Chronic Lymphocytic Leukemia: Newly Recognized Molecular and Clinical Heterogenity
Chronic Lymphocytic Leukemia: Newly Recognized Molecular and Clinical Heterogenity
Abstracts & Commentary
Sources: Damle RN, et al. Blood 1999;94:1840-1847; Hamblin TJ, et al. Blood 1999;94:1848-1854.
Chronic lymphocytic leukemia (cll) is the most common leukemia in Western nations. About 14,000 cases are diagnosed each year in the United States. The disease usually has an indolent course until tumor cells replace and crowd out normal marrow elements. About 25% of patients develop some autoimmune disease in the course of their CLL and about half the deaths in patients with CLL are related to infection, to which patients have a heightened sensitivity because of a hypogammaglobulinemia as well as a decrease in host response to pathogens.
The malignant cell in CLL is a small lymphocyte with no obvious differences from normal lymphocytes, except perhaps a tendency to break up and smudge under the force generated by making a peripheral blood smear. The cell expresses surface immunoglobulin and other B-cell markers. Its most characteristic feature is the expression of CD5, a marker that is usually considered a T-cell marker. CD5+ B cells usually reside in the follicular mantle of lymph nodes and correspond to the B1 subset of cells in mice, which usually respond to foreign antigens without the help of T cells, do not generate immunologic memory, and do not carry mutations in their immunoglobulin genes (which means they have not passed through a follicle center to undergo immunoglobulin gene mutation). Most antibodies that recognize self antigens are generated by CD5+ B cells.
As laboratories have pursued the characterization of CLL cells in more detail, a certain heterogeneity has emerged that complicates the picture. However, reports from two leading groups studying CLL have used the heterogeneity to define at least two subsets of patients with CLL with significant differences in their natural history.
One group, Damle and colleagues, composed of researchers from North Shore Hospital (associated with New York University), Long Island Jewish Hospital (associated with Albert Einstein Medical College), and the University of Genoa, noted that patients differed in the degree of expression of CD38 (a cell surface marker) and the presence of mutations in their immunoglobulin genes. CD38 is a protein that has enzymatic activity; it converts nicotinamide adenine dinucleotide to cyclic adenosine diphosphoribose, a compound that binds to ryanodine receptors and regulates calcium flux.1 In mice that have the CD38 gene knocked out, antibody responses to T-dependent antigens are greatly reduced.2 CD38 is normally expressed on B cells that occupy the germinal center, where B cells go to mutate their immunoglobulin and generate antibody that has higher affinity for an antigen. Damle et al noted that about half of their 47 patients with CLL had tumor cells expressing immunoglobulin molecules that closely resembled the germline sequence and were not mutated; nearly all of these patients had CD38 expression on more than 30% of the tumor cells. By contrast, about half of the cases had tumor cells expressing immunoglobulin molecules that had undergone somatic mutation, probably in a germinal center; all of these patients expressed either no CD38 or only low levels of CD38. Cases with and without mutations were similar in all other surface markers, such as CD23, CD5, and CD19. Only CD38 expression was different between the groups.
When patient characteristics were examined, significant differences were found. In cases with mutated immunoglobulin genes and low CD38 expression, men and women were equally represented. By contrast, in cases with cells expressing germline immunoglobulin genes and high levels of CD38, males predominated. Differences were also found in response to therapy; those with unmutated immunoglobulin molecules and high CD38 expression responded poorly to chemotherapy whereas those with mutated immunoglobulin molecules and low CD38 expression were responsive to therapy. Most interestingly, patients differed in natural history based upon the status of their immunoglobulin genes. CD38-high unmutated immunoglobulin gene cases had a poorer median survival (9 years) than CD38-low mutated immunoglobulin gene cases (median survival not reached but exceeding 18 years). The Rai stage distribution is difficult to assess given the small number of patients. However, for those with intermediate Rai stage on clinical grounds, the two subtypes maintained significant survival differences—nine years for unmutated, and 17 years for mutated.
