Article Tools
Hematologic Malignancies
Article Tools
Low ERG and BAALC Expression Identifies a New Subgroup of Adult Acute T-Lymphoblastic Leukemia With a Highly Favorable Outcome
Abstract
Expression of the genes ERG (v-ets erythroblastosis virus E26 oncogene homolog) and BAALC (brain and acute leukemia, cytoplasmic) shows similarity during hematopoietic maturation and predicts outcome in acute myeloid leukemia. We hypothesized that like ERG, BAALC expression might be of prognostic significance in acute T-lymphoblastic leukemia (T-ALL) and that ERG and BAALC expression together would better identify the patient's risk profile.
ERG and BAALC mRNA expression were determined by real-time reverse transcriptase polymerase chain reaction in 153 adults with T-ALL. Patients were designated low or high ERG expressers and low or high BAALC expressers.
High BAALC expression correlated with a higher frequency of early T-ALL (P < .0001), CD34 positivity (P < .0001), coexpression of myeloid markers (P = .0001), and high ERG expression (P = .03). High BAALC compared with low BAALC patients had an inferior relapse-free survival (RFS; P = .0008) and overall survival (OS; P = .0001). In contrast, patients with low expression of both ERG and BAALC (representing 41% of all T-ALL patients) had the most favorable outcome (P < .0001; 4-year RFS: low ERG/low BAALC 81%; P < .0001; 4-year OS: low ERG/low BAALC 69%). On multivariable analysis, low ERG/low BAALC expression was of independent favorable prognostic significance (RFS, P = .001; OS, P = .003).
Acute T-lymphoblastic leukemia (T-ALL) comprises approximately 25% of adult ALL and is a clinically heterogeneous disease, likely including distinct molecular entities.1 Treatment optimization in acute leukemia requires the accurate assignment of patients at diagnosis to specific risk groups. Progress has been made in the molecular characterization of clinically distinct subgroups of acute myeloid leukemia (AML) and B-lymphoblastic leukemia.2,3 Moreover, risk-adapted treatment designs have been successfully applied to these patients, guided by molecular markers with proven prognostic significance.4,5 In contrast, few molecular risk factors have been identified in T-ALL. With a long-term disease-free survival rate of only 32% to 46%, high-risk patients may be candidates for alternative treatment regimens.1,4 Additionally, molecular markers might more precisely define good-risk patients, who may not benefit from treatment intensification including stem-cell transplantation (SCT).
Presenting clinical features and immunophenotype are currently the most widely accepted prognostic factors in T-ALL.6,7 In search of novel molecular risk markers, we have previously identified high-level expression of the oncogenic ETS transcription factor ERG (v-ets erythroblastosis virus E26 oncogene homolog) as an independent adverse prognostic factor that identifies T-ALL patients with a high risk of relapse and inferior survival.8 ETS transcription factors are key players in lineage-specific regulation during commitment and differentiation of lymphocytes, with induction of ERG expression in early T-cell development and downregulation of ERG with further T-lineage commitment.9,10 In addition to its physiological role in early T-cell development, ERG has been implicated in AML. ERG is involved in the rare t(16;21)(p11;q22), where ERG is fused to the FUS gene,11 and overexpression has been observed in the prognostically inferior subgroup of AML with a complex karyotype.12 Moreover, in AML lacking chromosomal aberrations high level ERG expression has been associated with an immature phenotype and was an independent risk factor predicting inferior outcome.13
Similar findings have been observed for the gene BAALC (brain and acute leukemia, cytoplasmic), which is highly expressed in normal uncommitted progenitor cells and downregulated with the onset of differentiation.14,15 Furthermore, high BAALC expression was found in patients with AML carrying a complex karyotype and was of independent adverse prognostic significance in AML patients with normal cytogenetics.12,15 Like ERG expression, BAALC expression was also associated with a more immature leukemic phenotype, reflecting the genes' physiological expression in the hematopoietic progenitor compartment.
Because of these similarities in expression pattern, ERG and BAALC may both be markers of an immature leukemic subtype. We hypothesized that in addition to ERG expression, BAALC expression might be of prognostic significance in T-ALL, and that the combined assessment of ERG and BAALC could more precisely identify the patient's risk profile.
We studied 153 consecutive adult patients with newly diagnosed T-ALL who were enrolled between 1993 and 2003 on the German Acute Lymphoblastic Leukemia (GMALL) 05/93 and 06/99 multicenter protocols.17,18 The study included 105 patients previously assessed for ERG expression.8 The GMALL protocols included intensive chemotherapy as well as autologous (n = 8) or allogeneic (n = 25) SCT. All patients gave written informed consent to participate in the study according to the Declaration of Helsinki. The study was approved by the ethics board of the Johann Wolfgang Goethe-Universität Frankfurt, Germany.
