Endogenous Erythroid Colony Formation in Chronic Myeloid Leukemia: A Recurrent Finding Associated with Persistent Minimal Residual Disease Under Imatinib

Abstract

In vitro endogenous erythroid colony (EEC) formation is a common finding in BCR-ABL-negative myeloproliferative neoplasms. The aim of the present study was to determine the prevalence and the clinical significance of EEC growth in chronic myeloid leukemia (CML). Results of clonogeneic progenitor cell assays from 52 patients with newly diagnosed CML were correlated with disease characteristics at presentation and molecular response to imatinib. EECs (median 7 per dish, range 1–39) were detectable in 16 patients (31%). The proportion of patients with a high-risk Sokal score was lower in the EEC group (7% vs. 30%, respectively). The cumulative incidence of achieving a major molecular response after 2 years of imatinib was similar for both groups. However, patients with EECs were less likely to achieve a more profound decline of BCR-ABL transcripts. After 6 years of imatinib, the cumulative probability [95% CI] of reaching a ≥4 log reduction of BCR-ABL was 48% [16%; 92%] for patients of the EEC group and 84% [63%; 97%] for patients of the No EEC group. The probability [95% CI] of achieving a >4.5 log reduction of BCR-ABL after 7 years was 13% [2%; 61%] for patients with EECs and 52% [30%; 78%] for patients without EECs. In vitro EECs disappeared after achievement of a major molecular response in all evaluable patients. The data indicate that EEC formation is a recurrent finding in patients with CML which deserves further attention as a possible biomarker predicting the degree of molecular response to imatinib.

Introduction

E rythroid progenitors from normal individuals and patients with reactive conditions require exogenous erythropoietin to form colonies in semisolid cultures. In contrast, progenitor cells from patients with myeloproliferative neoplasms can form hemoglobinized colonies in the absence of exogenous cytokines. Endogenous erythroid colony (EEC) growth is a hallmark of polycythemia vera, and it is also frequently observed in patients with essential thrombocytosis and primary myelofibrosis [1 –4]. Contradictory results have been published with regard to the occurrence of EEC growth in chronic myeloid leukemia (CML) [5 –11]. However, these studies comprise about 50 patients in total, and most of them date back more than 15 years. Whether the detection of EECs is of any clinical significance in CML has never been investigated.

The discovery of tyrosine kinase inhibitors (TKIs) has dramatically improved the treatment options of CML. Currently, three drugs are FDA approved for front-line use. Imatinib, the first TKI, has been the preferred initial therapy for the past 10 years [12,13]. The recently licenced second-generation TKIs dasatinib and nilotinib have been shown to be superior to imatinib in terms of antileukemic potency [14 –17]. However, observation periods for these drugs are short, and long-term outcomes have to be awaited. Moreover, both drugs are significantly more expensive than imatinib. Thus, there is not yet sufficient evidence to make specific recommendations on which TKI should be used first line, and the challenge is to individually select the front-line strategy that can be expected to achieve the best possible outcome [18 –21]. Identification of baseline prognostic markers in addition to the currently used Sokal score may facilitate this decision process.

In the present study, we show that EEC formation is found in a substantial proportion of patients with untreated CML. Patients with and without EECs are compared with regard to disease characteristics at presentation as well as their molecular response to imatinib.

Methods

This retrospective study includes 52 consecutive patients with newly diagnosed CML, who received imatinib as front-line therapy between May 2003 and December 2009. Patients were essentially treated according to European LeukemiaNet (ELN) guidelines [12]. In case of intolerance or imatinib resistance, treatment was switched to a second-line TKI (nilotinib 2×400 mg daily or dasatinib 1×100 mg daily).

Results of pretreatment peripheral blood progenitor cell assays were reviewed and, based on the presence or absence of EECs, patients were split into two groups (“EEC” vs. “No EEC”). The study was approved by the ethics committee of the Medical University of Vienna (EK 383/2011).

Colony assay

For evaluation of endogenous colony growth, 50 to 100×10³/mL peripheral blood mononuclear cells were cultured in 0.8% methylcellulose (Sigma), 30% fetal calf serum (Linaris), 10% bovine serum albumin (Sigma), α-thioglycerol (10⁻⁴ M) (Fluka Chemie AG), and Iscove's-modified Dulbecco's medium (Gibco-Invitrogen). Simultaneously, stimulated colony growth was assessed by adding 5 ng/mL recombinant human (rh) interleukin-3 (R&D Systems), 10 ng/mL rh GM-CSF (R&D Systems), and 2 U/mL rh EPO (Roche) to the cultures. All cultures were plated in duplicate. After a culture period of 14 days at 37°C in 5% CO₂ and full humidity, cultures were examined under an inverted microscope. Aggregates of at least 50 hemoglobinized cells, easily recognizable by their red color, were classified as burst-forming units—erythroid (BFU-E). Aggregates with more than 40 translucent, compact, or dispersed cells were identified as colony-forming units—granulocyte/macrophage (CFU-GM). Pluripotent progenitors (CFU-GEMM) were identified by their heterogenous composition of translucent and hemoglobinized cells. The presence of at least two BFU-E in duplicate endogenous cultures was reported as being EEC positive.

