myelodysplastic syndrome - Nature

2 downloads 135 Views 507KB Size Report
Jun 29, 2015 - 1Allogeneic Blood and Marrow Transplant Program, Princess Margaret Cancer Center, University of Toronto, Toronto, Ontario, Canada and ...
Bone Marrow Transplantation (2015) 50, 1180–1186 © 2015 Macmillan Publishers Limited All rights reserved 0268-3369/15 www.nature.com/bmt

ORIGINAL ARTICLE

Outcomes of patients with therapy-related AML/myelodysplastic syndrome (t-AML/MDS) following hematopoietic cell transplantation N Alam1, EG Atenafu2, J Kuruvilla1, J Uhm1, JH Lipton1, HA Messner1, DH Kim1, M Seftel1 and V Gupta1 We studied outcomes of 65 consecutive patients with therapy-related AML/myelodyplastic syndrome (t-AML/MDS) who underwent allogeneic hematopoietic cell transplantation (HCT). Previously published scores of HCT-CI, CIBMTR, EBMT and Comorbidity-age index were also evaluated. Median follow-up of survivors was 72 months (range 16–204). At 2 years, overall survival (OS) was 34% (95% confidence interval (CI) 23–45). Nineteen patients (29%) had monosomal karyotype (MK). Patients with MK had an OS of 21% (95% CI 7–41) at 2 years. Abnormal adverse cytogenetics, unrelated donor, bone marrow graft and CIBMTR score were significant risk factors for OS on univariate analysis. On multivariate analysis, abnormal adverse cytogenetics (hazard ratio (HR) 2.7; 95% CI 1.02–7.2; P-value = 0.02) and unrelated donor (HR 2.7; 95% CI 1.5–5.0; P-value = 0.0013) were independent factors for survival. Non-relapse mortality (NRM) at 2 years was 31% (95% CI 15–47). Donor type was the only factor that was significant for NRM with matched related donors having an NRM of 20% (95% CI 0–42) whereas unrelated donors had NRM of 60% (95% CI 40–80; P-value = 0.0007). In conclusion, patients with t-AML/MDS have poor OS. Unrelated donor is a significant risk factor for both higher NRM and decreased OS. Cytogenetics are predictive for OS. Bone Marrow Transplantation (2015) 50, 1180–1186; doi:10.1038/bmt.2015.151; published online 29 June 2015

INTRODUCTION Therapy-related AML/myelodysplastic syndromes (t-AML/MDS) are now considered a single entity, called therapy-related myeloid neoplasms based on the current WHO classification.1 t-AML/MDS can be seen after chemotherapy and radiation for several malignancies, as well as after some disease-modifying agents in non-malignant diseases.2–4 The risk after adjuvant treatment for breast cancer has been described as 1.25% and after treatment for Hodgkin’s lymphoma as 0.9–3.2%.5–7 The risk of developing cytogenetic abnormalities after autologous transplantation can be in the range of 9%.8 The incidence of therapy-related myeloid neoplasm continues to rise with better therapies for primary malignancies and better supportive care, hence patients with cancers are able to live longer.5,9 These therapy-related myeloid neoplasms usually carry high-risk cytogenetics and have been shown to have a poor prognosis mainly with current treatments available.10–12 Chemotherapy is usually non-curative unless in the setting of therapy-related APL or selected patients with good risk cytogenetics. Hence, if a suitable donor is identified, these patients are usually referred for allogeneic hematopoietic cell transplantation (HCT) which is considered a potentially curative treatment. However, allogeneic HCT is associated with significant morbidity and mortality. Limited data are available as to which patients likely will benefit from this modality. One of the factors contributing towards these poor outcomes is adverse disease biology. The incidence of monosomal karyotype (MK) in de novo adult AML is around 10–19%,13–15 but in t-AML it can be up to 24% which is considerably higher.14 Previous studies

have not explored outcomes of monsosomal karyotype in the setting of t-AML/MDS.14,15 In addition, comorbidities are often a major issue for t-AML/MDS patients as they may have complications from previous therapies. Prior studies from registries did not include information about comorbidities in their analysis.16,17 We studied these factors in t-AML/MDS patients following allogeneic HCT at our center. PATIENTS AND METHODS Patients All consecutive patients who were diagnosed with t-AML/MDS and received allogeneic HCT between 1996 and 2012 at Princess Margaret Cancer Center, Toronto, Ontario were included. The study was approved by the Research Ethics Board (REB) and registry data access committee of university health network (REB #11-0732-CE). All patients undergoing transplant in our program are asked to indicate and sign if they agree with data collection and involvement in retrospective studies. Because this was a retrospective study, a separate informed consent was waived by REB. Details of transplant procedures and institutional policies have been previously published.18,19

Data The data were collected through retrospective chart review, as well as blood and marrow transplant program database. The following variables were collected and analyzed: previous malignancy, age at diagnosis of that malignancy (continuous), type and stage of prior malignancy, type of therapy, duration between prior and current malignancy, type of current malignancy, cytogenetics, age at transplant, gender, donor type, graft