Pretreatment bone marrow (BM) samples were collected centrally and enriched for the blast fraction by density-gradient centrifugation (Ficoll-Paque Plus; Amersham Biosciences, Uppsala, Sweden). Immunophenotyping of fresh samples was centrally performed in the GMALL reference laboratory at the Charité, Germany. Immunophenotyping was carried out as previously described.19,20 T-lineage leukemias were subclassified into pre–T-ALL or early T-ALL (cyCD3+, CD7+, CD5±, CD2−, sCD3−, CD4−/+, CD8−/+, CD1a− or cyCD3+, CD7+, CD5−, CD2+, sCD3−, CD4−, CD8−, CD1a−), thymic T-ALL (cyCD3+, CD7+, CD5±, CD2±, sCD3±, CD4+, CD8+, CD1a+), and mature T-ALL (cyCD3+, CD7+, CD5+, CD2+, sCD3±, CD4±, CD8±, CD1a−). Coexpression of myeloid markers was defined when leukemic blasts were positive for CD13 and/or CD33 expression.
Isolation of total RNA from mononuclear cells of pretreatment BM samples was carried out using the RNeasy Mini Kit (Qiagen, Hilden, Germany) according to manufacturer's instructions. Complementary DNA (cDNA) was synthesized using 500 ng of total RNA and avian myeloblastosis virus reverse transcriptase (RT-AMV; Roche, Mannheim, Germany) at 50°C for 60 minutes in the presence of RNase inhibitor (RNasin; Roche).
BAALC and ERG mRNA expression levels were determined by comparative real-time reverse transcriptase polymerase chain reaction (RT-PCR) as previously described.8,21 Absence or presence of HOX11 and HOX11L2 expression were determined in 141 of the samples by real-time RT-PCR.8
Comparisons of baseline clinical variables across groups were made using the Fisher's exact test for categoric data; the nonparametric Mann-Whitney U test was applied for continuous variables. A P value ≤ .05 (two-sided) was considered significant.
Patients analyzed for BAALC expression were initially divided into quartile groups according to levels of BAALC expression and subsequently dichotomized into the three lower quartiles (Q1, Q2, and Q3) and the upper quartile (Q4) of BAALC expression values. Q4 was chosen for the cut point because this cohort showed a clinically distinct outcome with an inferior relapse-free survival (RFS) compared with the remaining quartiles (Appendix Fig A1, online only). Patients were defined as low BAALC with expression levels in Q1 to Q3 (n = 111) and as high BAALC with expression levels in Q4 (n = 37). This classification was data driven; thus, results based on it should be considered exploratory.
ERG expression levels were dichotomized at their median, and patients were classified as having low ERG (n = 76) if they had expression values within the lower 50% and as high ERG (n = 76) if they had ERG expression values within the upper 50% of all measured values.8
Achievement of complete remission (CR) was assessed after completion of induction chemotherapy and required granulocytes at least 1,500/μL, platelets at least 100,000/μL, no peripheral blood (PB) blasts, BM cellularity more than 20% with maturation of all cell lines, less than 5% BM blasts, and absence of extramedullary leukemia. Primary resistance to chemotherapy was defined as more than 25% blasts in the BM or persistence of PB blasts after induction. Relapse was defined as the reappearance of circulating blasts, more than 5% BM blasts, or development of extramedullary leukemia.
The median follow-up was 54 months. After the exclusion of 33 patients who received SCT as consolidation treatment, 119 T-ALL patients were included for displayed analyses of overall survival (OS) and 97 for RFS.
OS was calculated using the Kaplan-Meier method, and the log-rank test was used to compare differences between survival curves. OS was measured from the protocol on-study date until the date of death regardless of cause, excluding for patients alive at last follow-up. RFS was defined only for patients who achieved CR, and was measured from the date of attaining CR until the date of relapse, excluding for death in CR (n = 2). Patients with no report of relapse by the end of the follow-up observation were censored on the date of last follow-up.
Cox proportional hazards models were constructed for RFS and OS. The following covariates were included in the full model of OS and RFS: BAALC expression (low v high), ERG expression (low v high), presence or absence of HOX11 and HOX11L2 expression, sex, WBC (< 100,000/μL v ≥ 100,000/μL), age (< median 29 v ≥ 29 years), immunophenotype (early, thymic, mature), mediastinal mass (presence v absence), CNS involvement (presence v absence), CD34 positivity, and expression of myeloid markers (CD13 and/or CD33). Stepwise forward selection was performed. The combined prognostic validity of ERG and BAALC expression was investigated in an exploratory multivariable analysis. Automatic variable selection identified a prognostically favorable subgroup of patients with low ERG and low BAALC values. All calculations were performed using the SPSS software package, version 12 (SPSS Inc, Chicago, IL).