Molecular disease monitoring

Molecular disease response was evaluated at 3- to 6 month intervals. Quantification of BCR-ABL transcripts was performed by real-time polymerase chain reaction (PCR) using the BCR-ABL Mbcr FusionQuant ^® Mega Kit (Ipsogen) and the LightCycler^® 2.0-System (Roche). Results were expressed according to the International Scale [12,22]. BCR-ABL negativity in real-time PCR was confirmed by a two-step PCR with nested primers, which allows the detection of 1 BCR-ABL positive cell in 10^4.5 normal cells [23]. Three levels of BCR-ABL transcript reduction from the standardized baseline were applied to assess molecular response: a ≥3-log reduction persisting for at least 6 months, defined as major molecular response (MMR) [12]; a ≥4-log reduction persisting for at least 6 months, defined as MR⁴ [24]; and undetectable BCR-ABL persisting for at least 6 months, defined as MR^4.5 [24]. Screening for mutations in the BCR-ABL TK domain was carried out by bidirectional direct sequencing of amplicons generated by semi-nested PCR amplification [25].

The Janus kinase 2 (JAK2) MutaScreen™ and the MutaQuant ^® Kit (Ipsogen) were used for the detection/quantification of the JAK2V617F mutation from DNA samples frozen at diagnosis. These kits show a cut-off limit for positivity at 2% and 0.1% JAK2V617F, respectively. JAK2 exon 12 mutation screening was performed by direct sequencing of PCR products containing JAK2 exon 12 sequences in both directions [26].

Immunohistochemistry

Immunohistochemistry was performed on Schaffer's solution fixed, ethylendiaminetetraacetic acid-decalcified, and paraffin-embedded diagnostic bone marrow biopsies on an automated Leica BOND-III stainer (Leica Biosystems). Anti-phosphorylated (p)-signal transducer and activator of transcription (STAT) 5 (Tyr694) rabbit monoclonal antibody (No. 9359; Cell Signaling Technology) was applied at a dilution of 1:100 after appropriate antigen retrieval. To assess specific staining, positive and negative controls were run simultaneously with the cases. For double staining procedures, pSTAT5 staining, which provided a nuclear brown signal developed with DAB chromogen, was followed by immunostaining with anti-CD71 mouse monoclonal antibody (171M-95; Cell Marque) at a dilution of 1:1,000, exhibiting a bright red membranous staining developed with Fast Red chromogen. Due to the different expression patterns of pSTAT5 and CD71, colabeling could, hence, easily be discerned in the erythroid cells.

Gene expression profiling and real-time PCR

Total RNA was extracted from bone marrow MNCs preserved at diagnosis using an RNeasy kit (Quiagen). Preparation of terminal-labeled cDNA, hybridization to genome-wide human Gene Level 1.0 ST GeneChips (Affymetrix), and scanning of the arrays were carried out according to the manufacturer's protocols (www.affymetrix.com). Robust multi-array average signal extraction, normalization, and filtering (to eliminate genes with extremely low expression) were performed as described [27,28] (www.bioconductor.org/). Significant differences between sample groups were identified using a previously described method termed “SAM Analysis” (significance analysis of microarrays) [29].

Validation of microarray data was performed by RT-PCR on a Biorad CFx96™ Real Time System using iQ SYBR Green Supermix (Bio-Rad). cDNA was prepared using the Taqman reverse trascriptase kit (Applied Biosystems). The housekeeping gene β-actin (ACTB forward:gatcaagatcattgctcctcctg; ACTB reverse: ctgcttgctgatccacatctg) was used as an endogenous control. All primers were designed with Primer3Plus (www.bioinformatics.nl/cgi-bin/primer3plus/primer3plus.cgi/) and obtained from Eurofins MWG Operon (Table 2). For normalization threshold cycles (C_t-values), all analyses were normalized to β-actin within each sample to obtain sample-specific ΔC_t values (C_t gene of interest −C_t β-actin). To compare the two groups (with/without EECs) and obtain fold expression levels, the 2^−ΔΔCt method was used [30].

Statistical analysis

Statistical analysis was performed using SAS 9.1 for Windows and R 2.14.0 (library cmprsk). Continuous variables were compared between groups using the two-sided two-sample t-test if normally distributed or the Mann–Whitney U test if not normally distributed. Normal distribution was then checked by looking at histograms. For categorical variables, differences between groups were analyzed by chi-square tests or Fisher's exact tests, if appropriate. The correlation between baseline BCR-ABL transcript level and number of EECs within the EEC group was analyzed by Spearman correlation. To compare the groups EEC versus No EEC with regard to the endpoint MMR, a competing risk analysis was applied with “MMR” as the event of interest and “switch to second-generation TK inhibitors” as the competing event. Cumulative incidence curves were plotted, and differences between groups were analyzed using the Gray's test. For the endpoints “MR⁴” and “MR^4.5,” differences between survival curves of the two groups were analyzed using log rank tests, and Kaplan–Meier curves were plotted. For all analyses, the significance level has been set to 0.05.

