Abstract
Background
The outcome of patients with relapsed acute myeloid leukemia (AML) remains unsatisfactory, with few available effective therapies. Increased understanding of the biology of the disease has led to the identification of novel therapeutic agents, several of which have been evaluated in recently conducted clinical trials.
Methods
We sought to determine whether the introduction of these agents as well as modern supportive care measures has already translated to better outcomes. We examined the outcomes of 1,056 patients with AML in first relapse treated between January 1993 and December 2013 at our institution.
Results
As previously reported, the independent prognostic factors for survival after first relapse, included age at relapse, cytogenetics, and duration of first complete remission (CR).
Conclusion
Upon multivariable analysis, treatment era was an independent predictor of survival, with significant improvement in overall survival between 2008–2013 as compared to prior time periods. Modern supportive care measures as well as participation in clinical trials of novel agents are already improving the outcomes in first relapse.
Keywords: acute myeloid leukemia, AML, relapse, outcomes
INTRODUCTION
Outcomes for most patients with acute myeloid leukemia (AML) remain poor.1 In particular, outcomes for patients with AML who experience relapse are unsatisfactory, in general.2 Although, approximately 60–80% of patients with AML achieve complete remission (CR), the majority will relapse, with the likelihood of attaining second or subsequent CR progressively diminished.3 Estey et al demonstrated that in patients with initial CR duration less than one year, the type of therapy received did not affect the overall outcome/survival; however, if the initial CR duration was greater than one year, there was a significant difference favoring those who received high dose cytarabine (HiDAC) containing regimens.4 At the extreme are patients who are refractory to their induction treatment who have a particularly dismal long-term prognosis.5 In this study we explored whether the introduction of modern supportive measures as well as novel non-myelosuppressive salvage strategies in the recent decade has translated to any improvements in the outcome.
METHODS
We reviewed the clinical data for all patients with AML who achieved CR at our institution from 1993–2013. Among 1,857 patients, 1,056 (57%) patients with available data on first relapse were identified during the study period. All patients signed consent forms for participation in research as approved by the Institutional Review Board of the University of Texas – MD Anderson Cancer Center. Four distinct time periods (treatment eras) were selected for comparison: 1993–1997, 1998–2002, 2003–2007, and 2008–2013. Demographic and clinical characteristics of the patients were examined and compared to determine predictors of long term outcome. Survival outcomes were evaluated using the Kaplan-Meier analysis and differences were compared using the log-rank test. Differences between cohorts were evaluated using the χ2 test for nominal variables and the Mann-Whitney U test and Fisher exact test for continuous values. Variables with a p-value ≤ 0.10 were included in a multivariate analysis.
RESULTS AND DISCUSSION
Table 1 shows the baseline demographics of the study population. The median age for the overall group (n=1,056) was 60 years (range 15–86 years). There were significant differences in terms of treatment era for median age at relapse (58 years in 1993–1997, 60 years in 1998–2002, 58 years in 2003–2007, and 62 years in 2008–2013, p-value 0.04) with the more recent cohort having an older population. There was a significant difference in terms of number of patients receiving allogeneic stem cell transplantation (SCT) in first CR (CR1) by treatment era (3 in 1993–1997, 8 in 1998–2002, 23 in 2003–2007, 41 in 2008–2013, p-value <0.001); Although there was no significant difference in terms of outcomes between the two groups with or without SCT in CR1 (p-value 0.55, Table 1b), it should be noted that the number of patients who underwent SCT in CR1 was small (6.8% of the cohort)
Table 1.
