Welcome international healthcare professionals

This site is no longer supported and will not be updated with new content. You are welcome to browse and download all content already included in the site. Please note you will have to register your email address to access the site.

You are here

EUTOS score predicts survival and cytogenetic response in patients with chronic phase chronic myeloid leukemia treated with first-line imatinib

Leukemia Research

Highlights

 

  • We validate Sokal, Euro, and EUTOS scores on 220 Chinese CP-CML patients.
  • EUTOS score could stratify CP-CML patients into 2 risk groups in OS and PFS significantly.
  • EUTOS score could distribute CP-CML patients into 2 risk groups in cumulative incidence of CCyR significantly by Fine and Gray model.
  • EUTOS score could predict the duration of first CCyR.
  • Low EUTOS index predicted for the achievement of CCyR.

Abstract

Sokal, Euro and newly developed EUTOS scoring systems were validated in 220 Chinese chronic phase chronic myeloid leukemia (CP-CML) patients treated with frontline imatinib. In the EUTOS low-risk and high-risk groups, the 5-year OS was 98.7% vs. 71.4% (P < 0.0001), and the 5-year cumulative incidence of complete cytogenetic response (CCyR) was 92.4% vs. 53.8% (P < 0.0001). EUTOS score also predicted progression-free survival and duration of CCyR. Low EUTOS index predicted for CCyR. However, Sokal and Euro scores mainly could not discriminate the intermediate-risk from high-risk group in either survival or CCyR. EUTOS score forecasts the prognosis of CP-CML patients treated with first-line imatinib.

Abbreviation: EUTOS - European Treatment and Outcome Study.

Keywords: EUTOS, Chronic myeloid leukemia, Complete cytogenetic response, Overall survival, Progression-free survival, Prognosis.

1. Introduction

Chronic myeloid leukemia (CML) is a myeloproliferative neoplasm caused by BCR-ABL, a constitutively active tyrosine kinase generated as a result of the t(9;22)(q34;q11) [1] . BCR-ABL tyrosine kinase inhibitors (TKI), including imatinib, and the subsequent second-generation (2G) TKIs, have a marked effect on the prognosis of CML. With the first-line imatinib treatment, at 5 years, overall survival (OS) and cumulative complete cytogenetic response (CCyR) rate was 89–93% and 80–87%, respectively [2] and [3]. However, there were 25% patients discontinued imatinib because of either suboptimal response or intolerance [4] . The survival after the progression into advanced phases is significantly shorter, no matter which therapy is used.

In order to predict the outcome of the patients, Sokal score [5] was developed in the era of chemotherapy, as well as Euro score [6] was proposed in the time of interferon-alfa administration. In 2011, the European Leukemia Network (ELN) revisited the baseline data of 2060 CP-CML patients treated with imatinib-based regimens and established the European Treatment and Outcome Study (EUTOS) score which was only based on spleen size and blood basophil percentage prior to any treatment [7] . In these patients, the new score predicted both CCyR and progression-free survival (PFS) better than that of Sokal or Euro score [7] .

In the present study, we performed a single-center retrospective study to validate the effectiveness of Sokal, Euro and EUTOS scoring systems in predicting survival and cytogenetic response in a cohort of Chinese early CP-CML patients treated with first-line imatinib regimens.

2. Patients and methods

2.1. Study population and risk groups dividing

From June 2003 to October 2012, 220 Chinese patients who diagnosed as CML in CP in our hospital and received imatinib (Gleevec, Novartis Pharma) as first-line therapy were reviewed. The patients had endured a maximum of six months from diagnosis to the imatinib treatment. Patients who received any cytoreductive treatment except for hydroxyurea and/or interferon-alfa (be used less than 3 months) before imatinib were excluded from the study. These patients received imatinib 400 mg daily, and the dose was adjusted according to their tolerance and response in order to maintain imatinib at or greater than 300 mg daily.

Sokal, Euro and EUTOS scores were calculated based on the patients’ clinical records prior to any treatment. And the calculation of the three scores was performed on the ELN network site ( http://www.leukemia-net.org ) according to the published formulas ( Table 1 ) [5], [6], and [7]. Using the risk definitions of the three scoring systems ( Table 1 ), we divided the patients into each of the risk groups.

Table 1 Calculation forms of relative risk for Sokal, Euro, and EUTOS scores.