Results

Sixteen (31%) patients exhibited spontaneous growth of erythroid blood progenitors. The median number of EECs per dish was 7 (range 1–39). Parallel colony assays from bone marrow were available in 24 patients. EECs were reported in 5 out of 6 patients of the EEC group and in none of 18 patients of the No EEC group. EECs were not observed in 30 control blood samples of patients with nonmalignant reactive leukocytosis who had also been referred for a colony assay during the study period.

Baseline disease characteristics

Pretreatment patient characteristics are listed in Table 1. All patients of the EEC group were in chronic phase of the disease. The No EEC group included four patients in advanced CML phase. Accordingly, the proportion of patients with high-risk Sokal and Hasford scores was lower in the EEC group. Both groups contained similar proportions of patients with low risk scores. Metaphase cytogenetics at diagnosis were available in 12 patients with EECs and in 35 patients without EECs. Additional cytogenetic changes accompanying the translocation t(9;22) were present in two patients of the EEC group [1× loss of Y and 1× t(9;10;22)(q34;q22;q11), 100% of metaphases, each] and in three patients of the No EEC group [1× del(11)(q23), 33% of metaphases; 1× t(2;10)(p12;q25), 25% of metaphases; 1× (8;17)(q13;q23), 40% of metaphases]. Hemoglobin levels were higher in patients with EECs, but this did not reach statistical significance. No difference was observed for WBC parameters and platelet counts as well as cytokine-stimulated colony growth of blood progenitors. Baseline BCR-ABL transcript levels were similar for both groups. Furthermore, within the EEC group, no significant correlation was found between BCR-ABL transcripts and numbers of EECs (P=0.28). One patient of the EEC group exhibited the JAK2V617F mutation. The frequency of mutated alleles was 46% in the bone marrow and 44% in the peripheral blood; the corresponding BCR-ABL:ABL ratios were 84% and 81%, respectively. The trephine biopsy showed a polycythemia vera-like morphology with minimal signs of concomitant fibrosis. None of the investigated patients exhibited a mutation of JAK2 exon 12.

Table 1.

Baseline Clinical and Hematological Characteristics of Patients with (EEC) and Without (No EEC) In Vitro Growth of Endogenous Erythroid Colonies

	EEC group (n=16)	No EEC group (n=36)	P value
Age, years			0.42
Mean±SD	55.6±19.8	51.1±17.7
Range	22–84	27–86
Sex			0.25
Female %	50	33
Male %	50	67
Disease phase			NA
Chronic phase %	100	89
Accelerated phase %	0	5.5
Blast phase %	0	5.5
Sokal risk score			0.05
Low %	43	53
Intermediate %	50	17
High %	7	30
Missing, n patients	2	0
Hasford risk score			0.31
Low %	43	47
Intermediate %	50	31
High %	7	22
Missing, n patients	2	0
Hemoglobin, g/dL			0.09
Mean±SD	12.9±2.0	11.9±2.1
Range	8.9–16.2	7.7–15.8
Platelets, 10⁹/L			0.79
Mean±SD	400±276	422±265
Range	142–1,017	112–1,476
White blood cells, 10⁹/L			0.95
Mean±SD	117.1±101.9	118.8±93.3
Range	18.4–316.6	17.5–375.3
Basophils,×10⁹/L			0.48
Median (Q1; Q3)	1.8 (0.9; 7.1)	2.9 (1.3; 7.5)
Range	0.3–19.7	0–24.8
Blasts, 10⁹/L			0.44
Median (Q1; Q3)	0.8 (0; 3.1)	1.2 (0; 6.7)
Range	0–9.5	0–27.4
Eosinophils, 10⁹/L			0.56
Median (Q1; Q3)	1.1 (4.5; 5.9)	2.6 (0.7; 7.7)
Range	0–18.3	0–22.1
LDH, U/L			0.35
Mean±SD	664±365	765±358
Range	252–1,705	185–1,627
Missing, n patients	0	1
BFU-E,/mL blood			0.96
Median (Q1; Q3)	12,876 (4,223; 25,173)	15,672 (3,430; 78,133)
Range	1,876–152,896	0–591,101
CFU-GM,/mL blood			0.80
Median (Q1; Q3)	5,969 (1,060; 21,874)	4,102 (1,708; 12,271)
Range	325–76,920	117–290,101
CFU-GEMM,/mL blood			0.44
Median (Q1; Q3)	130 (38; 522)	70 (0; 645)
Range	0–8,494	0–8,804
Missing, n patients	0	3
BCR-ABL:ABL ratio (%)			0.69
Mean±SD	65±25	67±23
Range	21–100	7–100
JAK2 mutations
JAK2 V617F, n patients	1	ND
JAK2 exon 12, n patients	0	ND

BFU-E, burst forming units-erythroid; CFU-GM, colony forming units-granulocyte/macrophage; CFU-GEMM, colony forming units-granulocyte/erythrocyte/macrophage/megakaryocyte; JAK2, Janus kinase 2; LDH, lactate dehydrogenase; NA, not applicable; ND, not determined; EEC, endogenous erythroid colony growth.

