Fax: (507) 266-4972
Respective clustering of unfavorable and favorable cytogenetic clones in myelofibrosis with myeloid metaplasia with homozygosity for JAK2V617F and response to erythropoietin therapy
Version of Record online: 10 MAR 2006
Copyright © 2006 American Cancer Society
Volume 106, Issue 8, pages 1739–1743, 15 April 2006
How to Cite
Tefferi, A., Strand, J. J., Lasho, T. L., Elliott, M. A., Li, C.-Y., Mesa, R. A. and Dewald, G. W. (2006), Respective clustering of unfavorable and favorable cytogenetic clones in myelofibrosis with myeloid metaplasia with homozygosity for JAK2V617F and response to erythropoietin therapy. Cancer, 106: 1739–1743. doi: 10.1002/cncr.21787
- Issue online: 4 APR 2006
- Version of Record online: 10 MAR 2006
- Manuscript Accepted: 19 OCT 2005
- Manuscript Revised: 28 SEP 2005
- Manuscript Received: 6 SEP 2005
- JAK2V617F mutation analysis;
- myeloid metaplasia;
- treatment response;
- tyrosine kinase mutation
Patients who have myelofibrosis with myeloid metaplasia (MMM) display recurrent, albeit nonspecific cytogenetic abnormalities that are diverse prognostically. For the current study, the authors explored the relation between specific cytogenetic clones and JAK2V617F mutational status in patients with MMM and the effects on treatment response to erythropoietin (Epo).
Concomitantly collected blood granulocytes and bone marrow were processed for JAK2V617F mutation analysis and cytogenetic studies, respectively. Genomic DNA was amplified by polymerase chain reaction, and fluorescent dye chemistry sequencing was performed by using the same primers that were used for amplification.
Among 105 study patients, cytogenetic abnormalities were detected in 47 patients (45%), and the JAK2V617F mutation was detected in 52 patients (50%). Comparison of mutational frequencies between favorable (normal, sole 13q−, or 20q− clones; n = 70 patients) and unfavorable (all other abnormalities; n = 35 patients) cytogenetic categories revealed a significantly different incidence of homozygous JAK2V617F between them (9% vs. 23%, respectively; P = .04). Furthermore, the mutant allele coexisted with several recurrent cytogenetic lesions. Among 25 patients who received Epo either alone (n = 17 patients) or in combination with hydroxyurea (n = 8 patients), 4 patients (16%) achieved a response, and none of them were homozygous for JAK2V617F. Conversely, a response was more likely (P = .0001) in the presence of favorable cytogenetic abnormalities (i.e., 3 of 4 responders carried sole 13q− or 20q− clones).
Unfavorable and favorable cytogenetic clones in MMM clustered with homozygosity for JAK2V617F and treatment response to Epo-based therapy, respectively. Cancer 2006. © 2006 American Cancer Society.
Myelofibrosis with myeloid metaplasia (MMM) is a clinicopathologic entity that is characterized by stem cell-derived clonal myeloproliferation, progressive anemia, hepatosplenomegaly from extramedullary hematopoiesis, and bone marrow histologic changes that include collagen fibrosis, osteosclerosis, and angiogenesis.1 Despite a flurry of recent activity,2–6 the molecular pathogenesis of MMM remains obscure. The recent description of a myeloid lineage-specific JAK2 tyrosine kinase mutation (JAK2V617F) in some (but not all) patients with MMM has suggested heterogeneity in the molecular basis of the disease that also may be apparent at the cytogenetic level.7 In this regard, it is noteworthy that specific cytogenetic anomalies in MMM have been associated with significantly different survival outcomes.8–10
The objective of the current study was to determine whether specific cytogenetic lesions in MMM are distributed differently among distinct JAK2V617F mutation categories. In addition, we tested the hypothesis that patients who have myelofibrosis with myeloid metaplasia (MMM) who carry homozygous JAK2V617F are at peak JAK-STAT signaling and, thus, are unlikely to experience further erythropoietic drive from exogenous erythropoietin (Epo) administration by reviewing a subset of the study population that received first-line therapy with Epo alone or in combination with hydroxyurea.