Of the 153 T-ALL patients, material in sufficient quality and quantity was available for the determination of BAALC in 148 and ERG in 152 cases, with 147 cases studied for expression of both genes. Patients were defined as low and high BAALC and low and high ERG as described in the statistical section.
Patients with high versus low BAALC expression were less frequently males (P = .01) and less often presented with CNS involvement (P = .04; Table 1). A significant correlation was observed between ERG and BAALC expression (P = .03). In addition, high expression of BAALC, as for ERG,8 was associated with the immature phenotype of early T-ALL (P < .0001; Table 1). High BAALC expression was further characterized by CD34 positivity (62% v 22% for low BAALC; P < .0001) and aberrant expression of the myeloid markers CD33 and/or CD13 (P = .0001; Table 1). There were no significant differences between patients with low and high ERG expression with respect to pretreatment WBC, age, sex, presenting mediastinal mass or CNS involvement, positivity for HOX11 or HOX11L2 expression, or expression of myeloid markers or CD34 (data not shown).
Compared with low BAALC expressers, patients with high BAALC had a lower CR rate (73% v 89%; P = .03) and a higher relapse rate (71% v 31%; P = .003; Table 1). Furthermore, high BAALC expression was highly predictive of an inferior RFS (P = .0008; 4-year RFS: high BAALC 21% [95% CI, 0% to 44%] v low BAALC 65% [95% CI, 53% to 77%]) and OS (P = .0001; 4-year OS: high BAALC 18% [95% CI, 2% to 34%] v low BAALC 58% [95% CI, 47% to 69%]; Fig 1). In accordance with the previous data,8 this extended study showed no difference in the CR rate between the low and high ERG groups, with high ERG expressers having a significantly higher relapse rate (56%) compared with low-ERG patients (25%; P = .004). In addition, patients with high ERG expression had a significantly shorter RFS (P = .002; 4-year RFS: high ERG 36% [95% CI, 18% to 54%] v low ERG 71% [95% CI: 58% to 85%]) and OS (P = .004; 4-year OS: high ERG 32% [95% CI, 18% to 46%] v low ERG 61% [95% CI, 48% to 74%]).
On the basis of ERG and BAALC expression status, we identified four different subgroups: low ERG/low BAALC (n = 61), low ERG/high BAALC (n = 12), high ERG/low BAALC (n = 50), and high ERG/high BAALC (n = 24). Patients with high expression of ERG and/or BAALC were characterized by an inferior outcome with no significant differences between the three groups; thus, these patients were considered a single group (Appendix Fig A2, online only).
Patients with low ERG/low BAALC expression compared with patients in the combined group were more frequently males (85% v 66%; P = .01; Table 2). Moreover, low-ERG/low-BAALC patients predominately showed a thymic immunophenotype (74% v 49% combined group; P = .006). In contrast, patients with high ERG and/or high BAALC expression exhibited immature features with CD34 positivity (42% v 18% of low-ERG/low-BAALC patients; P = .002), as well as expression of myeloid markers (P = .005; Table 2). No significant differences were seen between the two groups with respect to age, WBC, HOX11, and HOX11L2 status, or presenting with CNS involvement or mediastinal mass (data not shown).
Low-ERG/low-BAALC patients—representing 41% of all T-ALL patients—had a very favorable outcome with a significantly lower relapse rate (17% v 60%; P < .0001; Table 2) compared with patients in the combined group. Moreover, low-ERG/low-BAALC patients had a superior RFS (P < .0001; relapse-free at 4 years: 81% v 33%) and OS (P = .0001; alive at 4 years: 69% v 30%) compared with patients with either high ERG and/or high BAALC expression (Appendix Table A1, online only; Fig 2). In an exploratory multivariable analysis the combined prognostic validity of ERG and BAALC expression was investigated. In this model, patients with low ERG/low BAALC expression had a significantly lower risk of relapse (HR: 0.23, P = .001) and death (HR: 0.4, P = .003), once adjusting for immunophenotypic subtype, the only other prognostic factor (Table 3).
As mentioned in the statistical section, the analyses so far have excluded patients receiving SCT. However, when patients undergoing allogeneic and autologous SCT were included in the analyses and censored at time of SCT, outcome analyses showed comparable results. Furthermore, no differences were seen in additional analyses that included all irrespective of the postremission therapy (data not shown).