Molecular response to imatinib

The cumulative incidence of being in MMR after 2 years was 54% (95% CI [22%; 77%]) for the EEC group and 57% (95% CI [38%; 72%]) for the No EEC group (Fig. 1A). Patients with MMR were further evaluated with regard to a more profound decline of BCR-ABL. After 6 years of treatment, the cumulative probabilities of MR⁴ were 48% (95% CI [16%; 92%]) for the EEC group and 84% (95% CI [63%; 97%]) for the No EEC group, respectively (Fig. 1B). The cumulative risk of reaching MR^4.5 after 7 years of treatment was 13% (95% CI [2%; 61%]) for patients with EECs and 52% (95% CI [30%; 78%]) for patients without EECs (Fig. 1C). BCR-ABL kinase domain mutations occurred in two patients of the No EEC group (6%). None of the patients of the EEC group developed a mutation under imatinib.

FIG. 1.

Molecular response of patients with and without endogenous erythroid colony (EEC) growth to imatinib. (A) Estimated cumulative incidence curves with major molecular response (MMR; upper curves) and treatment switch (lower curves) as competing events. (B) Cumulative probability of MR⁴. (C) Cumulative probability of MR^4.5.

Molecular response after treatment switch

Four patients of the EEC group (25%) were switched to a second-generation TKI. One of them achieved an MMR after 6 months of treatment with nilotinib. The patient with the additional JAK2V617F mutation was switched to dasatinib and also obtained an MMR after 3 months. Concomitantly, the JAK2V617F mutation disappeared from peripheral blood and bone marrow. One year later, the patient achieved an MR^4.5. The other two patients failed to achieve an MMR to second-generation TKIs. Ten patients of the No EEC group (28%) were switched to either dasatinib or nilotinib, and two of them achieved an MMR after 4 and 14 months of nilotinib.

Follow up of EECs

In 10 patients of the EEC group, at least three follow-up investigations of colony growth were available. EECs completely disappeared in seven patients who achieved an MMR to imatinib. In the two patients with an MMR to second-generation TKIs, EECs permanently disappeared after treatment switch. In one patient who never achieved an MMR, EECs remained intermittently detectable.

Expression of pSTAT5 in diagnostic bone marrow biopsies

Expression patterns of pSTAT5 were examined by immunohistochemistry in 10 bone marrow specimens of patients with and without EEC growth, respectively. In both groups, pSTAT5 expression was detected in variable amounts of CD71-positive erythroid precursors, megakaryocytes, and cells of the myeloid lineage. No differences in terms of cellular distribution or intensity of staining were recognized between groups (Fig. 2).

FIG. 2.

Representative bone marrow biopsies of a patient with EECs (A) and a patient without EECs (B) double stained with anti- pSTAT5 (brown) and anti-CD71 (red). Individual CD71-positive erythropoietic cells (thick black arrows), some CD71-negative myeloid cells (white arrows), and occasional megakaryocytes (thin black arrows) show faint expression of pSTAT5.

Genes associated with EEC growth

Microarray analysis was performed to screen for potential genes involved with EEC growth. Twelve patients with EECs and 12 patients without EECs were matched for age, sex, and Sokal score, and a total of four Gene Chips per group was prepared by pooling three patients, respectively. The top differentially expressed genes (greater than fourfold change assessed by SAM software) are listed in Table 2. Interestingly, they were invariably expressed at lower levels in samples from patients with EEC growth when compared with samples from the No EEC group. To validate the results of GeneChip experiments, we performed RT-PCR using RNA samples from 15 patients of the EEC group and from 35 patients of the No EEC group. Two further genes, histamin receptor H4 (HRH4) and histidin decarboxylase (HDC), with a greater than threefold (but less than fourfold) expression difference in microarrays were additionally included, as they were topically related to genes FCER1A (Fc fragment of IgE, high affinity I, receptor for; alpha polypeptide), MS4A2 (membrane-spanning 4-domains, subfamily A, member 2 [Fc fragment of IgE, high affinity I, receptor for; beta polypeptide]), and CPA3 (carboxypeptidase A3, mast cell) of the top group. CML with EEC growth was significantly associated with lower mRNA levels of FCER1A, CPA3, HDC, and HRH4 when compared with CML without EEC growth. We next determined mRNA levels of these four genes in 16 unaffected bone marrows of patients with either Hodgkin's or Non-Hodgkin's lymphomas. As illustrated in Fig. 3, gene expression levels in EEC-positive CML were similar or even lower than in controls without CML.

FIG. 3.