MATERIALS AND METHODS
After approval by the Mayo Clinic Institutional Review board, the current study population was identified based on availability of bone marrow cytogenetic studies and concomitantly collected peripheral blood neutrophils for JAK2V617F mutation analysis. Diagnosis of MMM was according to World Health Organization criteria.11 The diagnosis of either postpolycythemic myeloid metaplasia (PPMM) or postthrombocythemic myeloid metaplasia (PTMM) required fulfillment of diagnostic criteria for MMM as well as documentation of an antecedent polycythemia vera (PV) or essential thrombocytopenia (ET) supported by bone marrow examination. Mutation analysis for JAK2V617F was performed in DNA derived from peripheral blood granulocytes. Genomic DNA was amplified by polymerase chain reaction analysis, and fluorescent dye chemistry sequencing was performed by using the same primers that were used for amplification.12 Based on previously published retrospective and prospective studies,8, 10 cytogenetic findings were considered to be either favorable (normal, sole 13q− or 20q− abnormalities) or unfavorable (all other abnormalities). For the purposes of the current study, an anemia response constituted either an increase >2 g/dL in hemoglobin level or the patient becoming red blood cell transfusion-independent. Comparisons between categoric variables were performed by calculating chi-square statistics. Comparisons between categoric and continuous variables were performed by using either the Mann–Whitney U test or the Kruskal–Wallis test. All data were analyzed by using SAS software (SAS Inc., Cary, NC).
Simultaneously collected bone marrow and peripheral blood granulocytes from patients with MMM were processed for cytogenetic studies and JAK2V617F mutation analysis, respectively. The current study was based on 105 consecutive patients (median age, 60 years; age range, 30-81 years; 73 males) who had ≥ 20 metaphases examined successfully for karyotypic abnormalities. Seventy-nine patients had agnogenic myeloid metaplasia (AMM), 16 patients had PPMM, and 10 patients had PTMM. Baseline Dupriez prognostic scores distribution for the study cohort were low-risk in 54% of patients, intermediate-risk in 37% of patients, and high-risk in 9% of patients.13 The median disease duration prior to mutation screening was 40 months (range, 0-358 months) with no significant difference among the 3 MMM variants (P = .99).
Cytogenetic abnormalities were detected in 47 patients (45%), and the JAK2V617F mutation was detected in 52 patients (50%; homozygous in 13%). The incidence of abnormal or unfavorable cytogenetic findings was similar among the 3 MMM variants; 43% and 30% in AMM, 50% and 44% in PPMM, and 50% and 40% in PTMM (P = .83 and P = .52, respectively). In contrast, both overall and homozygous JAK2V617F mutational frequency was significantly higher in PPMM (89%; homozygous in 38%) compared with either AMM (46%; homozygous in 8%) or PTMM (20%; homozygous in both instances; P = .0004). This particular observation has been reported previously.7
Overall mutational frequencies were similar between patients with normal (48%) and abnormal (51%) cytogenetic findings. However, homozygous JAK2V617F mutations occurred more frequently in the presence of unfavorable cytogenetic abnormalities (23% vs. 9%; P = .04). Although it was not statistically significant, the same trend was seen when the analysis was restricted to the 79 patients who had AMM (13% vs. 5%). Furthermore, among the 8 homozygous results seen in patients who had unfavorable cytogenetic profiles (n = 35 patients), 7 occurred among the 21 patients who had all metaphases affected by the specific cytogenetic lesion compared with only 1 among 14 patients who had a mixture of normal and abnormal metaphases (P = .04). Table 1 outlines specific cytogenetic lesions with their corresponding JAK2V617F mutational status and reveals the lack of mutual exclusivity between the JAK2V617F mutation and most MMM-characteristic cytogenetic abnormalities.