Within the largest and prognostically most favorable immunophenotypic subgroup of thymic T-ALL (n = 87), 45 patients lacked high expression of both ERG and BAALC. For patients with thymic T-ALL low ERG/low BAALC expression identified patients with highly favorable RFS (P = .002) with 87% of patients remaining relapse free at 4 years, and superior OS (P = .003) with a 4-year OS of 77% (Fig 3).
The analyses of the predictive value of ERG and BAALC expression within the prognostically unfavorable subgroups of early and mature T-ALL were limited by the small sample size. Whereas there was no survival difference in mature T-ALL, in early T-ALL, low ERG/low BAALC status was associated with a more favorable outcome (one of four relapsed) compared with patients with high ERG and/or high BAALC expression (eight of eight relapsed).
Treatment optimization in acute leukemia requires the accurate assignment of patients at diagnosis to specific risk groups to guide subsequent risk-adapted treatment stratification. In T-ALL, few prognostic molecular markers have been identified.
In this study, we have evaluated the impact of expression of the gene BAALC in conjunction with other clinical and molecular factors to predict outcome in T-ALL, and have shown for the first time to our knowledge that BAALC expression is of prognostic relevance in adult T-ALL. High BAALC expression was of adverse prognostic significance with a dismal long-term survival of 18% at 4 years and RFS of only 21% at 4 years.
The gene BAALC was initially found to be of prognostic significance in AML with normal cytogenetics.16 Like expression of the oncogenic ETS transcription factor ERG, expression of BAALC showed high expression in the hematopoietic progenitor compartment. Moreover, ERG and BAALC, coregulated during normal hematopoietic maturation, were found to be nearest neighbors in a gene expression profile of cytogenetically normal AML patients.22 We hypothesized that expression of both BAALC and ERG might be of prognostic importance in T-ALL.
As we have shown, high BAALC expression and high ERG expression identifies T-ALL patients with a high risk of relapse and inferior survival. Moreover, the combined assessment of BAALC and ERG allowed a more precise risk assessment with expression of high ERG and/or high BAALC predicting adverse RFS (4-year RFS 33%) and OS (4-year OS 30%). These high-risk patients may benefit from more intensive consolidation regimens including SCT. Encouraging results of allogeneic SCT in first CR have been reported in a retrospective study with a RFS of 74% at 3 years.23 Allogeneic SCT as part of postremission therapy has been shown to improve survival in high-risk adult ALL patients.24,25 However, allogeneic SCT is still associated with high treatment-related mortality.26 Thus, the identification of low-risk patients for whom these intensive regimens can be withheld is of major clinical importance.
Here, we demonstrate that low expression of both ERG and BAALC identifies T-ALL patients with a distinctly favorable outcome, with 81% of patients who receive standard therapy remaining relapse-free at 4 years. Moreover, the predictive value of ERG and BAALC expression was independent of the only other prognostic factor, the immunophenotypic subgroup. Risk factors for T-ALL patients treated within the GMALL protocols with proven adverse prognostic significance had been the immunophenotypic subtypes (early and mature T-ALL) and HOX11L2 positivity.24 In the current analysis including the new molecular markers ERG and BAALC as well as relevant clinical characteristics (age, WBC, > one course to CR) and molecular markers (HOX11, HOX11L2, immunophenotypes), only the immunophenotypic subgroup and expression of ERG and BAALC were of independent prognostic relevance. Within thymic T-ALL, the largest immunophenotypic subgroup, the molecular status of low ERG/low BAALC (representing 52% of thymic T-ALL patients) maintained its favorable prognosis with an exceptional 4-year RFS of 87%. Thus low ERG/low BAALC expression identifies a large subgroup of T-ALL patients with a remarkably good long-term outcome achieved by chemotherapy alone.
The increasing knowledge of molecular alterations will unravel novel pathways involved in leukemogenesis, and the identification of prognostically relevant molecular markers may facilitate more sophisticated risk-adapted treatment strategies. Furthermore, insights into altered pathways may be the basis for the development of new targeted therapies.24,27,28 In this respect, the intriguing finding that ERG and BAALC expression are of prognostic significance in AML as well as T-ALL underscores their potential involvement in myeloid and T-lineage leukemogenesis. We hypothesize that, because both genes are highly expressed in the normal hematopoietic progenitor compartment, high expression in AML as well as T-ALL signifies a stem cell–like disease with an immature progenitor cell as cell of origin. The association of ERG and BAALC with a more immature phenotype in both leukemic subtypes,8,13,16 and coexpression of myeloid markers in T-ALL with high BAALC expression may indicate a bilineage involvement of the leukemic blasts. Whereas in the multivariable analysis neither CD34 positivity nor presence of myeloid markers were of prognostic impact, the molecular status of BAALC and ERG expression revealed independent prognostic importance. Thus, BAALC and ERG may play a mechanistic role that contributes to the more aggressive leukemic phenotype in AML and T-ALL.