Relation of RT-PCR data to normal controls. Data are expressed as fold expression (mean±SEM) of genes FCER1A, HRH4, CPA3, and HDC in chronic myeloid leukemia (CML) groups (gray and black bars) versus normal controls (white bars). BM, bone marrow.

Table 2.

Fold Change of the Top Differentially Expressed Genes Using Microarrays and Validation by Real-Time Polymerase Chain Reaction

				Real-time PCR
Gene name	Symbol	GenBank ID	Microarray Fold change EEC/No EEC	Primer sequences	Fold change^a EEC/No EEC	P value
Fc fragment of IgE, high affinity I, receptor for; alpha polypeptide	FCER1A	NM_002001	0.25	FCER1A_425f tcagtgactggctgctccttc	0.15±0.03	0.037
				FCER1A_499r catggcacctgaggaagagg
Solute carrier family 14 (urea transporter), member 1 (Kidd blood group)	SLC14A1	NM_001128588	0.24	SLC14A1_109f gccactgccttctgctgcca	0.21±0.08	0.114
				SCL14A1_376r ctggctcctgttcctgcgacc
Interferon-induced protein with tetratricopeptide repeats 1-like	IFIT1L	NM_001010987	0.17	IFIT1L_433f tggacaaggtggagaacacttg	0.10±0.02	0.072
				IFIT1L_515r ttcctcacagtccacctctgg
Solute carrier family 4, anion exchanger, member 1 (erythrocyte membrane protein band 3, Diego blood group)	SLC4A1	NM_000342	0.24	SLC4A1_1910f gctgcgcaagttcaagaacag	0.18±0.08	0.184
				SLC4A1_2001r ccatgatcaggatggagatgg
STE20-related kinase adaptor beta	STRADB	NM_018571	0.21	STRADB_996f aagcatggacagaggcatagg	0.24±0.08	0.100
				STRADB_1084r tgtctcagtagttctggactcagc
Glycophorin A (MNS blood group)	GYPA	NM_002099	0.18	GYPA_188f ggtggcaatgcacacttcaac	0.34±0.19	0.251
				GYPA_286r ggagtggctgcatatgtgtcc
Killer cell lectin-like receptor subfamily K, member 1	KLRK1	NM_007360	0.25	KLRK1_683f aagatggctccattctctcacc	0.39±0.16	0.914
				KLRK1_760r tcgaggcatagagtgcacagtc
Carboxypeptidase A3 (mast cell)	CPA3	NM_001870	0.24	CPA3_123f agaaggtgttccgcgtgaag	0.16±0.02	0.016
				CPA3_222r gtggcacctggataccagaag
G protein-coupled receptor 128	GPR128	NM_032787	0.01	GPR128_1496f tgccgctgcaaccatactac	0.75±0.74	0.729
				GPR128_1605r acagacagtgcacatccaacg
Membrane-spanning 4-domains, subfamily A, member 2 (Fc fragment of IgE, high affinity I, receptor for; beta polypeptide)	MS4A2	NM_000139	0.20	MS4A2_502f aacactgccagcagcatagc	0.27±0.06	0.079
				MS4A2_598r tctggcaactgtggatgtgg
Aminolevulinate, delta-, synthase 2	ALAS2	NM_000032	0.23	ALAS2_476f ggaagccattttccggtcccca	0.27±0.06	0.175
				ALAS2_968r tcgcaccctggcaggatcttgg
Histamine receptor H4	HRH4	NM_021624	0.32	HRH4_438f acctgtcagtctcaaatgctgtgtc	0.28±0.04	0.031
				HRH4_525r cagcacccaaacggccacca
Histidine decarboxylase	HDC	NM_002112	0.29	HDC_2027f agcacaagctgtcctccttcc	0.06±0.03	0.028
				HDC_2128r tctgagcactcactggcacac

Values are means±SEM.

Bold numbers indicate significant p-values (<0.05).

EEC, endogenous erythroid colony growth; SEM, standard error of mean; PCR, polymerase chain reaction.

Discussion

The present study clearly indicates that in vitro growth of EECs is a recurrent phenomenon in BCR-ABL-positive CML. EECs in semisolid cultures are typically found in myeloproliferative neoplasms with mutations of the JAK2 gene that lead to constitutive activation of the erythropoesis triggering JAK2-STAT5 pathway [31]. Patients with BCR-ABL-positive CML usually lack mutations of the JAK2 gene except rare cases that have been reported to harbor both, the BCR-ABL translocation and the JAK2V617F mutation [32 –35]. In our series of EEC-positive cases, a JAK2V617F mutation was detected in 1 patient. The JAK2V617F and the BCR-ABL burden overlapped at diagnosis and simultaneously decreased on BCR-ABL-targeted therapy, indicating that the JAK2V617F mutant was harbored by the BCR-ABL-positive clone. EECs disappeared on TKI treatment in all evaluable patients and returned only in one patient who failed to achieve an MMR. Together, these results suggest that EEC growth in CML indicates a disease subtype with different biological properties rather than the presence of a second malignancy. It has also been previously shown that EECs cultured from bone marrow of patients with CML carry the Philadelphia chromosome [5].

