- Email: [email protected]
A cytogenetic model predicts relapse risk and survival in patients with acute myeloid leukemia undergoing hematopoietic stem cell transplantation in morphologic complete remission
G Model
ARTICLE IN PRESS
LR-5279; No. of Pages 5
Leukemia Research xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Leukemia Research journal homepage: www.elsevier.com/locate/leukres
A cytogenetic model predicts relapse risk and survival in patients with acute myeloid leukemia undergoing hematopoietic stem cell transplantation in morphologic complete remission Armin Rashidi, Amanda F. Cashen ∗ Division of Oncology, Washington University School of Medicine, St. Louis, MO, United States
a r t i c l e
i n f o
Article history: Received 25 August 2014 Received in revised form 8 November 2014 Accepted 10 November 2014 Available online xxx Keywords: Cytogenetic Acute myeloid leukemia Relapse Remission Survival Stem cell transplantation
a b s t r a c t Up to 30% of patients with acute myeloid leukemia (AML) and abnormal cytogenetics have persistent cytogenetic abnormalities (pCytAbnl) at morphologic complete remission (mCR). We hypothesized that the prognostic significance of pCytAbnl in patients undergoing allogeneic hematopoietic stem cell transplantation (HSCT) in mCR varies with cytogenetic risk group. We analyzed the data on 118 patients with AML and abnormal cytogenetics who underwent HSCT in mCR, and developed a risk stratification model based on pCytAbnl and cytogenetic risk group. The model distinguished three groups of patients (P < 0.01) with distinct outcomes: the group with pCytAbnl and unfavorable risk cytogenetics (n = 25) had the shortest median time to relapse (TTR; 5 months), relapse-free survival (RFS; 3 months), and overall survival (OS; 7 months). The group with favorable/intermediate risk cytogenetics and without pCytAbnl (n = 43) had the longest median TTR (not reached), RFS (57 months), and OS (57 months). The group with pCytAbnl and favorable/intermediate risk cytogenetics, or, without pCytAbnl but with unfavorable risk cytogenetics (n = 50) experienced intermediate TTR (18 months), RFS (9 months), and OS (18 months). In conclusion, a cytogenetic risk model identifies patients with AML in mCR with distinct rates of relapse and survival following HSCT. © 2014 Elsevier Ltd. All rights reserved.
1. Introduction Acute myeloid leukemia (AML) relapses after allogeneic hematopoietic stem cell transplantation (HSCT) in 30–35% of patients. Relapse occurs typically within the first 6 months posttransplantation and portends a dismal prognosis, with a median post-relapse overall survival (OS) of 3–6 months [1–3]. Chemotherapy and donor lymphocyte infusion provide only limited post-relapse median OS of less than 3 months [1]. Although second HSCT may offer durable remissions with 5-year OS rates approaching 30% [4], only a minority of patients are eligible to receive a second transplant. Given the currently poor prognosis of patients with relapsed AML following HSCT, risk-stratification of patients based on pretransplantation characteristics could help identify those patients who will not benefit from HSCT or guide the implementation of strategies to reduce the risk of posttransplant relapse.
Approximately 10–30% of patients with an abnormal karyotype at the time of diagnosis of AML have persistent cytogenetic abnormalities (pCytAbnl) at morphologic complete remission (mCR) [5–8]. In a non-transplant setting these patients have higher overall mortality and relapse risk compared to those without pCytAbnl [7], but the significance of pCytAbnl at the time of mCR in patients who undergo HSCT has been inconsistent and not clearly defined. Considering that unfavorable cytogenetics at diagnosis is associated with higher relapse rates and shorter relapse-free survival (RFS) after HSCT [9], we hypothesized that the prognostic significance of pCytAbnl in patients undergoing HSCT in mCR varies with the cytogenetic risk group. Using a large cohort of patients with AML and abnormal cytogenetics at diagnosis who underwent HSCT in mCR, we develop a simple risk stratification model that enables us to delineate patients with distinct rates of relapse, RFS, and OS. 2. Materials and methods 2.1. Patients
∗ Corresponding author at: 660 S. Euclid Avenue, Campus Box 8007, St. Louis, MO 63110, United States. Tel.: +1 314 454 8306; fax: +1 314 454 7551. E-mail address: [email protected] (A.F. Cashen).
By reviewing our computerized database, we identified consecutive adult patients (age ≥18 years) with AML with cytogenetic abnormalities at initial diagnosis, who underwent HSCT while in morphological CR between January 2009 and July 2013. The institutional review board of the Washington University School of
http://dx.doi.org/10.1016/j.leukres.2014.11.007 0145-2126/© 2014 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Rashidi A, Cashen AF. A cytogenetic model predicts relapse risk and survival in patients with acute myeloid leukemia undergoing hematopoietic stem cell transplantation in morphologic complete remission. Leuk Res (2014), http://dx.doi.org/10.1016/j.leukres.2014.11.007
G Model LR-5279; No. of Pages 5
ARTICLE IN PRESS A. Rashidi, A.F. Cashen / Leukemia Research xxx (2014) xxx–xxx
2
Medicine approved this retrospective analysis. The majority of patients were treated on clinical trials. All patients gave written informed consent before undergoing HSCT.
2.2. Cytogenetic analysis All samples were analyzed by standard cytogenetic techniques in our institution. At least 20 metaphases from the bone marrow were required to define a karyotype as normal. For cytogenetically abnormal samples, analyses with fewer than 20 metaphases were also acceptable. Following the International System for Human Cytogenetic Nomenclature [10], a cytogenetic abnormality was considered clonal when at least two metaphases had the same aberration in case of structural abnormalities or trisomies. The requirement for definition of clonal monosomies was their presence in at least three metaphases. Cytogenetic risk group was defined according to the guidelines of the Southwest Oncology Group and Eastern Cooperative Oncology Group [11].