1 Allogeneic Blood and Marrow Transplant Program, Princess Margaret Cancer Center, University of Toronto, Toronto, Ontario, Canada and 2Department of Biostatistics, Princess Margaret Cancer Center, University of Toronto, Toronto, Ontario, Canada. Correspondence: Dr N Alam, Allogeneic Blood and Marrow Transplant Program, Princess Margaret Cancer Center, University of Toronto, 610 University Avenue, Toronto, Ontario M5G2M9, Canada. E-mail: [email protected] Received 8 January 2015; revised 14 April 2015; accepted 13 May 2015; published online 29 June 2015

Patients with t-AML/MDS after HCT N Alam et al

1181 source, conditioning regimen, CMV status for donor and recipient, GvHD prophylaxis, acute and chronic GvHD, engraftment data and causes of death.

Prior therapy Therapy received before developing t-AML/MDS was considered intense if patients had received two or more lines of chemotherapy, as well as radiation and/or autologous transplantation. If they had received only one line of chemotherapy and/or radiation they were considered to be in the moderate group, and if they had received non-chemotherapy agents or only radiation they were considered to be in the mild group.

Donors Donors were divided into related and unrelated. Low-resolution serological typing was performed for HLA class I and high-resolution molecular typing for HLA class II at our center till 1998 for donors. From 1999, HLA class I was done by molecular low resolution till June 2011 whereas class II remained at molecular high resolution. After June 2011, HLA class I was also performed at molecular high resolution.

Cytogenetics Medical research council (MRC) classification was used for cytogenetics of AML.20 Revised International prognostic scoring system (R-IPSS) was used to classify cytogenetics of MDS.21 Cytogenetics were then further divided into normal, abnormal adverse and abnormal non-adverse. Patients whose cytogenetics was not known were kept as separate entity as unknown.

MK MK was defined as presence of two or more autosomal monosomies or one monosomy along with the presence of additional structural abnormalities as per previously published criteria.15

Risk scores We evaluated four previously published risk scores: HCT-CI, CIBMTR, EBMT and the recently published risk score of comorbidity-age index.16,17,22,23 The variables for these scores were calculated retrospectively based on co morbidities and other factors from retrospective chart reviews. HCT-CI scores were divided into 0, 1 and 41 based on the number of patients in each group. EBMT scores were assigned as per Kroger et al.17, as low risk which had score of 0 and 1, moderate risk for score of 2 and high risk as score of 3 and 4. Because HCT-CI assigns a score of 3 for prior solid tumors which could give a bias to a study like this, HCT-CI score was evaluated both ways: by assigning a score of 3 for prior history of solid tumors and without it. CIBMTR scores were divided into three groups with 0, 1 in one group, score of 2 in second group and score of 3,4 in third group based on the number of patients. Comorbidity-age index was divided into four groups; 0 in first group, 1 and 2 in second group, 3 and 4 in third and ⩾ 5 in fourth group as per the original paper.

Definitions Overall survival (OS) was defined as time from day of transplant to death from any cause. Patients who were alive at their last follow-up were censored. Non-relapse mortality (NRM) was defined as death from any cause other than relapse. Relapse was defined as time to onset of recurrent t-AML/MDS through morphologic evidence in bone marrow or extramedullary sites. Relapse of primary malignancy was defined as identification of original malignancy. Neutrophil engraftment was defined as an absolute neutrophil count of 0.5 × 109 or higher for 3 days. Platelet engraftment was defined as platelet count higher than 20 × 109 for 7 days unsupported by transfusion. Acute and chronic GvHD were graded according to previously published criteria.24,25

Statistical analysis Patient demographics, and disease- and transplant-related factors were summarized with descriptive statistics. The main outcome of interest was OS. This was calculated using the Kaplan–Meier product method. NRM was calculated with relapse as competing risk factor. Similarly, cumulative incidence of relapse was calculated with death due to NRM as competing risk factor. © 2015 Macmillan Publishers Limited

Log-rank test for OS and Fine and Gray’s method for cumulative incidence of relapse and NRM were used to evaluate the risk factors on univariate analysis. The risk factors evaluated are described in data section. Owing to small sample size, only a limited multivariate analysis was done using the risk factors that were significant on univariate analysis.