Constitutive activation of STAT5 has been shown to induce EEC formation in human primary cells as well as to be critical for the maintenance of CML [36,37]. We, therefore, hypothesized that EEC growth in CML might be associated with increased levels of pSTAT5. Our immunohistochemistry studies, however, gave no evidence that pSTAT5 is differentially expressed in bone marrows of patients with and without EEC growth. Moreover, the results of gene expression screening do not suggest that genes associated with erythroid proliferation/differentiation are differently regulated in CMLs with and without EECs (see also Supplementary Table S1; Supplementary Data are available online at www.liebertpub.com/scd). Instead, we found the basophil/mast cell-associated genes FCR1A, CPA3, HDC, and HRH4 to be significantly lower expressed in bone marrows of CMLs with EEC growth compared with CMLs without EECs. This difference cannot be simply explained by different degrees of basophilia, as (1) basophil numbers did not significantly differ between CML groups, and (2) mRNA levels of CMLs with EEC growth were similar or even lower than those of normal controls. The data rather suggest that basophil/mast cell-associated genes are up-regulated in CML without EEC growth and that the presence of EECs indicates a less determined differentiation state of leukemic cells with multi-lineage potential.

The few reports on EEC growth in CML date from the time before BCR-ABL targeted therapy. Therefore, we were particularly interested to evaluate whether patients with EECs exhibit different sensitivity to imatinib. In the present cohort, the cumulative incidence of achieving an MMR within 2 years was similar for both groups. However, patients with EECs had a lower probability to reach more profound reduction of BCR-ABL transcript levels. So far, reduction of BCR-ABL transcripts beyond the level of MMR has not been reported to translate into better survival. However, several studies have demonstrated that patients with undetectable BCR-ABL transcripts are most likely to maintain optimal treatment response [38 –42]. Furthermore, it has been shown that around 40%–50% of patients, who achieve a stable complete molecular response on imatinib, can stop therapy for at least several years without recurrence of molecular disease [43]. Thus, maximal disease reduction is a preferable goal of TKI treatment of CML [44,45].

In conclusion, our data suggest that EEC formation in CML deserves attention as a potential biomarker predicting the degree of molecular response to imatinib. Further prospective studies are questioned to strengthen the presented observations. If EEC growth actually turns out to indicate patients with inferior response to imatinib, the endogenous colony assay may become a valuable tool in the decision-making process for first-line therapy in CML.

Author Disclosure Statement

P.V. received consultant fee from Novartis and grants and payment for lectures from Novartis and BMS.

References

Mossuz

. 2006. Influence of the assays of endogenous colony formation and serum erythropoietin on the diagnosis of polycythemia vera and essential thrombocythemia. Semin Thromb Hemost, 32:246–250.

Tefferi

, Vardiman

. 2008. Classification and diagnosis of myeloproliferative neoplasms: the 2008 World Health Organization criteria and point-of-care diagnostic algorithms. Leukemia, 22:14–22.

Florensa

, Besses

, Woessner

, Sole

, Acin

, Pedro

, Florensa

, Sans-Sabrafen

. 1995. Endogenous megakaryocyte and erythroid colony formation from blood in essential thrombocythaemia. Leukemia, 9:271–273.

Lutton

, Levere

. 1979. Endogenous erythroid colony formation by peripheral blood mononuclear cells from patients with myelofibrosis and polycythemia vera. Acta Haematol, 62:94–99.

Dainiak

, Liu

, Dewey

, Kulkarni

. 1984. Chromosome analysis of isolated colony erythroblasts in chronic myelogenous leukaemia. Br J Haematol, 56:507–512.

Strobel

, Messner

, Löhr

, Fausa

. 1983. Fetal hemoglobin analysis of erythroid bursts in patients with chronic myelogenous leukemia (CML) Exp Hematol, 11:982–986.

Hara

, Kai

, Fushimi

, Taniwaki

, Ifuku

, Okamoto

, Ohe

, Fujita

, Noguchi

et al. 1981. Pluripotent, erythrocytic and granulocytic hemopoietic precursors in chronic granulocytic leukemia. Exp Hematol, 9:871–877.

Dobo

, Pineau

, Zandecki

, Hunault

, Hermouet

. 1999. Endogenous erythroid and megakaryocytic colony formation in serum-free, cytokine-free collagen gels. J Hematother Stem Cell Res, 8:601–607.

Adams

, Barrett

, Beard

, McCarthy

. 1988. Primary polycythaemia, essential thrombocythaemia and myelofibrosis—three facets of a single disease process? Acta Haematol, 79:33–37.

10.

Issaad

, Vainchenker

. 1994. Growth of erythroid colonies in chronic myelogenous leukemia is independent of erythropoietin only in the presence of steel factor. Blood, 84:3447–3456.

11.