2.3. Stem cell transplantation Patients who received myeloablative or reduced-intensity conditioning regimens were included. Standard mobilization protocols and apheresis techniques were used for the procurement of donor bone marrow or G-CSF-primed peripheral blood stem cells. Transplants from a matched sibling were favored over those from unrelated donors or HLA-haploidentical siblings. The most frequently used GvHD prophylaxis regimens were tacrolimus plus methotrexate, with or without mycophenolate mofetil. Tacrolimus was dose-adjusted to maintain blood levels of 5–15 ng/dL during the first 100 days posttransplant and then tapered, as clinically indicated. Methotrexate was most commonly given at a dose of 10 mg/m2 on day +1 and 7.5 mg/m2 on days +3 and +6. Filgrastim was administered subcutaneously daily beginning on day +7 posttransplant or otherwise if dictated by the protocol. Filgrastim was discontinued once the absolute neutrophil count recovered to more than 1.5 × 109 /L for 2 consecutive days. Cytomegalovirus (CMV) surveillance consisted of weekly screening by polymerase chain reaction (PCR), with preemptive use of ganciclovir or foscarnet if new positivity or rising titers were discovered.
2.4. Definitions pCytAbnl was defined as cytogenetic abnormalities detected by fluorescence in situ hybridization or metaphase cytogenetics at mCR that were present in the clone at the time of diagnosis of AML. According to the criteria reported by the International Working Group [12], mCR was defined as <5% blast cells, no Auer rods, and no cluster of blast cells on the bone marrow analysis, as well as no evidence of extramedullary leukemia. A cytogenetic abnormality at the time of diagnosis of AML that did not disappear on a bone marrow biopsy done performed within 2 months preceding HSCT (and before conditioning) and in mCR was considered persistent. A monosomal karyotype (MK) was defined as at least two autosomal monosomies or a single autosomal monosomy in the presence of one or more structural cytogenetic abnormalities [13]. Relapse was defined as ≥5% blasts in the bone marrow or development of extramedullary leukemia. Bone marrow analysis was performed on days +30 and +100 posttransplant, and every 3–6 months thereafter. For RFS, patients were considered to experience failures at the time of relapse or death from any cause. For OS, the event was death from any cause. Time to relapse (TTR) was measured from the day of transplantation. 2.5. Statistical analyses Unadjusted probabilities of OS and RFS were estimated using the Kaplan–Meier method. All outcomes were treated as time-to-event points. OS was measured from the date of transplantation until the date of death and censored on the date of last follow up if alive. RFS was measured from the date of transplantation and censored on the date of last follow up if alive and in remission. Variables with a normal distribution are presented as mean ± standard deviation (SD), and those with a skewed distribution as median (standard error of the mean; SEM). Analyses were performed using SPSS version 17.0 (SPSS, Chicago, IL), and a P value of <0.05 was considered statistically significant.
3. Results A total of 118 patients (age 51 ± 14 years, 58% males) met the inclusion criteria. AML was therapy-related in 21 (31%) patients, 15 (71%) of whom had a known myelodysplastic syndrome that
Fig. 1. Overall survival (A), relapse-free survival (B), and cumulative incidence of relapse (C) in patients with and without persistence of cytogenetic abnormalities (pCytAbnl).
Please cite this article in press as: Rashidi A, Cashen AF. A cytogenetic model predicts relapse risk and survival in patients with acute myeloid leukemia undergoing hematopoietic stem cell transplantation in morphologic complete remission. Leuk Res (2014), http://dx.doi.org/10.1016/j.leukres.2014.11.007
G Model LR-5279; No. of Pages 5
ARTICLE IN PRESS A. Rashidi, A.F. Cashen / Leukemia Research xxx (2014) xxx–xxx
transformed to AML. The most frequent FAB subtypes were M0/M1/M2 (46%), followed by M4/M5 (26%). Favorable, intermediate and unfavorable cytogenetic risk disease was present in 18%, 29%, and 53% of patients, respectively. A monosomal karyotype was present in 31% of patients. The majority (73%) of patients were in CR1. Thirty seven patients (31%) had pCytAbnl prior to transplant. Conditioning was myeloablative in 67% of patients and reduced-intensity in the remainder. The stem cell source was peripheral blood in 98% of patients and was provided by a matched sibling, a matched unrelated donor, and a haploidentical relative in 35%, 63% and 2% of patients, respectively. The donor-recipient ABO blood group match status was matched, minor mismatched, and major mismatched in 45%, 28%, and 27% of patients, respectively. The transplant was sex-mismatched in 39% and the donor was a female in 36% of transplants. The median RFS and OS rates for the entire group were 10 and 24 months, respectively. Both median RFS and median OS were significantly longer among patients without pCytAbnl (18 and 37 vs. 3 and 8 months, respectively; P = 0.002 for both; Fig. 1). The median TTR was 5 months for patients with pCytAbnl. The median TTR was not reached for patients without pCytAbnl (P < 0.001). Next we sought to evaluate whether the prognostic effect of pCytAbnl is modified by the cytogenetic risk group. We built a risk scoring system using pCytAbnl and cytogenetic risk group as adverse risk features (Table 1). These two features separated patients into three distinct groups (Fig. 2). The group with pCytAbnl and unfavorable risk cytogenetics (R2: two adverse risk features) had the highest relapse rate and shortest RFS and OS. The group without pCytAbnl but with favorable/intermediate risk
3
Table 1 Median time to relapse, relapse-free survival, and overall survival. Time to relapse (months)
Overall survival (months)
Relapse-free survival (months)
Model 1: pCytAbnl and cytogenetic risk group R0 (n = 43) Not reached 57 R1 (n = 50) 18 18 R2 (n = 25) 5 7
57 9 3
Model 2: pCytAbnl and monosomal karyotype Not reached 57 R0 (n = 61) R1 (n = 40) 22 11 3 7 R2 (n = 17)
22 5 3
OS, overall survival; pCytAbnl, persistent cytogenetic abnormalities; RFS, relapsefree survival; TTR, time to relapse; R0, no adverse risk feature; R1, one adverse risk feature; R2, two adverse risk features.