Scoring system Calculation Risk definition
Sokal score [5] Exp 0.0116 × (age − 43.4) + 0.0345 × (spleen − 7.51) + 0.188 × [(platelet count ÷ 700)2 − 0.563] + 0.0887 × (blast cells–2.1) Low risk: <0.8

Intermediate risk: 0.8–1.2

High risk: >1.2
Euro score [6] 0.666 when age ≥ 50 years + (0.042 × spleen) + 1.0956 when platelet count >1500 × 109 + (0.0584 × blast cells) + 0.20399 when basophils >3% + (0.0413 × eosinophils) × 100 Low risk: ≤780

Intermediate risk: 781–1480

High risk: > 1480
EUTOS score [7] Spleen × 4 + basophils × 7 Low risk: ≤87

High risk: >87

Age is in years. Spleen is in centimeters below the costal margin. Blast cells, esinophils, and basophils are in percent of peripheral blood differential. Exp: exponential function.

2.2. Cytogenetic analysis

Bone marrow aspiration for morphology and for cytogenetics was performed at baseline, every three or six months for the first year and then every six or twelve months in the following years. Cytogenetic response was measured in bone marrow cells and determined by metaphase analyses with R-banding technique after short-term culture.

2.3. Definitions

The definitions of CP, complete hematologic response (CHR), major cytogenetic response (MCyR), CCyR were made according to ELN recommendations [8] . OS and PFS were calculated from the start of imatinib therapy. The duration of CCyR was calculated from the date of the first CCyR to the date of CCyR loss or last cytogenetic evaluation, whichever came first.

2.4. Statistical analyses

Survival probabilities and duration of CCyR were estimated by the Kaplan–Meier method and compared by the log-rank test. By contrast, in estimating the cumulative incidence (CI) of CCyR, death was taken into account as competing risk [9] . The Fine and Gray model was used and the Gray test was applied for group comparison [10] and [11]. Univariate and multivariate analyses were performed to identify potential factors for prediction of CCyR. Multivariate analysis was performed using the logistic regression model. For survival, duration of CCyR, and CI of CCyR analyses, patients were censored at the time of stem cell transplantation. P-value < 0.05 was considered statistically significant. Calculations were carried out using SAS Version 9.3 software (Cary, NC, USA) and R Version 3.0.2 software.

2.5. Ethics

The protocol followed the Declaration of Helsinki and was approved by the ethical advisory board of Institute of Hematology and Blood Diseases Hospital. Informed consent was obtained from all patients before they entered the study.

3. Results

3.1. Treatments and responses

The median time from diagnosis to imatinib therapy was 0.5 month (interquartile range: 0.5–1.5 months). The median follow-up was 42 months (interquartile range: 19–65 months), 1 patient was lost to follow-up. During follow-up, 34 patients (15.4%) discontinued imatinib after a median time of 15.3 months (interquartile range: 9–38 months). Reasons of discontinuation included the loss of CHR or progression to advanced phases (n = 14), failure to achieve MCyR (n = 15), failure to achieve major molecular response (n = 4), and finding suitable stem cell donor (n = 1). After discontinued imatinib, 31 patients received the niluotinib (n = 21), dasatinib (n = 8), bosutinib (n = 2) or an allogeneic stem-cell transplantation (n = 3). In addition, 3 patients underwent transplantations after failure to 2G-TKIs. Clinical characteristics of all patients were detailed in Table 2 .

Table 2 The patient clinical characteristics (n = 220).

Characteristics No. %
Median age, years (range) 39.5 (15–69)
Gender: male 137 62.3
Splenomegaly 162 73.6
Spleen size ≥10 cm below the costal margin 76 34.5
Median leukocytes ×109/l (range) 133.3 (5.1–554.2)
Median Hb g/l (range) 113.5 (53–178)
Median platelets ×109/l (range) 419.5 (5–1867)
Median basophils in PB % (range) 4 (0–17)
Median eosinophils in PB % (range) 2 (0–16)
Median blast cells in PB % (range) 0 (0–12)

PB: peripheral blood; No.: number of patients.

The 5-year probabilities of OS, PFS were 91.9% (95% confidence interval: 87.6–96.2%), and 87.4% (95% confidence interval: 81.8–93.1%), respectively. And the 5-year CI of CCyR was 82.9% (95% confidence interval: 82.8–83.1%).