Weinberg

, Worsley

, Gilbert

, Cuttner

, Berk

, Alter

. 1989. Comparison of erythroid progenitor cell growth in vitro in polycythemia vera and chronic myelogenous leukemia: only polycythemia vera has endogenous colonies. Leuk Res, 13:331–338.

12.

Baccarani

, Cortes

, Pane

, Niederwieser

, Saglio

, Apperley

, Cervantes

, Deininger

, Gratwohl

et al. 2009. Chronic myeloid leukemia: an update of concepts and management recommendations of European LeukemiaNet. J Clin Oncol, 27:6041–6051.

13.

Hochhaus

, O'Brien

, Guilhot

, Druker

, Branford

, Foroni

, Goldman

, Müller

, Radich

et al. IRIS Investigators. 2009. Six-year follow-up of patients receiving imatinib for the first-line treatment of chronic myeloid leukemia. Leukemia, 23:1054–1061.

14.

Saglio

, Kim

, Issaragrisil

, le Coutre

, Etienne

, Lobo

, Pasquini

, Clark

, Hochhaus

et al. ENESTnd Investigators. 2010. Nilotinib versus imatinib for newly diagnosed chronic myeloid leukemia. N Engl J Med, 362:2251–2259.

15.

Kantarjian

, Shah

, Hochhaus

, Cortes

, Shah

, Ayala

, Moiraghi

, Shen

, Mayer

et al. 2010. Dasatinib versus imatinib in newly diagnosed chronic-phase chronic myeloid leukemia. N Engl J Med, 362:2260–2270.

16.

Kantarjian

, Hochhaus

, Saglio

, De Souza

, Finn

, Stenke

, Goh

, Rosti

, Nakamae

et al. 2011. Nilotinib versus imatinib for the treatment of patients with newly diagnosed chronic phase, Philadelphia chromosome-positive, chronic myeloid leukaemia: 24-month minimum follow-up of the phase 3 randomised ENESTnd trial. Lancet Oncol, 9:841–851.

17.

Kantarjian

, Shah

, Cortes

, Baccarani

, Agarwal

, Undurraga

, Wang

, Ipina

, Kim

et al. 2012. Dasatinib or imatinib in newly diagnosed chronic-phase chronic myeloid leukemia: 2-year follow up from a randomized phase 3 trial (DASISION) Blood, 119:1123–1129.

18.

Marin

. 2012. Initial choice of therapy among plenty for newly diagnosed chronic myeloid leukemia. Am Soc Hematol Ed Book, 2012:115–120.

19.

Baccarani

, Pileri

, Steegmann

J-L

, Muller

, Soverini

, Dreyling

. on behalf of the ESMO Guidelines Working Group. 2012. Chronic myeloid leukemia: ESMO clinical practice guidelines for diagnosis, treatment and follow-up (2012) Ann Oncol, 23:vii72–vii77.

20.

Shami

, Deininger

. 2012. Evolving treatment strategies for patients newly diagnosed with chronic myeloid leukemia: the role of second-generation BCR-ABL inhibitors as first-line therapy. Leukemia, 26:214–224.

21.

Valent

. 2011. Severe adverse events associated with the use of second-line BCR/ABL1 tyrosine kinase inhibitors: preferential occurrence in patients with comorbidities. Haematologica, 96:1395–1397.

22.

Müller

, Cross

, Erben

, Schenk

, Hanfstein

, Ernst

, Hehlmann

, Branford

, Saglio

, Hochhaus

. 2009. Harmonization of molecular monitoring of CML therapy in Europe. Leukemia, 23:1957–1963.

23.

Maurer

, Janssen

, Thiel

, van Denderen

, Ludwig

, Aydemir

, Heinze

, Fonatsch

, Harbott

, Reiter

. 1991. Detection of chimeric BCR-ABL genes in acute lymphoblastic leukaemia by the polymerase chain reaction. Lancet, 337:1055–1058.

24.

Cross

NCP

, White

, Müller

, Saglio

, Hochhaus

. 2012. Standardized definitions of molecular response in chronic myeloid leukemia. Leukemia, 26:2172–2175.

25.

Hochhaus

, Kreil

, Corbin

, La Rosee

, Müller

, Lahaye

, Hanfstein

, Schoch

, Cross

et al. 2002. Molecular and chromosomal mechanisms of resistance to imatinib (STI571) therapy. Leukemia, 16:2190–2196.

26.

Scott

, Tong

, Levine

, Scott

, Beer

, Stratton

, Futreal

, Erber

, McMullin

et al. 2007. JAK2 exon 12 mutations in polycythemia vera and idiopathic erythrocytosis. N Engl J Med, 356:459–468.

27.

Bilban

, Haslinger

, Prast

, Klinglmüller

, Woelfel

, Haider

, Sachs

, Otterbein

, Desoye

et al. 2009. Identification of novel trophoblast invasion-related genes: heme oxygenase-1 controls motility via peroxisome proliferator-activated receptor gamma. Endocrinology, 150:1000–1013.