cytogenetics (R0: no adverse risk features) had the lowest relapse rate and longest RFS and OS. The other group (R1: with pCytAbnl and favorable/intermediate risk cytogenetics, or, without pCytAbnl but with unfavorable risk cytogenetics) experienced intermediate relapse and survival rates. Clustering the two subgroups of R1 together was further justified by the observation that they were not different in TTR (P = 0.94), RFS (P = 0.88), or OS (P = 0.65) (Fig. S1). Notably, R0, R1, and R2 groups were not different in the frequency of myeloablative vs. reduced-intensity conditioning regimen (P = 0.40). Supplementary material related to this article can be found, in the online version, at http://dx.doi.org/10.1016/j.leukres. 2014.11.007.
Fig. 2. Cumulative incidence of relapse (A), relapse free survival (B) and overall survival (C) based on persistent cytogenetic abnormalities and cytogenetic risk groups. R0, R1, and R2 are defined in the text and in Table 1.
Please cite this article in press as: Rashidi A, Cashen AF. A cytogenetic model predicts relapse risk and survival in patients with acute myeloid leukemia undergoing hematopoietic stem cell transplantation in morphologic complete remission. Leuk Res (2014), http://dx.doi.org/10.1016/j.leukres.2014.11.007
G Model LR-5279; No. of Pages 5
ARTICLE IN PRESS A. Rashidi, A.F. Cashen / Leukemia Research xxx (2014) xxx–xxx
4
Fig. 3. Cumulative incidence of relapse (A), relapse free survival (B) and overall survival (C) based on persistent cytogenetic abnormalities and monosomal karyotype. R0, R1, and R2 are defined in the text and in Table 1.
Finally, we built a risk scoring system using pCytAbnl and presence or absence of MK (Table 1). For OS, the curves were slightly more separated and the corresponding hazard ratios were marginally larger using this model (Fig. 3C) compared to the model based on pCytoAbnl and traditional cytogenetic risk groups (Fig. 2C). However, the model with MK had an inferior performance compared to the one with cytogenetic risk groups in terms of TTR (Figs. 2A and 3A) and RFS (Figs. 2B and 3B). The separation between the curves for both TTR and RFS was more pronounced and the hazard ratios were larger in the model with cytogenetic risk groups compared the one with MK. 4. Discussion It is well established that active or high-burden disease at the time of HSCT negatively impacts outcomes [14–16]. Even minimal residual disease (MRD), as detected by multiparametric flow cytometry, is associated with increased risk of relapse and death [17]. Considering that MRD testing before HSCT is not readily available or commonly practiced, there is an urgent need for models using conventional methods to risk stratify patients before HSCT. Patients at higher risk of relapse could be considered for more frequent monitoring and bone marrow examinations, maintenance chemo- or immunotherapy, more intense conditioning regimens, or non-transplant treatment approaches. Pretransplantation pCytAbnl was associated with significantly shorter RFS and OS in two previous studies [5,17], a statistically non-significant trend for shorter RFS and OS in one study [8], a statistically non-significant trend for increased risk of relapse in one study [8], and no change in the risk of relapse in one study [17].
To address these inconsistencies, we hypothesized that cytogenetic risk group may act as a modifier for the outcome effects of pCytAbnl, given that poor risk cytogenetics at diagnosis is associated with shortened RFS in AML patients who undergo HSCT in mCR [9]. Using only two variables (pCytAbnl and conventional cytogenetic risk groups), our composite risk stratification model classified patients into three groups with distinct rates of relapse and survival (Fig. 2). Of note, our model applies only to patients with abnormal cytogenetics at the time of diagnosis of AML and mCR at the time of HSCT. Cytogenetic risk grouping by the monosomal karyotype classification was superior in a recent study in predicting the outcome of AML undergoing HSCT in mCR [18]. In the present study, although the MK composite model had a slightly better prognostic performance in terms of OS compared to the model with traditional cytogenetic risk group, it did not perform as well for RFS and TTR. The largest previous study of transplant outcomes in patients with pCytAbnl included 18 such patients, only three of which had favorable/intermediate risk cytogenetics [8]. This limited the authors’ analysis of the potential modifying effects of cytogenetic risk group on pCytAbnl. Patients with unfavorable cytogenetics and pCytAbnl had a significantly shorter RFS and higher relapse rates compared to those with favorable/intermediate cytogenetics and without pCytAbnl. Also, patients with unfavorable cytogenetics but without pCytAbnl had a significantly higher relapse rates compared to those with favorable/intermediate cytogenetics and without pCytAbnl. All other intergroup comparisons for relapse and survival were non-significant. Importantly, cytogenetic risk group and pCytAbnl had no significant effect on OS. Our larger sample of patients with pCytAbnl enabled us to build a composite risk model, and as a result, our outcomes for the three risk groups were
Please cite this article in press as: Rashidi A, Cashen AF. A cytogenetic model predicts relapse risk and survival in patients with acute myeloid leukemia undergoing hematopoietic stem cell transplantation in morphologic complete remission. Leuk Res (2014), http://dx.doi.org/10.1016/j.leukres.2014.11.007