| a: Characteristics of patients with relapsed AML | |
|---|---|
| Characteristics | N=1056 (range or %) |
| Age, median (range) | 60 (15 – 86) |
| Male | 572 (54.2) |
| De novo or secondary AML | |
| De novo | 852 (80.7) |
| Secondary | 204 (19.3) |
| Period when relapse occurred | 181 (17.1) |
| 1993–1997 | 286 (27.1) |
| 1998–2002 | 268 (25.4) |
| 2003–2007 | 321 (30.4) |
| 2008–2013 | |
| Cytogenetic | |
| Favorable | 74 (7) |
| Intermediate | 652 (61.7) |
| Adverse | 259 (24.5) |
| IM/ND** | 71 (6.7) |
| Therapy received*** | |
| HDAC | 860 (81.4) |
| HMA | 52 (4.9) |
| Targeted | 43 (4.1) |
| Other/combinations | 1031 (97.6) |
| Type of response before relapse | |
| Complete | 986 (93.4) |
| Complete without platelet recovery | 70 (6.6) |
| Stem cell transplant in CR1 | |
| Yes | 72 (6.8) |
| No | 984 (93.2) |
| Duration of CR1 in months, median (range) | 8 (0.3 – 144.4) |
| Median survival after relapse, months (95% CI) | 5.85 (5.38 – 6.32) |
| Median follow up after relapse, months (95% CI)**** | 58.8 (50.7 – 67) |
| b: Univariate Analysis of Predictors of Survival | |||||
|---|---|---|---|---|---|
| Characteristics | N=1056 (%) | Events (%) | Median OS (months) | 95% CI | p-value |
| Age at relapse | ≤.001 | ||||
| ≤65 years | 685 (65) | 587 (86) | 6.6 | 5.95 – 7.25 | |
| >65 years | 371 (35) | 332 (90) | 4.2 | 3.50 – 4.98 | |
| Diagnosis* | .004 | ||||
| AML | 703 (67) | 612 (87) | 6.1 | 5.56 – 6.65 | |
| AMML | 196 (19) | 167 (85) | 6.1 | 4.95 – 7.33 | |
| AMOL | 92 (9) | 80 (87) | 3.4 | 1.94 – 4.96 | |
| AEL | 46 (4.4) | 42 (91.3) | 5.0 | 2.31 – 7.75 | |
| Mega | 19 (1.8) | 18 (94.7) | 3.9 | 0 – 8.43 | |
| De novo or secondary AML | ≤.001 | ||||
| De novo | 852 (80.6) | 732 (85.9) | 6.2 | 5.60 – 6.76 | |
| Secondary | 204 (19.4) | 187 (91.7) | 4.9 | 3.97 – 6 | |
| Period when relapse occurred | .006 | ||||
| 1993 – 1997 | 181 (17.1) | 176 (97.2) | 4.8 | 3.91 – 5.69 | |
| 1998 – 2002 | 286 (27.1) | 268 (93.7) | 5.7 | 4.55 – 6.81 | |
| 2003 – 2007 | 268 (25.4) | 241 (89.9) | 6.0 | 5.19 – 6.90 | |
| 2008 – 2013 | 321 (30.4) | 234 (72.8) | 6.7 | 5.86 – 7.61 | |
| Cytogenetics | ≤.001 | ||||
| Favorable | 74 (7) | 53 (71.6) | 13.9 | 10.52 – 17.28 | |
| Intermediate | 652 (61.7) | 555 (85.1) | 6.6 | 5.89 – 7.25 | |
| Adverse | 259 (24.5) | 249 (96.1) | 3.4 | 2.89 – 3.95 | |
| IM/ND** | 71 (6.7) | 62 (87.3) | 5.3 | 3.15 – 7.43 | |
| Therapy received*** | |||||
| HDAC | 860 (81.4) | 757 (88) | 6.08 | 5.52 – 6.64 | .05 |
| No HDAC | 196 (18.6) | 162 (82.6) | 5.06 | 4.10 – 6.02 | |
| HMA | 53 (5) | 46 (86.8) | 3.45 | 2.08 – 4.82 | .007 |
| No HMA | 1004 (95) | 873 (86.9) | 5.98 | 5.50 – 6.46 | |
| Targeted | 43 (4.1) | 30 (69.8) | 9.59 | 4.86 – 14.32 | .002 |
| No Targeted | 1013 (95.9) | 889 (87.7) | 5.75 | 5.26 – 6.24 | |
| Other/combination | 1031 (97.6) | 898 (87.1) | 5.88 | 5.41 – 6.35 | .28 |
| No Other | 25 (2.4) | 21 (84) | 4.37 | 1.66 – 7.08 | |
| Type of response before relapse | .05 | ||||
| Complete | 986 (93.4) | 860 (87.2) | 6.05 | 5.57 – 6.53 | |
| Complete without platelet recovery | 70 (6.6) | 59 (84.3) | 4.37 | 3.20 – 5.54 | |
| Stem cell transplant in CR1 | 0.55 | ||||
| No | 984 (93.2) | 862 (87.6) | 5.9 | 5.41 – 6.35 | |