Hamblin and colleagues from the Royal Bournemouth Hospital and the Tenovus Research Laboratory performed a similar analysis of immunoglobulin gene mutations in CLL. They had previously shown that cases bearing unmutated immunoglobulin genes were more likely to harbor the trisomy 12 cytogenetic abnormality and have a poorer prognosis while those bearing mutated immunoglobulin genes more often had chromosome 13q 14 abnormalities and a better prognosis.3 Hamblin et al analyzed 84 cases; around 45% had unmutated immunoglobulin genes and 55% had mutated immunoglobulin genes. Median survival for Binet stage A patients with unmutated immunoglobulin genes was about eight years; for those with mutated immunoglobulin genes, median survival was more than 24 years.
COMMENTARY
The conclusion that emerges from these seminal studies is that the clinical disease we call CLL is really at least two diseases. One disease is derived from a memory B cell that has passed through a germinal center, mutated its immunoglobulin gene in response to an antigen, and been transformed. This disease is associated with interstitial deletions in the retinoblastoma gene product, affects men and women equally, and has an extremely indolent natural history. When treatment is needed, it responds well to therapy. The other disease is derived from an antigen-naive B cell from the marginal zone of lymph nodes, has never passed through a germinal center, has not mutated its immunoglobulin genes, and is probably related to mantle zone lymphoma. This disease is associated with trisomy 12, affects men predominantly, and has a more aggressive natural history (median survival 8 years). This disease is more refractory to chemotherapy.
While no one is anticipating a sudden expansion of molecular immunology labs capable of analyzing tumor-associated immunoglobulin genes for the presence of somatic mutations as a component of a routine diagnostic clinical pathology laboratory, the Damle et al study suggests that a surrogate marker for immunoglobulin gene mutation is CD38 expression. CD38 expression is routinely measureable using a standard flow cytometer. Thus, one imagines that a CLL panel will emerge in the diagnostic laboratory in which cells are labeled with anti-IgM or anti-CD19 to identify them as B cells, with anti-CD5 to identify them as CLL B cells, and with anti-CD38 to separate the good (CD38 low) from the bad (CD38 high). It is possible that once this separation of subsets has occurred, other markers that have been reported to assess prognosis in CLL such as beta 2-microglobulin levels, soluble CD23 levels, or soluble deoxythymidine kinase levels would add further clinical information.4-6 However, the question of the value of other markers needs to be addressed in a large study in which sufficient numbers of patients are analyzed and the two major subcategories of CLL are separated.
In my view, both studies overinterpret the data when they try to conclude that these differences make some sort of sense because the disease that gives rise to CLL with unmutated immunoglobulin genes is derived from a more primitive B cell than the disease that gives rise to CLL with mutated immunoglobulin genes. Disease natural history is determined by many factors and the level of differentiation of the normal cell that is transformed is one of the least reliable predictors of natural history. The most clinically aggressive lymphoid malignancy is Burkitt’s lymphoma, which is a neoplasm of mature germinal center B cells, not a tumor derived from the most primitive lymphoid precursor. It is difficult for me to understand how this old canard keeps surviving peer review. In general, one can learn little about the natural history of a lymphoid neoplasm on the basis of knowing where in normal lymphoid differentiation the tumor arose.
What’s next for CLL? Certainly these papers mark a point of departure in the history of the study of this disease. It is hoped that once the two newly defined subsets are recognized prospectively, it may be possible that study of the more aggressive, treatment-refractory form will yield clues to its management. For the favorable prognosis subset (CD38 low), we do not have curative therapy, but it isn’t completely clear that we need it. Median survivals in excess of 24 years in a disease that mainly affects people older than the age of 60 do not lead one to prioritize its treatment as high as say breast, colon, lung, and prostate cancer.
References
1. Adebanjo OA, et al. Nat Cell Biol 1999;1:409-414.
2. Cockayne DA, et al. Blood 1998;92:1324-1333.
3. Oscier DG, et al. Blood 1997;89:4153-4159.
4. DiGiovanni S, et al. Acta Haematol 1989;81:181-185.
5. Sarfati M, et al. Blood 1996;88:4259-4264.
6. Hallek M, et al. Blood 1999;93:1732-1738.
Chronic lymphocytic leukemia now appears to be at least two distinct diseases based on what features?
a. Clinical stage and serum-soluble CD23 levels
b. Clinical stage and serum beta-2-microglobulin levels
c. CD38 and CD5 expression on the tumor cells
d. CD38 expression and immunoglobulin gene mutations in tumor cells
e. CD5 expression and trisomy 12
Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.