G Model LR-5279; No. of Pages 5
ARTICLE IN PRESS A. Rashidi, A.F. Cashen / Leukemia Research xxx (2014) xxx–xxx
distinctly different. Larger studies in the future may focus on patients with more homogenous clinical and transplant characteristics to avoid potential limitations related to sample heterogeneity. In conclusion, we have developed a composite model based on pretransplantation pCytAbnl and cytogenetic risk group for patients with AML who undergo HSCT in mCR. The model identifies three groups of patients with distinct rates of relapse and survival. With the increasingly popular approach of risk-based individualized therapy in AML, the results of this study can be incorporated into larger scale models to guide treatment. Escalation of therapy, implementation of maintenance strategies, and closer follow up for higher risk patients may lead to improved outcomes, while de-escalation of therapy for lower risk patients can help avoid unnecessarily aggressive and potentially toxic measures in those who likely enjoy favorable outcomes. Conflicts of interest
[6] [7]
[8]
[9]
[10] [11]
[12]
The authors have no conflicts of interest. [13]
Sources of funding [14]
This study had no source of funding. References [1] Arellano ML, Langston A, Winton E, et al. Treatment of relapsed acute leukemia after allogeneic transplantation: a single center experience. Biol Blood Marrow Transplant 2007;13:116–23. [2] Pollyea DA, Artz AS, Stock W, et al. Outcomes of patients with AML and MDS who relapse or progress after reduced intensity allogeneic hematopoietic cell transplantation. Bone Marrow Transplant 2007;40:1027–32. [3] Thanarajasingam G, Kim HT, Cutler C, et al. Outcome and prognostic factors for patients who relapse after allogeneic hematopoietic stem cell transplantation. Biol Blood Marrow Transplant 2013;19:1713–8. [4] Eapen M, Giralt SA, Horowitz MM, et al. Second transplant for acute and chronic leukemia relapsing after first HLA-identical sibling transplant. Bone Marrow Transplant 2004;34:721–7. [5] Chen Y, Cortes J, Estrov Z, et al. Persistence of cytogenetic abnormalities at complete remission after induction in patients with acute myeloid leukemia:
[15]
[16]
[17]
[18]
5
prognostic significance and the potential role of allogeneic stem-cell transplantation. J Clin Oncol 2011;29:2507–13. Freireich EJ, Cork A, Stass SA, et al. Cytogenetics for detection of minimal residual disease in acute myeloblastic leukemia. Leukemia 1992;6:500–6. Marcucci G, Mrozek K, Ruppert AS, et al. Abnormal cytogenetics at date of morphologic complete remission predicts short overall and disease-free survival, and higher relapse rate in adult acute myeloid leukemia: results from cancer and leukemia group B study 8461. J Clin Oncol 2004;22:2410–8. Oran B, Popat U, Rondon G, et al. Significance of persistent cytogenetic abnormalities on myeloablative allogeneic stem cell transplantation in first complete remission. Biol Blood Marrow Transplant 2013;19:214–20. Chevallier P, Labopin M, Milpied N, et al. Impact of cytogenetics risk on outcome after reduced intensity conditioning allo-SCT from an HLA-identical sibling for patients with AML in first CR: a report from the acute leukemia working party of EBMT. Bone Marrow Transplant 2012;47:1442–7. Oran B, de Lima M. Prevention and treatment of acute myeloid leukemia relapse after allogeneic stem cell transplantation. Curr Opin Hematol 2011;18:388–94. Slovak ML, Kopecky KJ, Cassileth PA, et al. Karyotypic analysis predicts outcome of preremission and postremission therapy in adult acute myeloid leukemia: a Southwest Oncology Group/Eastern Cooperative Oncology Group Study. Blood 2000;96:4075–83. Cheson BD, Bennett JM, Kopecky KJ, et al. Revised recommendations of the International Working Group for Diagnosis, Standardization of Response Criteria, Treatment Outcomes, and Reporting Standards for Therapeutic Trials in Acute Myeloid Leukemia. J Clin Oncol 2003;21:4642–9. Breems DA, Van Putten WL, De Greef GE, et al. Monosomal karyotype in acute myeloid leukemia: a better indicator of poor prognosis than a complex karyotype. J Clin Oncol 2008;26:4791–7. Saliba RM, Komanduri KV, Giralt S, et al. Leukemia burden delays lymphocyte and platelet recovery after allo-SCT for AML. Bone Marrow Transplant 2009;43:685–92. Wong R, Shahjahan M, Wang X, et al. Prognostic factors for outcomes of patients with refractory or relapsed acute myelogenous leukemia or myelodysplastic syndromes undergoing allogeneic progenitor cell transplantation. Biol Blood Marrow Transplant 2005;11:108–14. Michallet M, Thomas X, Vernant JP, et al. Long-term outcome after allogeneic hematopoietic stem cell transplantation for advanced stage acute myeloblastic leukemia: a retrospective study of 379 patients reported to the Societe Francaise de Greffe de Moelle (SFGM). Bone Marrow Transplant 2000;26: 1157–63. Walter RB, Gooley TA, Wood BL, et al. Impact of pretransplantation minimal residual disease, as detected by multiparametric flow cytometry, on outcome of myeloablative hematopoietic cell transplantation for acute myeloid leukemia. J Clin Oncol 2011;29:1190–7. Hemmati PG, Schulze-Luckow A, Terwey TH, et al. Cytogenetic risk grouping by the monosomal karyotype classification is superior in predicting the outcome of acute myeloid leukemia undergoing allogeneic stem cell transplantation in complete remission. Eur J Haematol 2014;92:102–10.