| Yes | 72 (6.8) | 57 (79.2) | 5.2 | 3.40 – 6.99 | |
| Duration of CR1 | ≤0.001 | ||||
| <=8 months | 540 (51.1) | 499 (92.4) | 3.9 | 3.54 – 4.42 | |
| >8 months | 516 (48.9) | 420 (81.4) | 9.3 | 7.88 – 10.77 | |
| c: Multivariate Analysis of Predictors of Survival | |||
|---|---|---|---|
| Characteristics | p-value | HR | 95% CI |
| Age at relapse | ≤0.01 | 1.02 | 1.01–1.02 |
| Diagnosis* | |||
| AML (reference) | 1.00 | ||
| AMML | 0.19 | 1.13 | 0.94–1.36 |
| AMOL | ≤0.01 | 1.60 | 1.25–2.04 |
| AEL | 0.41 | 0.87 | 0.63–1.21 |
| Mega | 0.20 | 1.51 | 0.80–2.84 |
| Period when relapse occurred | ≤0.01 | ||
| 1993–1997 | ≤0.01 | 1.51 | 1.21–1.87 |
| 1998–2002 | 0.01 | 1.30 | 1.08–1.56 |
| 2003–2007 | 0.30 | 1.10 | 0.92–1.33 |
| 2008–2013 (reference) | 1.00 | ||
| Cytogenetics | ≤0.01 | ||
| Favorable (reference) | 1.00 | ||
| Intermediate | 0.05 | 1.34 | 0.10–1.81 |
| Adverse | ≤0.01 | 2.28 | 1.64–3.16 |
| Duration of CR1 | ≤0.01 | 0.98 | 0.97–0.98 |
Abbreviations
FAB=French American British AML classification
IM/ND=insufficient metaphase/not done
Treatment types are not mutually exclusive, HiDAC=high dose cytarabine (ARA-C), HMA=hypomethylating agent
Reverse Kaplan-Meier analysis
Abbreviations
Diagnosis type: AML=acute myeloid leukemia, other, AMML=acute myelomonocytic leukemia, AMOL=acute monocytic leukemia, AEL=acute erythroid leukemia, Mega=acute megakaryocytic leukemia
IM/ND=insufficient metaphase/not done
Treatment types are not mutually exclusive, HiDAC=high dose cytarabine (ARA-C), HMA=hypomethylating agent
Abbreviations
Diagnosis type: AML=acute myeloid leukemia, other, AMML=acute myelomonocytic leukemia, AMOL=acute monocytic leukemia, AEL=acute erythroid leukemia, Mega=acute megakaryocytic leukemia
Predictors of survival at the time of first relapse
Factors significant for predicting differences in survival included: age >65 years at relapse, AML subtype by FAB classification, secondary compared to de novo AML, treatment era, adverse cytogenetics, and CR1 duration greater or less than 12 months. These are, in general, confirmatory of relevant factors reported by other studies.6
Mutlivariate analysis of predictors of survival at first relapse
Significant variables (p-value < 0.10) identified in the univariate analysis were included in a multivariate analysis. Age >65 years at relapse; FAB morphological diagnoses of acute myelomonocytic leukemia, acute monocytic leukemia, and acute erythroid leukemia; as well as intermediate and adverse-risk cytogenetics, duration of CR1, and the era of treatment were independent predictors of survival after relapse. Survival was significantly better for patients relapsing in the era 2003–2013 compared to those relapsing 1993–2003 (6.3 months versus 5.3 months, p-value 0.004).(Figure 1a) Compared to patients relapsing between 2008–2013, patients relapsing between 1998–2002 and 1993–1997 had a survival disadvantage of 30% (HR 1.297; 95% CI= 1.076–1.564) and 50% (HR 1.505; 95% CI= 1.214–1.866), respectively. The difference with the 2003–2007 treatment era was not statistically significant. Figure 1b demonstrates survival differences by the treatment era when divided into 5-year intervals.