|Cytogenetic Status||No.||Homozygous||Heterozygous||Wild Type||Epo Therapy (n = 25) and Response|
|Normal||58||5||23||30||N = 13 (1 responded and was heterozygous)|
|del(20)(q11q13)||7||1||3||3||N = 2 (both responded, 1 was heterozygous, and the was other wild type)|
|del(11)(q13q23)||2||1||0||1||N = 1 (no response)|
|der(6)t(1;6)(q21;p21)||1||1||0||0||N = 1 (no response)|
|T(11;14)(q23;q13), del(13)(q12q14)||1||1||0||0||N = 1 (no response(|
|del(13)(q12q14)||3||0||1||2||N = 1 (was heterozygous and responded)|
|der(9)t(1;9)(q21;q21.2), +9||1||0||1||0||N = 1 (no response)|
|+8||1||0||1||0||N = 1 (no response)|
|+8, +9||1||0||1||0||N = 1 (no response)|
|t(1;7)(q10;p10)||1||0||0||1||N = 1 (no response)|
|t(8;9)(p21;p22), −7, +mar||1||0||0||1|
|der(9)t(9;12)(q13;q13)||1||0||0||1||N = 1 (no response)|
|15q−||1||0||0||1||N = 1 (no response)|
Information on first-line Epo therapy was available in 25 of the 105 study patients. These 25 patients (age range, 39-74 years; 16 males) received Epo either alone (n = 17 patients) or in combination with hydroxyurea (n = 8 patients). Four patients (16%) achieved a response, and none of them were homozygous for JAK2V617F (3 heterozygous and 1 wild type). Conversely, response to Epo-based therapy was more likely (P = .0001) in the presence of favorable cytogenetic abnormalities: All 3 patients who had favorable cytogenetic abnormalities (either 13q− or 20q− clones) responded, whereas none of the 9 patients who had unfavorable cytogenetic clones and only 1 of the 13 patients who had normal cytogenetics responded. Concomitantly measured serum Epo levels were available for review in 8 of the 25 patients who received Epo-based therapy, including in 2 of the 4 anemia responders. The serum Epo levels in the 2 patients who responded to Epo therapy (46 mU/mL and 84 mU/mL) did not differ significantly from levels in the 6 patients who did not respond (10 mU/mL, 32 mU/mL, 41 mU/mL, 51 mU/mL, 98 mU/mL, and 120 mU/mL; P = .74). The patient who had the lowest Epo level was homozygous for JAK2V617F and did not respond to Epo-based therapy.
The current results demonstrated a significant association between homozygous JAK2V617F and unfavorable cytogenetic profile in MMM, each of which has been associated previously with either a time-acquired phenomenon14 or inferior survival,8, 10 respectively. Accordingly, the observed clustering between the 2 markers may reflect the underlying stage of clonal evolution rather than a direct correlation. This contention is supported further by the demonstrated difference in the prevalence of JAK2V617F homozygosity among patients who had unfavorable cytogenetic profiles, depending on whether or not all metaphase chromosomes were affected. An alternative explanation considers the induction of genetic instability as a result of JAK2V617F-mediated cell proliferation, which is unrestricted in the absence of the wild-type allele and may play a dominant negative role.6 Conversely, a detailed examination of karyotype, as outlined in Table 1, did not appear to support the possibility that specific cytogenetic lesions are assorted differentially with either the wild-type or mutant JAK2V617F allele. In particular, simultaneous expression of homozygous JAK2V617F and the recently described MMM-specific der(6)t(1;6)(q21p21)2 in 1 study patient excludes the possibility of mutual exclusivity and, instead, supports a collaborative multistep process in disease pathogenesis.
The second observation in the current study suggested a greater likelihood of treatment response to Epo-based drug therapy in MMM patients with prognostically favorable cytogenetic abnormalities. In general, current drug therapy for MMM remains suboptimal, and the response to specific agents is unpredictable.15 Of the currently available drugs, those with demonstrated efficacy for alleviating anemia include Epo,16 corticosteroids,17 androgens,18 danazol,19 and thalidomide alone20 or in combination with prednisone.21 Response rates with these drugs generally have been meager, and accurate prediction of response has not been possible in most instances. The current results suggest that the presence of favorable cytogenetic abnormalities may predict response to Epo-based therapy independent of the serum Epo level, provided the levels are not too high (e.g., < 200 mU/mL). However, the preliminary nature of the specific observation is underscored, and a larger sample size with serum Epo levels available on all patients will be required to validate the current finding and to determine the serum Epo-independent predictive value of cytogenetics on treatment response. Conversely, the lack of Epo response in a JAK2V617F homozygote with relatively low Epo levels raises the possibility that such patients are at peak JAK-STAT signaling and, thus, are unlikely to experience further erythropoietic drive from exogenous Epo administration. Finally, it should be recalled that a previous study in MMM had suggested that the presence of any cytogenetic abnormality made it less likely for a response to occur in the setting of androgen therapy.18 The number of androgen-treated patients in the current study was too small to comment on our observation in this particular regard.
- 11WHO histological classification of chronic myeloproliferative diseases. In: JaffeES, HarrisNL, SteinH, VardimanJW, editors. World Health Organization classification of tumors: tumours of the haematopoietic and lymphoid tissues. Lyon: International Agency for Research on Cancer Press; 2001., , .
- 17Agnogenic myeloid metaplasia. Acton, MA: Publishing Science Group; 1975..