3.2. OS, PFS according to the 3 scoring systems

According to Sokal score, 101 (45.9%), 79 (35.9%) and 40 (18.2%) patients were in the low-, intermediate- and high-risk group, respectively. We found that in the case of OS, Sokal score stratified the intermediate- and high-risk groups significantly (p = 0.034), but failed to stratify the low- and intermediate-risk groups (p = 0.106; Fig. 1 A). In the case of PFS, Sokal score could discriminate the patients of 3 risk groups significantly (p = 0.012 for the low- and intermediate-risk groups, p = 0.046 for the intermediate- and high-risk groups; Fig. 1 D). The 5-year OS was 97.1% for the low-risk group, 91.2% for the intermediate-risk group and 80.8% for the high-risk group ( Table 3 ).

gr1

Fig. 1 OS of patients according to Sokal score (A), to Euro score (B), and to EUTOS score (C); PFS of patients according to Sokal score (D), to Euro score (E), and to EUTOS score (F).

Table 3 The distribution results according to the considered risk stratifications.

fx1
 

OS: overall survival; PFS: progression-free survival; No.: number of patients; CCyR: complete cytogenetic response; CI: cumulative incidence.

In terms of Euro score, low-risk score was observed in 110 (50%) patients, intermediate-risk score in 86 (39.1%) patients, high-risk score in 24 (10.8%) patients. Euro score resulted in statistically significant difference between the low-risk and intermediate-risk groups in both OS (p = 0.0002) and PFS (p < 0.0001), but not between the intermediate-risk and high-risk groups in OS (p = 0.897) and PFS (p = 0.932, Fig. 1 B and E). The 5-year OS were 98.4%, 84.7% and 86.5% for the low-, intermediate-, and high-risk group, respectively ( Table 3 ).

By contrast, EUTOS score stratified patients into 2 risk groups in both OS and PFS with obvious statistical significance (p < 0.0001 for OS; p < 0.0001 for PFS, Fig. 1 C and F). Low-risk score was observed in 165 (75%) patients, high-risk score in 55 (25%) patients. The 5-year OS were 98.7% and 71.4% for the low- and high-risk groups ( Table 3 ).

3.3. Cumulative incidence of CCyR according to the 3 scoring systems

Considering Sokal score, the 5-year CI of CCyR was 95% for the low-risk group, 78.2% for the intermediate-risk group, and 58.9% for the high-risk group. The CI of CCyR curves of the low- and intermediate-risk groups differed significantly with a p-value of 0.002; but it showed no difference between the intermediate- and high-risk groups (p = 0.053; Table 3 and Fig. 2 A).

gr2

Fig. 2 Cumulative incidences of CCyR according to Sokal score (A), to Euro score (B), and to EUTOS score (C).

According to Euro score, the 5-year CI of CCyR was 93.6%, 73.4% and 63.7% for the low-, intermediate-, and high-risk groups, respectively ( Table 3 ). Although Euro score stratified the low- and intermediate-risk groups significantly (p = 0.0005), it failed again to discriminate the intermediate- and high-risk groups (p = 0.517; Fig. 2 B).

As for EUTOS score, the low-risk group had significant advantage in achieving CCyR over the whole observation time compared with the high-risk group (p < 0.0001, Fig. 2 C). The 5-year CI of CCyR was 92.4% for the low-risk group and 53.8% for the high-risk group ( Table 3 ). Interestingly, we found that all the CCyR curves of the 3 scoring systems seemed stable after 2 years of imatinib treatment.

3.4. Duration of CCyR according to the 3 scoring systems

To evaluate whether the prognostic scores can be used for prediction of duration of CCyR, we investigated the CCyR duration curves based on the 3 scoring systems. Regarding Sokal score, the low-risk group showed a longer CCyR than the intermediate-risk group ( Fig. 3 A). The ratios of maintaining the CCyR of the low- and intermediate-risk groups were 85.4% and 75.8% (p = 0.022; Table 3 ). Nevertheless, there was no difference between the intermediate- and high-risk groups (p = 0.226; Table 3 and Fig. 3 A).

gr3

Fig. 3 Probabilities of maintaining the CCyR of patients according to Sokal score (A), to Euro score (B), and to EUTOS score (C).

According to Euro score, the intermediate-risk group showed shorter CCyR duration as compared with the neighboring low-risk group (ratio of maintaining CCyR, 71.6% vs. 85.6%, p = 0.002; Table 3 ), but no difference was found when compared with the high-risk group (ratio of maintaining CCyR, 71.6% vs. 81.2%, p = 0.56; Table 3 and Fig. 3 B).