Please cite this article in press as: Rashidi A, Cashen AF. A cytogenetic model predicts relapse risk and survival in patients with acute myeloid leukemia undergoing hematopoietic stem cell transplantation in morphologic complete remission. Leuk Res (2014), http://dx.doi.org/10.1016/j.leukres.2014.11.007
ARTICLE IN PRESS
LR-5279; No. of Pages 5
Leukemia Research xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Leukemia Research journal homepage: www.elsevier.com/locate/leukres
A cytogenetic model predicts relapse risk and survival in patients with acute myeloid leukemia undergoing hematopoietic stem cell transplantation in morphologic complete remission Armin Rashidi, Amanda F. Cashen ∗ Division of Oncology, Washington University School of Medicine, St. Louis, MO, United States
a r t i c l e
i n f o
Article history: Received 25 August 2014 Received in revised form 8 November 2014 Accepted 10 November 2014 Available online xxx Keywords: Cytogenetic Acute myeloid leukemia Relapse Remission Survival Stem cell transplantation
a b s t r a c t Up to 30% of patients with acute myeloid leukemia (AML) and abnormal cytogenetics have persistent cytogenetic abnormalities (pCytAbnl) at morphologic complete remission (mCR). We hypothesized that the prognostic significance of pCytAbnl in patients undergoing allogeneic hematopoietic stem cell transplantation (HSCT) in mCR varies with cytogenetic risk group. We analyzed the data on 118 patients with AML and abnormal cytogenetics who underwent HSCT in mCR, and developed a risk stratification model based on pCytAbnl and cytogenetic risk group. The model distinguished three groups of patients (P < 0.01) with distinct outcomes: the group with pCytAbnl and unfavorable risk cytogenetics (n = 25) had the shortest median time to relapse (TTR; 5 months), relapse-free survival (RFS; 3 months), and overall survival (OS; 7 months). The group with favorable/intermediate risk cytogenetics and without pCytAbnl (n = 43) had the longest median TTR (not reached), RFS (57 months), and OS (57 months). The group with pCytAbnl and favorable/intermediate risk cytogenetics, or, without pCytAbnl but with unfavorable risk cytogenetics (n = 50) experienced intermediate TTR (18 months), RFS (9 months), and OS (18 months). In conclusion, a cytogenetic risk model identifies patients with AML in mCR with distinct rates of relapse and survival following HSCT. © 2014 Elsevier Ltd. All rights reserved.
1. Introduction Acute myeloid leukemia (AML) relapses after allogeneic hematopoietic stem cell transplantation (HSCT) in 30–35% of patients. Relapse occurs typically within the first 6 months posttransplantation and portends a dismal prognosis, with a median post-relapse overall survival (OS) of 3–6 months [1–3]. Chemotherapy and donor lymphocyte infusion provide only limited post-relapse median OS of less than 3 months [1]. Although second HSCT may offer durable remissions with 5-year OS rates approaching 30% [4], only a minority of patients are eligible to receive a second transplant. Given the currently poor prognosis of patients with relapsed AML following HSCT, risk-stratification of patients based on pretransplantation characteristics could help identify those patients who will not benefit from HSCT or guide the implementation of strategies to reduce the risk of posttransplant relapse.
Approximately 10–30% of patients with an abnormal karyotype at the time of diagnosis of AML have persistent cytogenetic abnormalities (pCytAbnl) at morphologic complete remission (mCR) [5–8]. In a non-transplant setting these patients have higher overall mortality and relapse risk compared to those without pCytAbnl [7], but the significance of pCytAbnl at the time of mCR in patients who undergo HSCT has been inconsistent and not clearly defined. Considering that unfavorable cytogenetics at diagnosis is associated with higher relapse rates and shorter relapse-free survival (RFS) after HSCT [9], we hypothesized that the prognostic significance of pCytAbnl in patients undergoing HSCT in mCR varies with the cytogenetic risk group. Using a large cohort of patients with AML and abnormal cytogenetics at diagnosis who underwent HSCT in mCR, we develop a simple risk stratification model that enables us to delineate patients with distinct rates of relapse, RFS, and OS. 2. Materials and methods 2.1. Patients
∗ Corresponding author at: 660 S. Euclid Avenue, Campus Box 8007, St. Louis, MO 63110, United States. Tel.: +1 314 454 8306; fax: +1 314 454 7551. E-mail address: [email protected] (A.F. Cashen).
By reviewing our computerized database, we identified consecutive adult patients (age ≥18 years) with AML with cytogenetic abnormalities at initial diagnosis, who underwent HSCT while in morphological CR between January 2009 and July 2013. The institutional review board of the Washington University School of
http://dx.doi.org/10.1016/j.leukres.2014.11.007 0145-2126/© 2014 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Rashidi A, Cashen AF. A cytogenetic model predicts relapse risk and survival in patients with acute myeloid leukemia undergoing hematopoietic stem cell transplantation in morphologic complete remission. Leuk Res (2014), http://dx.doi.org/10.1016/j.leukres.2014.11.007
G Model LR-5279; No. of Pages 5
ARTICLE IN PRESS A. Rashidi, A.F. Cashen / Leukemia Research xxx (2014) xxx–xxx
2
Medicine approved this retrospective analysis. The majority of patients were treated on clinical trials. All patients gave written informed consent before undergoing HSCT.
2.2. Cytogenetic analysis All samples were analyzed by standard cytogenetic techniques in our institution. At least 20 metaphases from the bone marrow were required to define a karyotype as normal. For cytogenetically abnormal samples, analyses with fewer than 20 metaphases were also acceptable. Following the International System for Human Cytogenetic Nomenclature [10], a cytogenetic abnormality was considered clonal when at least two metaphases had the same aberration in case of structural abnormalities or trisomies. The requirement for definition of clonal monosomies was their presence in at least three metaphases. Cytogenetic risk group was defined according to the guidelines of the Southwest Oncology Group and Eastern Cooperative Oncology Group [11].