Figure 1.
a: Overall Survival, by Earliest and Most Recent Treatment Era
b: Overall Survival, by Treatment Era
We then performed multivariate analysis using Cox proportional hazard regression in the subgroup of patients treated with HiDAC-based therapy, which was the most frequently used therapy given throughout all of the treatment eras. The variables included were the same as the previous multivariate analysis. Importantly, this model confirmed the results of the previous analysis and found that compared to patients relapsing between 2008–2013, patients relapsing between 1998–2002 and 1993–1997 had a survival disadvantage of 33% (HR=1.332;95%CI= 1.071–1.657) and 57% (HR=1.566; 95% CI= 1.229–1.995), respectively. Again, the difference with the 2003–2007 treatment era was not statistically significant.
Breems et al6, studied outcomes in 1,540 patients with AML from HOVON/SAKK clinical trials from 1987–2001. Among 1108 who achieved CR, 667(60%) relapsed. They reported four independent predictors of overall survival after first relapse including unfavorable cytogenetics at diagnosis, duration of first CR, older age, and prior history of SCT. Similarly, Kurosawa et al examined the outcomes of patients in first relapse at their institution.7 Among 1,535 patients with AML who achieved CR, 1,015 (66%) relapsed, with about a half achieving a second CR (CR2). On multivariate analysis, duration of CR1 of one year or greater, achievement of CR2, and SCT as salvage therapy were independent predictors of survival. The authors concluded that attaining CR2 followed by the ability to undergo SCT represented the best chance of achieving long term survival.
Our results are consistent with prior analyses of patients with AML in first relapse, with age, cytogenetics and duration of CR1 as important predictors of survival in first relapse. In our study, the outcomes of patients who underwent SCT in CR1 was not significantly different from those who did not. However, it is important to note that the number of patients who underwent SCT in CR1 was small (6.8% of the cohort) thereby making this analysis limited in value. Importantly, we noted that era of treatment is an important predictor suggesting that more modern strategies of therapy as well as supportive care may already be impacting the outcomes of these patients. Of interest, this improvement in the more recent era could not be attributed to the patient characteristics. In fact, patients treated in the more recent era were significantly older, with no difference in the proportion of patients with adverse cytogenetics.