As far as EUTOS score was concerned, patients belonging to the low-risk had notably longer duration of CCyR than the high-risk groups (Ratio of maintaining CCyR, 87% vs. 52.6%, p < 0.0001; Table 3 and Fig. 3 C).

3.5. Factors predictive for CCyR

Finally, we analyzed the pretherapy prognostic factors for the achievement of CCyR by univariate and multivariate analyses. In univariate analyses, factors associated with CCyR were: EUTOS score (p < 0.0001), Sokal score (p = 0.002), Euro score (p = 0.050), hemoglobin level (p = 0.001), peripheral blood basophils (p = 0.004), peripheral blood blasts (p = 0.004) and spleen size (p = 0.006). Nevertheless, in a multivariate analysis, low EUTOS score was the only significant factor predicting CCyR (OR = 0.206, 95% confidence interval: 0.094–0.453, p < 0.0001).

4. Discussion

In CML, many baseline factors have been reported to influence the response to TKIs and survival, such as clonal chromosome abnormalities in Ph+ cells, specific multidrug resistance polymorphisms and the detailed molecular dissection of the genome [8], [12], [13], and [14]. However, these data were limited in research and have not been performed in daily clinical practice. Hence, the prognostic evaluation at diagnosis is still based on the clinical features. In the past three decades, risk stratification for CML patients primarily relied on Sokal and Euro scores which were developed in the pre-TKI era. These two scores were also proved to be valuable in predicting prognosis in TKI-treated patients [4], [15], and [16]. During the TKI era, EUTOS score was proposed based on more than 2000 CP-CML patients treated with imatinib-based regimens. The new score was generated to stratify CP-CML patients with different cumulative probability of achieving CCyR within 18 months and different 5-year PFS [7] .

In the present study, the EUTOS score was predictive for OS and PFS among patients in CP-CML treated with imatinib. The results are in agreement with other groups [7], [17], [18], [19], [20], [21], and [22]. Conversely, Sokal score failed to stratify the low-risk patients from the intermediate-risk patients in OS, while Euro score could not discriminate the intermediate-risk patients from the high-risk patients in OS and PFS.

We also found that the EUTOS score was highly predictive for CCyR. CI of CCyR was significantly stratified by EUTOS scoring system into two risk groups. By contrast, the intermediate- and high-risk patients of Sokal and Euro scoring systems lacked statistical significance in the CI of CCyR. Moreover, our study showed that the curves of CI of CCyR in the 3 scoring systems seemed stable after 2 years of imatinib treatment, which was in agreement with the report from Pavlik et al. [23] . It suggested that after 2 years of regular imatinib therapy, the CCyR cumulative ratios of CML patients reach to a plateau. Regarding duration of CCyR, the group with low EUTOS score had a significant longer CCyR duration than the group with high EUTOS score, whereas the Sokal and Euro scores failed again to discriminate the intermediate- and high-risk patients. The results indicate that EUTOS score is not only able to predict the probability of CCyR, but also the duration of CCyR. Moreover, multivariate analysis identified low EUTOS score as the only factor associated with the achievement of CCyR.

In our study, the EUTOS high-risk group did not share with the Sokal and Euro high-risk groups. In detail, of the 55 EUTOS high-risk patients, 20 patients belonged to Sokal high-risk and 14 to Euro high-risk. Conversely, the most Euro high-risk patients (20/24) also belonged to the Sokal high-risk group. Age was an important determinant for the Sokal and Euro score, but it was not included in EUTOS score. In the era of TKIs, older age appeared to be not associated with a worse outcome [24] . That might be the reason that the prognostic power of Sokal and Euro scores partly reduced.