2.3. Stem cell transplantation Patients who received myeloablative or reduced-intensity conditioning regimens were included. Standard mobilization protocols and apheresis techniques were used for the procurement of donor bone marrow or G-CSF-primed peripheral blood stem cells. Transplants from a matched sibling were favored over those from unrelated donors or HLA-haploidentical siblings. The most frequently used GvHD prophylaxis regimens were tacrolimus plus methotrexate, with or without mycophenolate mofetil. Tacrolimus was dose-adjusted to maintain blood levels of 5–15 ng/dL during the first 100 days posttransplant and then tapered, as clinically indicated. Methotrexate was most commonly given at a dose of 10 mg/m2 on day +1 and 7.5 mg/m2 on days +3 and +6. Filgrastim was administered subcutaneously daily beginning on day +7 posttransplant or otherwise if dictated by the protocol. Filgrastim was discontinued once the absolute neutrophil count recovered to more than 1.5 × 109 /L for 2 consecutive days. Cytomegalovirus (CMV) surveillance consisted of weekly screening by polymerase chain reaction (PCR), with preemptive use of ganciclovir or foscarnet if new positivity or rising titers were discovered.
2.4. Definitions pCytAbnl was defined as cytogenetic abnormalities detected by fluorescence in situ hybridization or metaphase cytogenetics at mCR that were present in the clone at the time of diagnosis of AML. According to the criteria reported by the International Working Group [12], mCR was defined as <5% blast cells, no Auer rods, and no cluster of blast cells on the bone marrow analysis, as well as no evidence of extramedullary leukemia. A cytogenetic abnormality at the time of diagnosis of AML that did not disappear on a bone marrow biopsy done performed within 2 months preceding HSCT (and before conditioning) and in mCR was considered persistent. A monosomal karyotype (MK) was defined as at least two autosomal monosomies or a single autosomal monosomy in the presence of one or more structural cytogenetic abnormalities [13]. Relapse was defined as ≥5% blasts in the bone marrow or development of extramedullary leukemia. Bone marrow analysis was performed on days +30 and +100 posttransplant, and every 3–6 months thereafter. For RFS, patients were considered to experience failures at the time of relapse or death from any cause. For OS, the event was death from any cause. Time to relapse (TTR) was measured from the day of transplantation. 2.5. Statistical analyses Unadjusted probabilities of OS and RFS were estimated using the Kaplan–Meier method. All outcomes were treated as time-to-event points. OS was measured from the date of transplantation until the date of death and censored on the date of last follow up if alive. RFS was measured from the date of transplantation and censored on the date of last follow up if alive and in remission. Variables with a normal distribution are presented as mean ± standard deviation (SD), and those with a skewed distribution as median (standard error of the mean; SEM). Analyses were performed using SPSS version 17.0 (SPSS, Chicago, IL), and a P value of <0.05 was considered statistically significant.
3. Results A total of 118 patients (age 51 ± 14 years, 58% males) met the inclusion criteria. AML was therapy-related in 21 (31%) patients, 15 (71%) of whom had a known myelodysplastic syndrome that
Fig. 1. Overall survival (A), relapse-free survival (B), and cumulative incidence of relapse (C) in patients with and without persistence of cytogenetic abnormalities (pCytAbnl).
Please cite this article in press as: Rashidi A, Cashen AF. A cytogenetic model predicts relapse risk and survival in patients with acute myeloid leukemia undergoing hematopoietic stem cell transplantation in morphologic complete remission. Leuk Res (2014), http://dx.doi.org/10.1016/j.leukres.2014.11.007
G Model LR-5279; No. of Pages 5
ARTICLE IN PRESS A. Rashidi, A.F. Cashen / Leukemia Research xxx (2014) xxx–xxx
transformed to AML. The most frequent FAB subtypes were M0/M1/M2 (46%), followed by M4/M5 (26%). Favorable, intermediate and unfavorable cytogenetic risk disease was present in 18%, 29%, and 53% of patients, respectively. A monosomal karyotype was present in 31% of patients. The majority (73%) of patients were in CR1. Thirty seven patients (31%) had pCytAbnl prior to transplant. Conditioning was myeloablative in 67% of patients and reduced-intensity in the remainder. The stem cell source was peripheral blood in 98% of patients and was provided by a matched sibling, a matched unrelated donor, and a haploidentical relative in 35%, 63% and 2% of patients, respectively. The donor-recipient ABO blood group match status was matched, minor mismatched, and major mismatched in 45%, 28%, and 27% of patients, respectively. The transplant was sex-mismatched in 39% and the donor was a female in 36% of transplants. The median RFS and OS rates for the entire group were 10 and 24 months, respectively. Both median RFS and median OS were significantly longer among patients without pCytAbnl (18 and 37 vs. 3 and 8 months, respectively; P = 0.002 for both; Fig. 1). The median TTR was 5 months for patients with pCytAbnl. The median TTR was not reached for patients without pCytAbnl (P < 0.001). Next we sought to evaluate whether the prognostic effect of pCytAbnl is modified by the cytogenetic risk group. We built a risk scoring system using pCytAbnl and cytogenetic risk group as adverse risk features (Table 1). These two features separated patients into three distinct groups (Fig. 2). The group with pCytAbnl and unfavorable risk cytogenetics (R2: two adverse risk features) had the highest relapse rate and shortest RFS and OS. The group without pCytAbnl but with favorable/intermediate risk
3
Table 1 Median time to relapse, relapse-free survival, and overall survival. Time to relapse (months)
Overall survival (months)
Relapse-free survival (months)
Model 1: pCytAbnl and cytogenetic risk group R0 (n = 43) Not reached 57 R1 (n = 50) 18 18 R2 (n = 25) 5 7
57 9 3
Model 2: pCytAbnl and monosomal karyotype Not reached 57 R0 (n = 61) R1 (n = 40) 22 11 3 7 R2 (n = 17)
22 5 3
OS, overall survival; pCytAbnl, persistent cytogenetic abnormalities; RFS, relapsefree survival; TTR, time to relapse; R0, no adverse risk feature; R1, one adverse risk feature; R2, two adverse risk features.
