytoxan, and Rituxan combination chemotherapy
By January 2008 she really had minimal clinical disease, with only a single anterior node palpable, and as you can see here her lymphocyte count was normal, and the rest of her CBC counts appeared okay.
By June 2008, there is no radiologic evidence of disease and her CBC remained normal.
A bone marrow was done and the bone marrow biopsy looked normal cellular with normal hematopoiesis.
On special stains, a single very small lymphoid aggregate composed almost exclusively of PAX-5 positive B-cells was identified. No CB-3 positive T-cells were identified in this aggregate.
If we look at the flow studies done on the peripheral blood, we can see that approximately 200,000 events were collected, and a small monoclonal kappa population was identified that accounted for 0.02% of the population.
Bone marrow flow showed a similar finding, with approximately 0.4% of the cells demonstrating a monoclonal kappa pattern of expression, thus there was a marked reduction of tumor burden in this patient, but a small amount of CLL was still detectable.
Our goals today were to distinguish Chronic Lymphocytic Leukemia (CLL) from other B-cell chronic lymphoproliferative disorders, to understand the definition of the term monoclonal B-cell lymphocytosis (MBL), and to realize the issues associated with minimal residual disease (MRD) detection in CLL. Hopefully we have been able to achieve these objectives with you. I look forward to discussing risk stratification in CLL and the role of the clinical laboratory in a future Hot Topic with you.