Since the publication of EUTOS score, at least 12 independent studies have investigated its prognostic power. The majority of the studies involving more than 4000 CP-CMP patients confirmed the new score [7], [17], [18], [19], [20], [21], [22], [25], and [26]. However, Hammersmith hospital and the Japan groups did not find EUTOS score predict the OS, PFS and CCyR, in 282 cases and in 145 cases treated with imatinib [27] and [28]. And MD Aderson cancer center in 465 cases receiving imatinib (400 or 800 mg daily) or 2G-TKIs reported that EUTOS score could predict CCyR, but not OS, event-free survival or transformation-free survival [29] . This discrepancy could be, at least in part, explained by the following reasons. First, it was reported that Asian CML patients commonly harbor the polymorphism affecting the sensitivity to TKIs and CML in Asia has an incidence rather lower than Western countries and tends to afflicting a younger population [14] and [30]. In this analysis, a high level of significance was found in EUTOS score which was consistent with the reports from Korea and Singapore [17] and [25]. It suggested that genetic polymorphisms and ethnical difference may exist in Asian CML patients. Secondly, statistical methods in assessment of CCyR differed in studies. Among the 12 studies, 4 studies [18], [19], [28], and [29] used Kaplan–Meier method and log-rank test, 4 studies [7], [20], and [23] including the original report by Hasford et al. and the present study used cumulative incidence model and Gray test, and other studies did not show the statistical method. Cumulative incidence model and Gray test allow for subdistribution of a competing risk to CCyR which is reported superior to standard survival models [9], [10], [11], and [31]. All the 4 reports using cumulative incidence model and Gray test showed significant different CI of CCyR according to EUTOS score. Meanwhile, among the 4 studies using Kaplan–Meier method and log-rank test, 2 of them showed no difference in CI of CCyR. Therefore, validation for CI of CCyR should be conducted using a homogenous statistical method.

However, this analysis has some shortcomings. Retrospective nature and a single-center report are major limitations of this project. Without consecutive CP-CML patients, the patients who could not afford the expense of imatinib would be excluded from the study. A larger number of patients from multi-center and prospective study may be needed to show significant difference in the future. On the other hand, nonadherence might be the vital reason to suboptimal response and is more prevalent than patients, physicians and family members believe [32] . Good adherence to TKIs is an important monitoring target [8] . Although assessment of adherence is not available in the study, it might also influence the results of the validation of the scoring systems.

In conclusion, our analysis indicates that EUTOS score predict survival and cytogenetic response in CP-CML patients treated with first-line imatinib. As far as we know, this is the first report that based on a series of Chinese patients.

Conflict of interest statement

Jianxiang Wang acts as consultant of Novartis and Bristol Myers Squibb. The other co-authors declare no competing financial interests.

Acknowledgements

The authors thank all the doctors, laboratorians and nurses who give the diagnoses and treatments to the CML patients in our hospital. We also thank the whole staff in the medical record department who works hard for preserving the precious medical records. And we acknowledge Prof. Min Wang for her critical reading of the manuscript.

This work was supported by National Public Health Grand Research Foundation (201202017), National Science Technology Major Project (2011ZX09302-007-04) of China and National Natural Science Foundation of China (81270635, 81370633).

Contributions. JW conceived the study; ZT performed the statistical analysis; ZT and BL collected data and wrote the paper. All authors treated the patients, reviewed and approved the manuscript.