cytogenetics (R0: no adverse risk features) had the lowest relapse rate and longest RFS and OS. The other group (R1: with pCytAbnl and favorable/intermediate risk cytogenetics, or, without pCytAbnl but with unfavorable risk cytogenetics) experienced intermediate relapse and survival rates. Clustering the two subgroups of R1 together was further justified by the observation that they were not different in TTR (P = 0.94), RFS (P = 0.88), or OS (P = 0.65) (Fig. S1). Notably, R0, R1, and R2 groups were not different in the frequency of myeloablative vs. reduced-intensity conditioning regimen (P = 0.40). Supplementary material related to this article can be found, in the online version, at http://dx.doi.org/10.1016/j.leukres. 2014.11.007.
Fig. 2. Cumulative incidence of relapse (A), relapse free survival (B) and overall survival (C) based on persistent cytogenetic abnormalities and cytogenetic risk groups. R0, R1, and R2 are defined in the text and in Table 1.
Please cite this article in press as: Rashidi A, Cashen AF. A cytogenetic model predicts relapse risk and survival in patients with acute myeloid leukemia undergoing hematopoietic stem cell transplantation in morphologic complete remission. Leuk Res (2014), http://dx.doi.org/10.1016/j.leukres.2014.11.007
G Model LR-5279; No. of Pages 5
ARTICLE IN PRESS A. Rashidi, A.F. Cashen / Leukemia Research xxx (2014) xxx–xxx
4
Fig. 3. Cumulative incidence of relapse (A), relapse free survival (B) and overall survival (C) based on persistent cytogenetic abnormalities and monosomal karyotype. R0, R1, and R2 are defined in the text and in Table 1.
Finally, we built a risk scoring system using pCytAbnl and presence or absence of MK (Table 1). For OS, the curves were slightly more separated and the corresponding hazard ratios were marginally larger using this model (Fig. 3C) compared to the model based on pCytoAbnl and traditional cytogenetic risk groups (Fig. 2C). However, the model with MK had an inferior performance compared to the one with cytogenetic risk groups in terms of TTR (Figs. 2A and 3A) and RFS (Figs. 2B and 3B). The separation between the curves for both TTR and RFS was more pronounced and the hazard ratios were larger in the model with cytogenetic risk groups compared the one with MK. 4. Discussion It is well established that active or high-burden disease at the time of HSCT negatively impacts outcomes [14–16]. Even minimal residual disease (MRD), as detected by multiparametric flow cytometry, is associated with increased risk of relapse and death [17]. Considering that MRD testing before HSCT is not readily available or commonly practiced, there is an urgent need for models using conventional methods to risk stratify patients before HSCT. Patients at higher risk of relapse could be considered for more frequent monitoring and bone marrow examinations, maintenance chemo- or immunotherapy, more intense conditioning regimens, or non-transplant treatment approaches. Pretransplantation pCytAbnl was associated with significantly shorter RFS and OS in two previous studies [5,17], a statistically non-significant trend for shorter RFS and OS in one study [8], a statistically non-significant trend for increased risk of relapse in one study [8], and no change in the risk of relapse in one study [17].
To address these inconsistencies, we hypothesized that cytogenetic risk group may act as a modifier for the outcome effects of pCytAbnl, given that poor risk cytogenetics at diagnosis is associated with shortened RFS in AML patients who undergo HSCT in mCR [9]. Using only two variables (pCytAbnl and conventional cytogenetic risk groups), our composite risk stratification model classified patients into three groups with distinct rates of relapse and survival (Fig. 2). Of note, our model applies only to patients with abnormal cytogenetics at the time of diagnosis of AML and mCR at the time of HSCT. Cytogenetic risk grouping by the monosomal karyotype classification was superior in a recent study in predicting the outcome of AML undergoing HSCT in mCR [18]. In the present study, although the MK composite model had a slightly better prognostic performance in terms of OS compared to the model with traditional cytogenetic risk group, it did not perform as well for RFS and TTR. The largest previous study of transplant outcomes in patients with pCytAbnl included 18 such patients, only three of which had favorable/intermediate risk cytogenetics [8]. This limited the authors’ analysis of the potential modifying effects of cytogenetic risk group on pCytAbnl. Patients with unfavorable cytogenetics and pCytAbnl had a significantly shorter RFS and higher relapse rates compared to those with favorable/intermediate cytogenetics and without pCytAbnl. Also, patients with unfavorable cytogenetics but without pCytAbnl had a significantly higher relapse rates compared to those with favorable/intermediate cytogenetics and without pCytAbnl. All other intergroup comparisons for relapse and survival were non-significant. Importantly, cytogenetic risk group and pCytAbnl had no significant effect on OS. Our larger sample of patients with pCytAbnl enabled us to build a composite risk model, and as a result, our outcomes for the three risk groups were
Please cite this article in press as: Rashidi A, Cashen AF. A cytogenetic model predicts relapse risk and survival in patients with acute myeloid leukemia undergoing hematopoietic stem cell transplantation in morphologic complete remission. Leuk Res (2014), http://dx.doi.org/10.1016/j.leukres.2014.11.007
G Model LR-5279; No. of Pages 5
ARTICLE IN PRESS A. Rashidi, A.F. Cashen / Leukemia Research xxx (2014) xxx–xxx