For high-risk T-ALL and AML patients, alternative treatment strategies are warranted to improve outcome. Further insights into the molecular pathways associated with risk factors including BAALC and ERG may provide guidance for the development of novel targeted drugs. Because the failure of current therapeutic regimens to achieve long-term relapse-free survival is likely related to chemotherapy-resistance, ERG- and/or BAALC-expressing leukemic blasts may contribute to drug resistance. Interestingly, BAALC and ERG were also found to be significantly overexpressed in the more aggressive BCR-ABL–positive compared with BCR-ABL–negative acute lymphoblastic leukemia.29
To the contrary, low expression of BAALC and ERG may identify a leukemic subtype with a high sensitivity to chemotherapeutic agents resulting in a favorable long-term relapse-free survival. Remarkably, a similar observation was found in AML, where the only patients with a long-term survival rate greater than 50% were patients with low expression of both ERG and BAALC.13
In conclusion, this study demonstrates the prognostic value of the combined assessment of ERG and BAALC expression in adult T-ALL patients. The low ERG/low BAALC expression status identifies a highly favorable risk group for which postremission therapy with SCT may not be indicated. Independent studies should be performed to confirm that the combined assessment of BAALC and ERG expression can improve treatment stratification in adult T-ALL.
The author(s) indicated no potential conflicts of interest.
Conception and design: Claudia D. Baldus, Wolf K. Hofmann
Financial support: Claudia D. Baldus, Eckhard Thiel, Wolf K. Hofmann
Administrative support: Claudia D. Baldus, Nicola Gökbuget, Dieter Hoelzer, Eckhard Thiel, Wolf K. Hofmann
Provision of study materials or patients: Stefan Schwartz, Nicola Gökbuget, Dieter Hoelzer
Collection and assembly of data: Peter Martus, Thomas Burmeister, Stefan Schwartz, Nicola Gökbuget, Dieter Hoelzer, Wolf K. Hofmann
Data analysis and interpretation: Claudia D. Baldus, Peter Martus, Thomas Burmeister, Clara D. Bloomfield, Eckhard Thiel, Wolf K. Hofmann
Manuscript writing: Claudia D. Baldus, Clara D. Bloomfield, Wolf K. Hofmann
Final approval of manuscript: Claudia D. Baldus, Peter Martus, Stefan Schwartz, Nicola Gökbuget, Clara D. Bloomfield, Eckhard Thiel, Wolf K. Hofmann
Fig 1. Impact of BAALC expression on survival. Kaplan-Meier analyses of (A) relapse-free survival (RFS) and (B) overall survival (OS) showing a significantly inferior outcome for patients with high BAALC expression compared with patients with low BAALC expression. Patients undergoing stem cell transplantation as consolidation treatment were excluded for the analyses of OS and RFS.
Fig 2. Impact of ERG in combination with BAALC expression on survival. Kaplan-Meier analyses of (A) relapse-free survival (RFS) and (B) overall survival (OS) showing a favorable outcome for patients with low expression of ERG and BAALC (low ERG/low BAALC) compared with patients with high ERG and/or high BAALC expression (combined group). Patients undergoing stem cell transplantation as consolidation treatment were excluded for the analyses of OS and RFS.
Fig 3. Impact of ERG in combination with BAALC expression in thymic acute T-lymphoblastic leukemia (T-ALL). Kaplan-Meier analyses of (A) relapse-free survival (RFS) and (B) overall survival (OS) showing a favorable outcome for patients with low expression of ERG and BAALC (low ERG/low BAALC) compared with patients with high ERG and/or high BAALC expression (combined group) in thymic T-ALL. Patients undergoing stem cell transplantation as consolidation treatment were excluded for the analyses of OS and RFS.
Fig A1. Impact of BAALC expression (in quartiles [Q]) on survival. Kaplan-Meier analysis of relapse-free survival (RFS) according to BAALC expression categorized in Q1 to Q4 of expression levels; patients in the upper Q4 were defined as high BAALC expressers. Patients undergoing stem-cell transplantation as consolidation treatment were excluded for the analysis of RFS.
