The trusted source for
healthcare information and
abstract & commentary
Source: Lopez-Guillermo A, et al. Blood 1999;93:3081-3087.
In a number of malignancies characterized by chromosomal translocations, heterogeneity in the molecular features of the translocation has been examined for its influence on prognosis. For example, in patients with the t(9;22) translocation and chronic myeloid leukemia, the translocation involves the juxtaposition of the abl gene on chromosome 9 with the bcr gene on chromosome 22. However, the site of the break on chromosome 22 is variable. The vast majority of breaks occur within a 5.8 Kb long region that has been called the major breakpoint cluster region (Mbcr). When the break occurs toward the 3’ end of the Mbcr, the chimeric gene created contains exon 3 of bcr (the gene product is called a2b3 to signify the joining of the second abl exon with the third exon of bcr). When the break occurs more 5’, exon 3 of bcr is deleted and the a2 exon is fused to the b2 exon creating a2b2. A couple of reports1 suggested that patients with a 3’ break might have a shorter chronic phase than those with a 5’ break. However, most investigators have not found any clinical correlation between the chromosome 22 breakpoint and prognosis.2-4
Similarly, the t(14;18) translocation in follicular lymphoma is characterized by heterogeneity in the chromosome 18 breakpoint. About two-thirds of the rearrangements occur at the major breakpoint region (MBR) located in the 3' untranslated portion of the last exon, about 10% occur in the minor cluster region (mcr) located about 30 Kb downstream of the bcl-2 gene, and in the remaining cases (about 15%), no translocation is detected by any technique. The efforts to relate the molecular anatomy of the translocation to prognosis have generally failed to reach firm conclusions because of heterogeneous histologic diagnoses (some papers included cases of diffuse large B-cell lymphoma of follicular center origin), heterogeneous treatments, small numbers of cases ( < 100), and failure to take into account the distribution of other possibly independent prognostic factors such as performance status, lactate dehydrogenase levels, and others.
Lopez-Guillermo and colleagues at M.D. Anderson Cancer Center have analyzed breakpoints in 247 patients with follicular lymphoma who were treated mainly with anthracycline-based combination chemotherapy. It is difficult to be certain, but these patients appear to have been treated mainly before the use of fludarabine in follicular lymphoma. The breakpoints occurred in the MBR in 175 patients (71%), in the mcr in 27 patients (11%), and no rearrangement was detected in 45 patients (18%). No significant differences were noted in the expression of bcl-2 or bax proteins. No significant differences were seen in the clinical characteristics. However, patients with either type of translocation had a significantly higher complete response rate (mcr 96%, MBR 90%) than did patients with germline tumors (71%). Failure-free survival at three years was also different among the groups: mcr tumors had the best outcome, 95% were failure-free at three years; for MBR, 76%; and for germline, 57%. When patients were stratified on the basis of lactate dehydrogenase levels and beta-2-microglobulin levels, the number of patients became much smaller, but the relationships among the three groups were retained and a Wilcoxon test gave a P-value of 0.08, implying that the type of rearrangement was largely independent of these other risk factors.
On the basis of this report, Lopez-Guillermo et al would order the prognosis of patients with follicular lymphoma based upon the nature of the translocation as follows: mcr > MBR > germline. Almost no one with mcr fails to achieve a complete response or relapses, but this form is rare. Those with MBR comprise the majority of cases and they follow the typical follicular lymphoma relapse curve with a steady rate of relapse without hint of a plateau. Those without rearrangements respond poorly to therapy, and their relapse curve resembles that associated with aggressive histology lymphoma. The relapses that occur usually happen within the first year and those in remission at three years after treatment appear less likely to relapse.
It is difficult to know how to interpret these data. While intriguing, it is not obvious why the subsets should be different. They all overexpress bcl-2 protein, and it is clear that bcl-2 overexpression is not sufficient to cause this neoplasm. It is possible that the different subsets are associated with a distinct pattern of mutations in other genes. But such an analysis has not yet been undertaken. If the subsets do differ in the spectrum of genes that are mutated, it would be difficult to attribute the difference to the bcl-2.
However, this level of ignorance is not the most pressing matter. What is most important is to define whether these observations are true; because if they are true, then no mcr patient should have treatment withheld, and those with MBR and germline genetics should be stratified and balanced in any future study as their prognosis is different. It would be valuable to try and ascertain the reproducibility of these findings in a group of patients like those who have been treated conservatively at Stanford over the years. Is the natural history of the subsets really distinct? Would patients who have been watched without treatment for 10 years or more disproportionately be those with mcr rearrangements?
It appears that therapy for follicular lymphoma is improving. The fludarabine-based combinations appear to be capable of inducing molecular-complete remissions in the majority of treated patients, and such remissions appear to be more long-lasting than remissions from other types of chemotherapy. Furthermore, the introduction of rituximab promises to further enhance the efficacy of the treatment in that the antibody may sensitize cells to the chemotherapy.5 It would be important to know whether there are molecular markers that could distinguish subsets of patients who do well, and those for whom a different approach would be indicated. Molecular characterization of the bcl-2 breakpoint is an intriguing lead that needs to be followed up, but support for its role has not yet reached the level of evidence for practice to be modified on its basis.
1. Mills KI, et al. Leukemia 1989;3:837-840.
2. Jaubert J, et al. Br J Haematol 1990;74:30-35.
3. Tefferi A, et al. Leukemia 1990;4:839-842.
4. Opalka B, et al. Leukemia 1991;5:452-456.
5. Czuczman MS, et al. J Clin Oncol 1999;17:268-276.
a. Those with breakpoints in the minor cluster region (mcr) appear to have the best prognosis.
b. Those with breakpoints in the mcr appear to have the worst prognosis.
c. Those with breakpoints in the major breakpoint cluster (MBR) appear to have the best prognosis.
d. Those with breakpoints in the MBR appear to have the worst prognosis.
e. Those with germline bcl-2 appear to have the best prognosis.