distinctly different. Larger studies in the future may focus on patients with more homogenous clinical and transplant characteristics to avoid potential limitations related to sample heterogeneity. In conclusion, we have developed a composite model based on pretransplantation pCytAbnl and cytogenetic risk group for patients with AML who undergo HSCT in mCR. The model identifies three groups of patients with distinct rates of relapse and survival. With the increasingly popular approach of risk-based individualized therapy in AML, the results of this study can be incorporated into larger scale models to guide treatment. Escalation of therapy, implementation of maintenance strategies, and closer follow up for higher risk patients may lead to improved outcomes, while de-escalation of therapy for lower risk patients can help avoid unnecessarily aggressive and potentially toxic measures in those who likely enjoy favorable outcomes. Conflicts of interest
[6] [7]
[8]
[9]
[10] [11]
[12]
The authors have no conflicts of interest. [13]
Sources of funding [14]
This study had no source of funding. References [1] Arellano ML, Langston A, Winton E, et al. Treatment of relapsed acute leukemia after allogeneic transplantation: a single center experience. Biol Blood Marrow Transplant 2007;13:116–23. [2] Pollyea DA, Artz AS, Stock W, et al. Outcomes of patients with AML and MDS who relapse or progress after reduced intensity allogeneic hematopoietic cell transplantation. Bone Marrow Transplant 2007;40:1027–32. [3] Thanarajasingam G, Kim HT, Cutler C, et al. Outcome and prognostic factors for patients who relapse after allogeneic hematopoietic stem cell transplantation. Biol Blood Marrow Transplant 2013;19:1713–8. [4] Eapen M, Giralt SA, Horowitz MM, et al. Second transplant for acute and chronic leukemia relapsing after first HLA-identical sibling transplant. Bone Marrow Transplant 2004;34:721–7. [5] Chen Y, Cortes J, Estrov Z, et al. Persistence of cytogenetic abnormalities at complete remission after induction in patients with acute myeloid leukemia:
[15]
[16]
[17]
[18]
5
prognostic significance and the potential role of allogeneic stem-cell transplantation. J Clin Oncol 2011;29:2507–13. Freireich EJ, Cork A, Stass SA, et al. Cytogenetics for detection of minimal residual disease in acute myeloblastic leukemia. Leukemia 1992;6:500–6. Marcucci G, Mrozek K, Ruppert AS, et al. Abnormal cytogenetics at date of morphologic complete remission predicts short overall and disease-free survival, and higher relapse rate in adult acute myeloid leukemia: results from cancer and leukemia group B study 8461. J Clin Oncol 2004;22:2410–8. Oran B, Popat U, Rondon G, et al. Significance of persistent cytogenetic abnormalities on myeloablative allogeneic stem cell transplantation in first complete remission. Biol Blood Marrow Transplant 2013;19:214–20. Chevallier P, Labopin M, Milpied N, et al. Impact of cytogenetics risk on outcome after reduced intensity conditioning allo-SCT from an HLA-identical sibling for patients with AML in first CR: a report from the acute leukemia working party of EBMT. Bone Marrow Transplant 2012;47:1442–7. Oran B, de Lima M. Prevention and treatment of acute myeloid leukemia relapse after allogeneic stem cell transplantation. Curr Opin Hematol 2011;18:388–94. Slovak ML, Kopecky KJ, Cassileth PA, et al. Karyotypic analysis predicts outcome of preremission and postremission therapy in adult acute myeloid leukemia: a Southwest Oncology Group/Eastern Cooperative Oncology Group Study. Blood 2000;96:4075–83. Cheson BD, Bennett JM, Kopecky KJ, et al. Revised recommendations of the International Working Group for Diagnosis, Standardization of Response Criteria, Treatment Outcomes, and Reporting Standards for Therapeutic Trials in Acute Myeloid Leukemia. J Clin Oncol 2003;21:4642–9. Breems DA, Van Putten WL, De Greef GE, et al. Monosomal karyotype in acute myeloid leukemia: a better indicator of poor prognosis than a complex karyotype. J Clin Oncol 2008;26:4791–7. Saliba RM, Komanduri KV, Giralt S, et al. Leukemia burden delays lymphocyte and platelet recovery after allo-SCT for AML. Bone Marrow Transplant 2009;43:685–92. Wong R, Shahjahan M, Wang X, et al. Prognostic factors for outcomes of patients with refractory or relapsed acute myelogenous leukemia or myelodysplastic syndromes undergoing allogeneic progenitor cell transplantation. Biol Blood Marrow Transplant 2005;11:108–14. Michallet M, Thomas X, Vernant JP, et al. Long-term outcome after allogeneic hematopoietic stem cell transplantation for advanced stage acute myeloblastic leukemia: a retrospective study of 379 patients reported to the Societe Francaise de Greffe de Moelle (SFGM). Bone Marrow Transplant 2000;26: 1157–63. Walter RB, Gooley TA, Wood BL, et al. Impact of pretransplantation minimal residual disease, as detected by multiparametric flow cytometry, on outcome of myeloablative hematopoietic cell transplantation for acute myeloid leukemia. J Clin Oncol 2011;29:1190–7. Hemmati PG, Schulze-Luckow A, Terwey TH, et al. Cytogenetic risk grouping by the monosomal karyotype classification is superior in predicting the outcome of acute myeloid leukemia undergoing allogeneic stem cell transplantation in complete remission. Eur J Haematol 2014;92:102–10.
Please cite this article in press as: Rashidi A, Cashen AF. A cytogenetic model predicts relapse risk and survival in patients with acute myeloid leukemia undergoing hematopoietic stem cell transplantation in morphologic complete remission. Leuk Res (2014), http://dx.doi.org/10.1016/j.leukres.2014.11.007