Fig A2. Impact of ERG in combination with BAALC expression on survival. Kaplan-Meier analyses of (A) relapse-free survival (RFS) and (B) overall survival (OS) showing a favorable outcome for patients with low expression of ERG and BAALC (low ERG/low BAALC) compared with the other subgroups. Patients undergoing stem-cell transplantation as consolidation treatment were excluded for the analyses of OS and RFS.
|
| Characteristic | BAALC Low (n = 111) | BAALC High (n = 37) | P |
|---|---|---|---|
| WBC ( × 109/L) | .31 | ||
| Median | 40.6 | 25.8 | |
| Range | 1-853 | 1-341 | |
| Age, years | .98 | ||
| Median | 29 | 30 | |
| Range | 16-65 | 17-59 | |
| Sex, % | .01 | ||
| Male | 79 | 56 | |
| Mediastinal mass | .12 | ||
| No. yes | 84 | 22 | |
| Total No. | 107 | 34 | |
| % | 78 | 65 | |
| CNS involvement | .04 | ||
| No. yes | 14 | 0 | |
| Total No. | 103 | 31 | |
| % | 14 | 0 | |
| HOX11 (n = 141) | .11 | ||
| No. positive | 45 | 10 | |
| Total No. | 102 | 35 | |
| % | 44 | 28 | |
| HOX11L2 (n = 141) | .5 | ||
| No. positive | 8 | 4 | |
| Total No. | 102 | 35 | |
| % | 8 | 11 | |
| ERG expression (n = 147) | .03 | ||
| No. high | 50 | 24 | |
| Total No. | 111 | 36 | |
| % | 45 | 67 | |
| Early T-ALL (n = 31) | < .0001* | ||
| No. | 14 | 17 | |
| % | 13 | 46 | |
| Thymic T-ALL (n = 87) | |||
| No. | 75 | 12 | |
| % | 68 | 32 | |
| Mature T-ALL(n = 33) | |||
| No. | 22 | 8 | |
| % | 26 | 22 | |
| CD34 positivity | < .0001 | ||
| No. positive | 25 | 23 | |
| Total No. | 111 | 37 | |
| % | 22 | 62 | |
| Myeloid markers | .0001 | ||
| No. present | 19 | 18 | |
| Total No. | 88 | 28 | |
| % | 22 | 64 | |
| CR | .03 | ||
| No. | 99 | 27 | |
| % | 89 | 73 | |
| Relapse† | |||
| No. | 23 | 12 | .003 |
| Total No. | 70 | 17 | |
| % | 31 | 71 |
Abbreviations: T-ALL, acute T-lymphoblastic leukemia; CR, complete remission.
*Overall P value comparing the frequency of early, thymic, and mature T-ALL between BAALC low and high groups.
†Patients receiving stem cell transplantation in first CR and patients taken off protocol were not assessed for the relapse rate.
|
| Characteristic | Low ERG/Low BAALC | High ERG and/or High BAALC (combined group) | P |
|---|---|---|---|
| No. of patients | 61 | 86 | |
| WBC, × 109/L | .4 | ||
| Median | 40.0 | 36.4 | |
| Sex, % | .01 | ||
| Male | 85 | 66 | |
| Early T-ALL (n = 30) | .006 | ||
| No. | 6 | 24 | |
| % | 10 | 28 | |
| Thymic T-ALL (n = 87) | |||
| No. | 45 | 42 | |
| % | 74 | 49 | |
| Mature T-ALL (n = 30) | |||
| No. | 10 | 20 | |
| % | 16 | 23 | |
| CD34 positivity | .002 | ||
| No. positive | 11 | 36 | |
| Total No. | 61 | 86 | |
| % | 18 | 42 | |
| Myeloid markers | .005 | ||
| No. present | 7 | 30 | |
| Total No. | 47 | 69 | |
| % | 15 | 43 | |
| CR | .5 | ||
| No. | 52 | 70 | |
| % | 85 | 81 | |
| Relapse† | < .0001 | ||
| No. | 7 | 27 | |
| Total No. | 42 | 45 | |
| % | 17 | 60 |
Abbreviations: T-ALL, acute T-lymphoblastic leukemia; CR, complete remission.
*Overall P value comparing the frequency of early, thymic, and mature T-ALL between BAALC low and high groups.
†Patients receiving stem cell transplantation in first CR and patients who were taken off protocol were not assessed for the relapse rate.
|
| End Point Variable* | Risk of Relapse (RFS) | Risk of Death (OS) | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| HR | 95% CI | P | HR | 95% CI | P | |||||
| ERG/BAALC | ||||||||||
| Low ERG/low BAALC v combined group | 0.23 | 0.1 to 0.5 | .001 | 0.40 | 0.2 to 0.7 | .003 | ||||
| Immunophenotype | ||||||||||
| Thymic v early | 0.23 | 0.1 to 0.5 | .0005 | 0.30 | 0.2 to 0.6 | .0002 | ||||
| Thymic v mature | 0.42 | 0.2 to 1.0 | .055 | 0.46 | 0.2 to 0.9 | .025 | ||||
Abbreviations: RFS, relapse-free survival; OS, overall survival; HR, hazard ratio.