Several underlying factors may be responsible for the improvement in outcomes. First, limiting the more intensive chemotherapy-based induction regimens to the younger population is associated with less toxicity and the ability to tolerate subsequent salvage strategies.8 Our group has previously reported that, in patients with newly diagnosed AML, aged 65 years or older, treatment with hypomethylating agents is associated with similar outcomes compared to intensive chemotherapy.9. In that analysis of 671 patients treated between years 2000 and 2010 (with either hypomethylating agents or the more traditional intensive cytotoxic chemotherapy), there was no statistically significant difference in either the 2-year relapse free survival rates or the median overall survival between the two groups, suggesting that treatment with hypomethylating agents is associated with similar long term outcomes as intensive chemotherapy. Second, a number of novel targeted therapies have been introduced that may potentially be having positive influence on the outcomes, especially in older patients. For example, the availability of clinical trials of fms-like tyrosine kinase-3 (FLT3) inhibitors may be the reason for an improved outcome in the specific population of patients with mutated FLT3.10 Takahashi et al. recently reportedthat among 120 patients with AML and FLT3 internal tandem duplication (FLT3-ITD) mutation, there was a statistically significant improvement in OS for the patients with primary refractory disease or CR1 duration less than 12 months who were treated with FLT3 inhibitors Importantly, on multivariate analysis, the use of FLT3 inhibitors inpatients with FLT3-ITD mutation, was an independent predictor for improved OS.10 Found in approximately 25% of patients with AML, FLT3-ITD serves not only as a marker of poor prognosis, but also a potential drug target with several clinical trials of agents targeting this kinase, underway.11–13 Reports of single agent activity10 as well as combination of FLT3 inhibitors with other agents in the salvage setting10,14 have been promising. Third, the introduction of the hypomethylating agents decitabine15,16 and azacytidine17,18,19 have provided us with less toxic strategies for treating older, infirm patients both in the frontline and salvage settings.14,20
In summary, despite patients with AML presenting at an older age at diagnosis and relapse, we have shown that the outcomes of patients with AML in first relapse have improved over time, and that the era of therapy is an independent prognostic factors for overall survival after first relapse with an improved survival in the last several years.
Acknowledgments
The University of Texas MD Anderson Cancer Center is supported in part by the National Institutes of Health through a Cancer Center Support Grant (P30CA16672).
Footnotes
Authorship Contributions: NP, HK, GGM, JC, FR evaluated the patients, gathered the data, and analyzed the data. NP, SP, MCT and FR analyzed the data and wrote the manuscript. All authors reviewed the manuscript and gave final approval.
Conflicts/Disclosures: The authors have no disclosures or conflicts of interest with regards to this manuscript.
References
- 1.Burnett A, Wetzler M, Lowenberg B. Therapeutic advances in acute myeloid leukemia. J Clin Oncol. 2011;29:487–94. doi: 10.1200/JCO.2010.30.1820. [DOI] [PubMed] [Google Scholar]
- 2.Keating MJ, Kantarjian H, Smith TL, et al. Response to salvage therapy and survival after relapse in acute myelogenous leukemia. J Clin Oncol. 1989;7:1071–80. doi: 10.1200/JCO.1989.7.8.1071. [DOI] [PubMed] [Google Scholar]
- 3.Estey E, Kornblau S, Pierce S, Kantarjian H, Beran M, Keating M. A stratification system for evaluating and selecting therapies in patients with relapsed or primary refractory acute myelogenous leukemia. Blood. 1996;88:756. [PubMed] [Google Scholar]
- 4.Estey EH. Treatment of relapsed and refractory acute myelogenous leukemia. Leukemia. 2000;14:476–9. doi: 10.1038/sj.leu.2401568. [DOI] [PubMed] [Google Scholar]