28.

Irizarry

, Bolstad

, Collin

, Cope

, Hobbs

, Speed

. 2003. Summaries of Affymetrix GeneChip probe level data. Nucleic Acids Res, 31:e15.

29.

Tusher

, Tibshirani

, Chu

. 2001. Significance analysis of microarrays applied to the ionizing radiation response. Proc Natl Acad Sci USA, 98:5116–5121.

30.

Livak

, Schmittgen

. 2001. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) method. Methods, 25:402–408.

31.

James

, V Ugo

, Casadevall

, Constantinescu

, Vainchenker

. 2005. A JAK2 mutation in myeloproliferative disorders: pathogenesis and therapeutic and scientific prospects. Trends Mol Med, 11:546–554.

32.

Krämer

, Reiter

, Kruth

, Erben

, Hochhaus

, Müller

, Cross

, Jones

, Ho

, Hensel

. 2007. JAK2-V617F mutation in a patient with Philadelphia-chromosome-positive chronic myeloid leukaemia. Lancet Oncol, 8:658–660.

33.

Bornhäuser

, Mohr

, Oelschlaegel

, Bornhäuser

, Jacki

, Ehninger

, Thiede

. 2007. Concurrent JAK2(V617F) mutation and BCR-ABL translocation within committed myeloid progenitors in myelofibrosis. Leukemia, 21:1824–1826.

34.

Hussein

, Bock

, Seegers

, Flasshove

, Henneke

, Buesche

, Kreipe

. 2007. Myelofibrosis evolving during imatinib treatment of a chronic myeloproliferative disease with coexisting BCR-ABL translocation and JAK2V617F mutation. Blood, 109:4106–4107.

35.

Inami

, Inokuchi

, Okabe

, Kosaka

, Mitamura

, Yamagucci

, Dan

. 2007. Polycythemia associated with the JAK2V617F mutation emerged during treatment of chronic myelogenous leukemia. Leukemia, 21:1103–1104.

36.

Garcon

, Rivat

, James

, Lacout

, Camara-Clayette

, Ugo

, Lecluse

, Bennaceur-Griscelli

, Vainchenker

. 2006. Constitutive activation of STAT5 and Bcl-xL overexpression can induce endogenous erythroid colony formation in human primary cells. Blood, 108:1551–1554.

37.

, Wolff

, Li

, Zhang

, Ilaria

. 2006. STAT5 signaling is required for the efficient induction and maintenance of CML in mice. Blood, 107:4917–4925.

38.

Press

, Galderisi

, Yang

, Rempfer

, Willis

, Mauro

, Druker

, Deininger

. 2007. A half-log increase in BCR-ABL RNA predicts a higher risk of relapse in patients with chronic myeloid leukemia with an imatinib-induced complete cytogenetic response. Clin Cancer Res, 13:6136–6143.

39.

Branford

, Seymour

, Grigg

, Arthur

, Rudzki

, Lynch

, Hughes

. 2007. BCR-ABL messenger RNA levels continue to decline in patients with chronic phase chronic myeloid leukemia treated with imatinib for more than 5 years and approximately half of all first-line treated patients have stable undetectable BCR-ABL using strict sensitivity criteria. Clin Cancer Res, 13:7080–7085.

40.

Hughes

, Branford

. 2009. Monitoring disease response to tyrosine kinase inhibitor therapy in CML. Am Soc Hematol Ed Book, 2009:477–487.

41.

Kantarjian

, O'Brien

, Shan

, Huang

, Garcia-Manro

, Faderl

, Ravandi-Kashani

, Verstovsek

, Beth Rios

, Cortes

. 2008. Cytogenetic and molecular responses and outcome in chronic myelogenous leukemia: need for new response definitions? Cancer, 112:837–845.

42.

Colombat

, Fort

, Chollet

, Marit

, Roche

, Preudhomme

, Reiffers

, Praloran

, Mahon

. 2006. Molecular remission in chronic myeloid leukemia patients with sustained complete cytogenetic remission after imatinib mesylate treatment. Haematologica, 91:162–168.

43.

Mahon

, Réa

, Guilhot

, Huguet

, Nicolini

, Legros

, Charbonnier

, Guerci

et al. on behalf of the Intergroupe Français des Leucémies Myéloïdes Chroniques. 2010. Discontinuation of imatinib in patients with chronic myeloid leukaemia who have maintained complete molecular remission for at least 2 years: the prospective, multicentre Stop Imatinib (STIM) trial. Lancet Oncol, 11:1029–1035.

44.

Cortes

, Hochhaus

, Hughes

, Kantarjian

. 2011. Front-line and salvage therapies with tyrosine kinase inhibitors and other treatments in chronic myeloid leukemia. J Clin Oncol, 29:524–531.

45.

Melo

, Ross

. 2011. Minimal residual disease and discontinuation of therapy in chronic myeloid leukemia: can we aim at a cure? Hematology Am Soc Hematol Educ Program, 2011:136–142.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.03 MB