*Variables considered for model inclusion are listed in the statistical analysis section.
|
| Survival | Low ERG/Low BAALC | High ERG and/or High BAALC (combined group) | P |
|---|---|---|---|
| RFS | < .0001 | ||
| No. of patients | 46 | 48 | |
| Median (months) | NR | 25 | |
| % RFS at 4 years | 81 | 33 | |
| 95% CI | 68 to 94 | 17 to 49 | |
| OS | .0001 | ||
| No. of patients | 52 | 64 | |
| Median (months) | NR | 26 | |
| % alive at 4 years | 69 | 30 | |
| 95% CI | 55 to 82 | 18 to 43 |
Abbreviations: T-ALL, acute T-lymphoblastic leukemia; RFS, relapse-free survival; NR, not reached.
published online ahead of print at www.jco.org on July 23, 2007.
Supported in part by grants from the Deutsche Krebshilfe (Max Eder Nachwuchsförderung) and Deutsche Forschungsgesellschaft (DFG BA 3363/1-1; C.D. Baldus); and the Leukemia Clinical Research Foundation (C.D. Bloomfield).
Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
We thank Alexandra Bittroff-Leben, Verena Serbent, Cornelia Schlee, Regina Reutzel, and Barbara Komischke for their excellent technical assistance.
| 1. | Hoelzer D, Goekbuget N, Ottmann O, et al: Acute lymphoblastic leukemia. Hematology Am Soc Hematol Educ Program 162,2002-192, Medline, Google Scholar |
| 2. | Armstrong SA, Look AT: Molecular genetics of acute lymphoblastic leukemia. J Clin Oncol 23::6306,2005-6315, Link, Google Scholar |
| 3. | Mrozek K, Bloomfield CD: Chromosome aberrations, gene mutations and expression changes, and prognosis in adult acute myeloid leukemia. Hematology Am Soc Hematol Educ Program 169,2006-177, Medline, Google Scholar |
| 4. | Thomas X, Boiron JM, Huguet F, et al: Outcome of treatment in adults with acute lymphoblastic leukemia: Analysis of the LALA-94 trial. J Clin Oncol 22::4075,2004-4086, Crossref, Medline, Google Scholar |
| 5. | Haferlach T, Kern W, Schoch C, et al: A new prognostic score for patients with acute myeloid leukemia based on cytogenetics and early blast clearance in trials of the German AML Cooperative Group. Haematologica 89::408,2004-418, Medline, Google Scholar |
| 6. | Yanada M, Jinnai I, Takeuchi J, et al: Clinical features and outcome of T-lineage acute lymphoblastic leukemia in adults: A low initial white blood cell count, as well as a high count predict decreased survival rates. Leuk Res 31::907,2007-914, Crossref, Medline, Google Scholar |
| 7. | Vitale A, Guarini A, Ariola C, et al: Adult T-cell acute lymphoblastic leukemia: Biologic profile at presentation and correlation with response to induction treatment in patients enrolled in the GIMEMA LAL 0496 protocol. Blood 107::473,2006-479, Crossref, Medline, Google Scholar |
| 8. | Baldus CD, Burmeister T, Martus P, et al: High expression of the ETS transcription factor ERG predicts adverse outcome in acute T-lymphoblastic leukemia in adults. J Clin Oncol 24::4714,2006-4720, Link, Google Scholar |
| 9. | Maroulakou IG, Bowe DB: Expression and function of Ets transcription factors in mammalian development: A regulatory network. Oncogene 19::6432,2000-6442, Crossref, Medline, Google Scholar |
| 10. | Anderson MK, Hernandez-Hoyos G, Diamond RA, et al. Precise developmental regulation of Ets family transcription factors during specification and commitment to the T cell lineage. Development 126::3131,1999-3133 148, Medline, Google Scholar |
| 11. | Kong XT, Ida K, Ichikawa H, et al: Consistent detection of TLS/FUS-ERG chimeric transcripts in acute myeloid leukemia with t(16;21)(p11;q22) and identification of a novel transcript. Blood 90::1192,1997-1199, Medline, Google Scholar |
| 12. | Baldus CD, Liyanarachchi S, Mrozek K, et al: Acute myeloid leukemia with complex karyotypes and abnormal chromosome 21: Amplification discloses overexpression of APP, ETS2, and ERG genes Proc Natl Acad Sci U S A 101::3915,2004-3920, Crossref, Medline, Google Scholar |