References

  • [1] A. Quintas-Cardama, J. Cortes. Molecular biology of bcr-abl1-positive chronic myeloid leukemia. Blood. 2009;113:1619-1630 Crossref
  • [2] B.J. Druker, F. Guilhot, S.G. O’Brien, I. Gathmann, H. Kantarjian, N. Gattermann, et al. Five-year follow-up of patients receiving imatinib for chronic myeloid leukemia. N Engl J Med. 2006;355:2408-2417 Crossref
  • [3] U. Proetel, N. Pletsch, M. Lauseker, M.C. Muller, B. Hanfstein, S.W. Krause, et al. Older patients with chronic myeloid leukemia (≥65 years) profit more from higher imatinib doses than younger patients: a subanalysis of the randomized CML-Study IV. Ann Hematol. 2014;
  • [4] H. de Lavallade, J.F. Apperley, J.S. Khorashad, D. Milojkovic, A.G. Reid, M. Bua, et al. Imatinib for newly diagnosed patients with chronic myeloid leukemia: incidence of sustained responses in an intention-to-treat analysis. J Clin Oncol. 2008;26:3358-3363 Crossref
  • [5] J.E. Sokal, E.B. Cox, M. Baccarani, S. Tura, G.A. Gomez, J.E. Robertson, et al. Prognostic discrimination in good-risk chronic granulocytic leukemia. Blood. 1984;63:789-799
  • [6] J. Hasford, M. Pfirrmann, R. Hehlmann, N.C. Allan, M. Baccarani, J.C. Kluin-Nelemans, et al. A new prognostic score for survival of patients with chronic myeloid leukemia treated with interferon alfa. Writing Committee for the Collaborative CML Prognostic Factors Project Group. J Natl Cancer Inst. 1998;90:850-858
  • [7] J. Hasford, M. Baccarani, V. Hoffmann, J. Guilhot, S. Saussele, G. Rosti, et al. Predicting complete cytogenetic response and subsequent progression-free survival in 2060 patients with CML on imatinib treatment: the EUTOS score. Blood. 2011;118:686-692 Crossref
  • [8] M. Baccarani, M.W. Deininger, G. Rosti, A. Hochhaus, S. Soverini, J.F. Apperley, et al. European LeukemiaNet recommendations for the management of chronic myeloid leukemia: 2013. Blood. 2013;122:872-884 Crossref
  • [9] M. Pfirrmann, A. Hochhaus, M. Lauseker, S. Saussele, R. Hehlmann, J. Hasford. Recommendations to meet statistical challenges arising from endpoints beyond overall survival in clinical trials on chronic myeloid leukemia. Leukemia. 2011;25:1433-1438 Crossref
  • [10] R.J. Gray. A class of K-sample tests for comparing the cumulative incidence of a competing risk. Ann Stat. 1988;16:1141-1154
  • [11] J.P. Fine, R.J. Gray. A proportional hazards model for the subdistribution of a competing risk. J Am Stat Assoc. 1999;94:496-509
  • [12] S. Luatti, F. Castagnetti, G. Marzocchi, C. Baldazzi, G. Gugliotta, I. Iacobucci, et al. Additional chromosomal abnormalities in Philadelphia-positive clone: adverse prognostic influence on frontline imatinib therapy: a GIMEMA Working Party on CML analysis. Blood. 2012;120:761-767 Crossref
  • [13] S. Dulucq, S. Bouchet, B. Turcq, E. Lippert, G. Etienne, J. Reiffers, et al. Multidrug resistance gene (MDR1) polymorphisms are associated with major molecular responses to standard-dose imatinib in chronic myeloid leukemia. Blood. 2008;112:2024-2027 Crossref
  • [14] K.P. Ng, A.M. Hillmer, C.T. Chuah, W.C. Juan, T.K. Ko, A.S. Teo, et al. A common BIM deletion polymorphism mediates intrinsic resistance and inferior responses to tyrosine kinase inhibitors in cancer. Nat Med. 2012;18:521-528 Crossref
  • [15] G. Saglio, D.W. Kim, S. Issaragrisil, P. le Coutre, G. Etienne, C. Lobo, et al. Nilotinib versus imatinib for newly diagnosed chronic myeloid leukemia. N Engl J Med. 2010;362:2251-2259 Crossref
  • [16] A. Hochhaus, S.G. O’Brien, F. Guilhot, B.J. Druker, S. Branford, L. Foroni, et al. Six-year follow-up of patients receiving imatinib for the first-line treatment of chronic myeloid leukemia. Leukemia. 2009;23:1054-1061 Crossref
  • [17] H. Than, L. Kuan, C.H. Seow, W. Li, J.C. Allen Jr., C. Chuah. The EUTOS score is highly predictive for clinical outcome and survival in Asian patients with early chronic phase chronic myeloid leukemia treated with imatinib. ASH Annual Meeting Abstract, vol. 120 (, 2012) 3758
  • [18] B. Uz, Y. Buyukasik, H. Atay, E. Kelkitli, M. Turgut, O. Bektas, et al. EUTOS CML prognostic scoring system predicts ELN-based ‘event-free survival’ better than Euro/Hasford and Sokal systems in CML patients receiving front-line imatinib mesylate. Hematology. 2013;18:247-252 Crossref