- 5.Ravandi F, Cortes J, Faderl S, et al. Characteristics and outcome of patients with acute myeloid leukemia refractory to 1 cycle of high-dose cytarabine-based induction chemotherapy. Blood. 2010;116:5818–23. doi: 10.1182/blood-2010-07-296392. quiz 6153. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Breems DA, Van Putten WL, Huijgens PC, et al. Prognostic index for adult patients with acute myeloid leukemia in first relapse. J Clin Oncol. 2005;23:1969–78. doi: 10.1200/JCO.2005.06.027. [DOI] [PubMed] [Google Scholar]
- 7.Kurosawa S, Yamaguchi T, Miyawaki S, et al. Prognostic factors and outcomes of adult patients with acute myeloid leukemia after first relapse. Haematologica. 2010;95:1857–64. doi: 10.3324/haematol.2010.027516. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Kern W, Aul C, Maschmeyer G, et al. Superiority of high-dose over intermediate-dose cytosine arabinoside in the treatment of patients with high-risk acute myeloid leukemia: results of an age-adjusted prospective randomized comparison. Leukemia. 1998;12:1049–55. doi: 10.1038/sj.leu.2401066. [DOI] [PubMed] [Google Scholar]
- 9.Quintas-Cardama A, Ravandi F, Liu-Dumlao T, et al. Epigenetic therapy is associated with similar survival compared with intensive chemotherapy in older patients with newly diagnosed acute myeloid leukemia. Blood. 2012;120:4840–5. doi: 10.1182/blood-2012-06-436055. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Takahashi K, Kantarjian H, Pemmaraju N, et al. Salvage therapy using FLT3 inhibitors may improve long-term outcome of relapsed or refractory AML in patients with FLT3-ITD. Br J Haematol. 2013;161:659–66. doi: 10.1111/bjh.12299. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Patel JP, Gonen M, Figueroa ME, et al. Prognostic relevance of integrated genetic profiling in acute myeloid leukemia. N Engl J Med. 2012;366:1079–89. doi: 10.1056/NEJMoa1112304. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Pemmaraju N, Kantarjian H, Andreeff M, Cortes J, Ravandi F. Investigational FMS-like tyrosine kinase 3 inhibitors in treatment of acute myeloid leukemia. Expert Opin Investig Drugs. 2014 doi: 10.1517/13543784.2014.911839. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Kayser S, Levis MJ. FLT3 tyrosine kinase inhibitors in acute myeloid leukemia: clinical implications and limitations. Leuk Lymphoma. 2013 doi: 10.3109/10428194.2013.800198. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Ravandi F, Alattar ML, Grunwald MR, et al. Phase 2 study of azacytidine plus sorafenib in patients with acute myeloid leukemia and FLT-3 internal tandem duplication mutation. Blood. 2013;121:4655–62. doi: 10.1182/blood-2013-01-480228. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Kantarjian H, Issa JP, Rosenfeld CS, et al. Decitabine improves patient outcomes in myelodysplastic syndromes: results of a phase III randomized study. Cancer. 2006;106:1794–803. doi: 10.1002/cncr.21792. [DOI] [PubMed] [Google Scholar]
- 16.Blum W, Garzon R, Klisovic RB, et al. Clinical response and miR-29b predictive significance in older AML patients treated with a 10-day schedule of decitabine. Proc Natl Acad Sci U S A. 2010;107:7473–8. doi: 10.1073/pnas.1002650107. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Fenaux P, Mufti GJ, Hellstrom-Lindberg E, et al. Efficacy of azacitidine compared with that of conventional care regimens in the treatment of higher-risk myelodysplastic syndromes: a randomised, open-label, phase III study. Lancet Oncol. 2009;10:223–32. doi: 10.1016/S1470-2045(09)70003-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Fenaux P, Gattermann N, Seymour JF, et al. Prolonged survival with improved tolerability in higher-risk myelodysplastic syndromes: azacitidine compared with low dose ara-C. Br J Haematol. 2010;149:244–9. doi: 10.1111/j.1365-2141.2010.08082.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Fenaux P, Mufti GJ, Hellstrom-Lindberg E, et al. Azacitidine prolongs overall survival compared with conventional care regimens in elderly patients with low bone marrow blast count acute myeloid leukemia. J Clin Oncol. 2010;28:562–9. doi: 10.1200/JCO.2009.23.8329. [DOI] [PubMed] [Google Scholar]
- 20.Blum W, Klisovic RB, Hackanson B, et al. Phase I study of decitabine alone or in combination with valproic acid in acute myeloid leukemia. J Clin Oncol. 2007;25:3884–91. doi: 10.1200/JCO.2006.09.4169. [DOI] [PubMed] [Google Scholar]