| 13. | Marcucci G, Baldus CD, Ruppert AS, et al: Overexpression of the ETS-related gene, ERG, predicts a worse outcome in acute myeloid leukemia with normal karyotype: A Cancer and Leukemia Group B study. J Clin Oncol 23::9234,2005-9242, Link, Google Scholar |
| 14. | Tanner SM, Austin JL, Leone G, et al: BAALC, the human member of a novel mammalian neuroectoderm gene lineage, is implicated in hematopoiesis and acute leukemia. Proc Natl Acad Sci U S A 98::13901,2001-13906, Crossref, Medline, Google Scholar |
| 15. | Baldus CD, Tanner SM, Kusewitt DF, et al: A. BAALC, a novel marker of human hematopoietic progenitor cells. Exp Hematol 31::1051,2003-1056, Medline, Google Scholar |
| 16. | Baldus CD, Tanner SM, Ruppert AS, et al: BAALC expression predicts clinical outcome of de novo acute myeloid leukemia patients with normal cytogenetics: A Cancer and Leukemia Group B study. Blood 102::1613,2003-1618, Crossref, Medline, Google Scholar |
| 17. | Hoelzer D, Arnold R, Buechner T, et al: Characteristics, outcome and risk factors in adult T-lineage acute lymphoblastic leukemia (ALL). Blood 94:, 1999 (abstr 2926a) Google Scholar |
| 18. | Brüggemann M, Raff T, Flohr T, et al: Clinical significance of minimal residual disease quantification in adult patients with standard risk acute lymphoblastic leukemia. Blood 107::1116,2006-1123, Crossref, Medline, Google Scholar |
| 19. | Bene MC, Castoldi G, Knapp W, et al: Proposals for the immunological classification of acute leukemias. Leukemia 9::1783,1995-1786, Medline, Google Scholar |
| 20. | Schwartz S, Rieder H, Schläger B, et al: Expression of the human homologue of rat NG2 in adult acute lymphoblastic leukemia: Close association with MLL rearrangement and a CD10(-)/CD24(-)/CD65s(+)/CD15(+) B-cell phenotype. Leukemia 17::1589,2003-1595, Crossref, Medline, Google Scholar |
| 21. | Baldus CD, Thiede C, Soucek S, et al: BAALC expression and FLT3 internal tandem duplication mutations in acute myeloid leukemia patients with normal cytogenetics: Prognostic implications. J Clin Oncol 24::790,2006-797, Link, Google Scholar |
| 22. | Baldus CD, Radmacher M, Marcucci G, et al: Co-expression of BAALC with ERG, TCF4, and CD133 indicates a shared pathway in acute myeloid and lymphoblastic leukemia. Blood 106:, 2005 (abstr 773a) Google Scholar |
| 23. | Thomas X, Danaila C, Le QH, et al: Long-term follow-up of patients with newly diagnosed adult acute lymphoblastic leukemia: A single institution experience of 378 consecutive patients over a 21-year period. Leukemia 15::1811,2001-1822, Crossref, Medline, Google Scholar |
| 24. | Gokbuget N, Hoelzer D. Treatment of adult acute lymphoblastic leukemia. Hematology Am Soc Hematol Educ Program 133,2006-141, Medline, Google Scholar |
| 25. | Yanada M, Matsuo K, Suzuki T, Naoe T: Allogeneic hematopoietic stem cell transplantation as part of postremission therapy improves survival for adult patients with high-risk acute lymphoblastic leukemia: A metaanalysis. Cancer 106::2657,2006-2663, Crossref, Medline, Google Scholar |
| 26. | Pui CH, Pei D, Sandlund JT, et al: Risk of adverse events after completion of therapy for childhood acute lymphoblastic leukemia. J Clin Oncol 23::7936,2005-7941, Link, Google Scholar |
| 27. | Lowenberg B: On the road to new drugs in acute myeloid leukemia. J Clin Oncol 25::1,2007-2, Link, Google Scholar |
| 28. | Downing JR, Mullighan CG. Tumor-specific genetic lesions and their influence on therapy in pediatric acute lymphoblastic leukemia. Hematology Am Soc Hematol Educ Program 118,2006-122, Medline, Google Scholar |
| 29. | Juric D, Lacayo NJ, Ramsey MC, et al: Differential gene expression patterns and interaction networks in BCR-ABL-positive and -negative adult acute lymphoblastic leukemias. J Clin Oncol 25::1341,2007-1349, Link, Google Scholar |