  • [19] M. Tiribelli, M. Bonifacio, E. Calistri, G. Binotto, E. Maino, L. Marin, et al. EUTOS score predicts long-term outcome but not optimal response to imatinib in patients with chronic myeloid leukaemia. Leuk Res. 2013;37:1457-1460 Crossref
  • [20] V.S. Hoffmann, M. Baccarani, D. Lindoerfer, F. Castagnetti, A. Turkina, A. Zaritsky, et al. The EUTOS prognostic score: review and validation in 1288 patients with CML treated frontline with imatinib. Leukemia. 2013;27:2016-2022 Crossref
  • [21] M. Breccia, P. Finsinger, G. Loglisci, R. Latagliata, M. Mancini, A. Salaroli, et al. The EUTOS score identifies chronic myeloid leukeamia patients with poor prognosis treated with imatinib first or second line. Leuk Res. 2012;36:e209-e210 Crossref
  • [22] M. Tiribelli, M. Bonifacio, E. Calistri, G. Binotto, E. Maino, L. Marin, et al. EUTOS score identifies cases with poor outcome in patients with early chronic phase chronic myeloid leukemia, though not predictive for optimal response to imatinib. ASH Annual Meeting Abstract, vol. 120 (, 2012) 3778
  • [23] T. Pavlik, E. Janousova, J. Mayer, K. Indrak, M. Jarosova, H. Klamova, et al. Current survival measures reliably reflect modern sequential treatment in CML: correlation with prognostic stratifications. Am J Hematol. 2013;88:790-797 Crossref
  • [24] G. Gugliotta, F. Castagnetti, F. Palandri, M. Breccia, T. Intermesoli, A. Capucci, et al. Frontline imatinib treatment of chronic myeloid leukemia: no impact of age on outcome, a survey by the GIMEMA CML Working Party. Blood. 2011;117:5591-5599 Crossref
  • [25] S-A. Yahng, E-J. Jang, S.Y. Choi, Y-J. Oh, J-H. Bang, J-E. Park, et al. Comparison of Sokal, Hasford and EUTOS Scores in terms of long-term treatment outcome according to the risks in each prognostic model: a single center data analyzed in 255 early chronic phase chronic myeloid leukemia patients treated with frontline imatinib mesylate. ASH Annual Meeting Abstract, vol. 120 (, 2012) 2794
  • [26] M. Bonifacio, G. Binotto, E. Calistri, E. Maino, M. Tiribelli. Gruppo Triveneto LMC. EUTOS score predicts early optimal response to imatinib according to the revised 2013 ELN recommendations. Ann Hematol. 2014;93:163-164 Crossref
  • [27] D. Marin, A.R. Ibrahim, J.M. Goldman. European Treatment and Outcome Study (EUTOS) score for chronic myeloid leukemia still requires more confirmation. J Clin Oncol. 2011;29:3944-3945 Crossref
  • [28] E. Yamamoto, S. Fujisawa, M. Hagihara, M. Tanaka, K. Fujimaki, K. Kishimoto, et al. European Treatment and Outcome Study score does not predict imatinib treatment response and outcome in chronic myeloid leukemia patients. Cancer Sci. 2014;105:105-109
  • [29] E. Jabbour, J. Cortes, A. Nazha, S. O’Brien, A. Quintas-Cardama, S. Pierce, et al. EUTOS score is not predictive for survival and outcome in patients with early chronic phase chronic myeloid leukemia treated with tyrosine kinase inhibitors: a single institution experience. Blood. 2012;119:4524-4526 Crossref
  • [30] W.Y. Au, P.B. Caguioa, C. Chuah, S.C. Hsu, S. Jootar, D.W. Kim, et al. Chronic myeloid leukemia in Asia. Int J Hematol. 2009;89:14-23 Crossref
  • [31] H. Putter, M. Fiocco, R.B. Geskus. Tutorial in biostatistics: competing risks and multi-state models. Stat Med. 2007;26:2389-2430 Crossref
  • [32] L. Noens, M.A. van Lierde, R. De Bock, G. Verhoef, P. Zachee, Z. Berneman, et al. Prevalence, determinants, and outcomes of nonadherence to imatinib therapy in patients with chronic myeloid leukemia: the ADAGIO study. Blood. 2009;113:5401-5411 Crossref

Footnotes

a Department of Clinical Hematology, Institute of Hematology & Blood Diseases Hospital, Chinese Academy of Medical Science & Peking Union Medical College, Tianjin 300020, People's Republic of China

b Department of Hemopathology, Institute of Hematology & Blood Diseases Hospital, Chinese Academy of Medical Science & Peking Union Medical College, Tianjin 300020, People's Republic of China

c State Key Laboratory of Experimental Hematology, Institute of Hematology & Blood Diseases Hospital, Chinese Academy of Medical Science & Peking Union Medical College, Tianjin 300020, People's Republic of China

lowast Corresponding author at: Department of Clinical Hematology, Institute of Hematology & Blood Diseases Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, 288 Nanjing Road, Tianjin 300020, People's Republic of China. Tel.: +86 22 23909280; fax: +86 22 27232515.