RESULTS Patient, disease and transplant-related characteristics A total of 65 patients had allogeneic HCT for t-AML/MDS at Princess Margaret Cancer Center from 1996 to 2012. Patient characteristics, disease- and transplant-related data are summarized in Table 1. Among patients with AML (n = 32), 25 (78%) were transplanted in first CR, 4 (13%) in second CR and 3 (9%) had allogeneic transplant with active disease. All MDS patients had o 10% blasts in their marrow prior to HCT. Prior malignancy was hematological in 39 patients which included Hodgkin’s disease (n = 11; 28%), acute promyelocytic leukemia (n = 1; 2%), de novo AML (n = 1; 2%), multiple myeloma (n = 3; 8%), B-cell NHL (n = 17; 44%), T-cell lymphoma (n = 4; 10%) and chronic lymphocytic leukemia (n = 2; 5%). There were 24 patients who developed t-AML/MDS after solid tumors. These comprised breast cancer (n = 14; 58%), thyroid (n = 2, 8%) prostate (n = 1, 4%), ependymoma (n = 1; 4%), sarcoma (n = 3; 13%) endometrial cancer (n = 1; 4%), giant-cell tumor (n = 1; 4%) and testicular cancer (n = 1; 4%). Two patients (3%) developed t-AML/MDS after being treated for Behcet’s disease and psoriatic arthritis. Median age was 52.8 (range 18.9–69.4) years. In all, 35 (53.8%) patients received transplant from related donors, whereas 30 (46.2%) were unrelated. Median follow-up of survivors was 71.9 (range 15.9–204.1) months. Therapy prior to HCT Thirty one AML patients received induction chemotherapy prior to transplant and went into CR before transplantation. Out of those, 20 patients received 1 or 2 cycles of consolidation before transplant and still remained in good condition to go ahead with HCT. One patient was transplanted not in remission. Among 33 patients with MDS, 8 patients received cytoreductive therapy before allogeneic HCT. Out of those one patient received induction and six received azacitidine and one received decitabine. As this study is from 1996 to 2012, several conditioning regimens were used over the years. OS The estimated OS of the whole population at 2 years was 34% (95% confidence interval (CI) 22–45; Figure 1). On univariate analysis, cytogenetics, donor type and graft source were significant risk factors for OS (Table 2). In our analysis, we divided cytogenetics into normal, abnormal adverse and abnormal non-adverse as described in Patients and methods section. Cytogenetics were unknown for nine patients most of whom had transplant before the year 2000. On univariate analysis, patients with abnormal adverse cytogenetics had an OS of 21% (95% CI 9–38) at 2 years, whereas patients with abnormal non-adverse cytogenetics had an OS of 53% (95% CI 29–72). Patients with normal cytogenetics had an OS of 44% (95% CI 14–72) and patients with unknown cytogenetics had an OS of 22% (95% CI 3–51; P-value = 0.04; Table 2). Donor type and graft source were also significant risk factors on univariate analysis. Unrelated donors had an OS of 23% (95% CI 10–39), whereas related donors had an overall survival of 43% (95% CI 26–58) at 2 years, P-value 0.003 (Figure 2b). Graft source of bone marrow had an OS of 31% (95% CI 11–53), whereas PBSCs Bone Marrow Transplantation (2015) 1180 – 1186

Patients with t-AML/MDS after HCT N Alam et al

1182 Table 1.

Patient and HCT characteristics

Characteristics of patients Number of patients Median age at HCT, years (range) Gender Male Female Prior malignancy Solid malignancy Hematological malignancy Other Prior therapy Intense (chemotherapy ⩾ 2, radiation, autologous transplant) Moderate (chemotherapy, radiation) Mild (non-chemo agents, radiation) NA Median time from previous diagnosis to current diagnosis, years (range) Current disease MDS AML Cytogenetics Favorable Intermediate Adverse NA Donor type Matched related donor Unrelated donor Graft type Bone marrow PBSC GvHD prophylaxis CSA/MMF CSA/MTX Campath/CSA Others NA Conditioning regimen Myelablative Reduced intensity CMV serostatus R − /D − R+/D − R − /D+ R+/D+ NA Performance status at HCT 0 ⩾1 NA HCT-CI score 0 1 41 NA EBMT score Score 0/1 Score 2 Score 3/4 CIBMTR score Score 0/1 Score 2 Score 3/4 Comorbidity-age index Score 0 Score 1/2 Score 3/4 Score ⩾ 5 NA

Overall population, n (%) Patients without MK, n (%) Patients with MK, n (%) P-value 65 (100) 52.8 (18.9–69.4)

46 (71) 52.9 (18.9–69.4)

19 (29) 52.6 (33.6–66.2)

27 (42) 38 (58)

18 (39) 28 (61)

9 (47) 10 (53)

24 (37) 39 (60) 2 (3)

20 (43) 26 (57)

4 (21) 13 (68) 2 (11)

21 (32)

11 (24)

10 (53)

36 7 1 7

28 6 1 7

0.60a 0.54 0.15

(55) (11) (2) (0.9–31)

(61) (13) (2) (0.9–31)

0.10a

8 (42) 1 (5) 6.7 (2–21)

0.67a 0.07

33 (51) 32 (49)

20 (43) 26 (57)

13 (68) 6 (32)

2 26 28 9

2 24 11 9

(4) (52) (24) (20)

2 (11) 17 (89)

35 (54) 30 (46)

23 (50) 23 (50)

12 (63) 7 (37)

16 (25) 49 (75)

13 (28) 33 (72)

3 (16) 16 (84)

18 26 12 7 2

10 20 10 5 1

(3) (40) (43) (14)

o0.0001a

0.33

(28) (40) (18) (11) (3)

(22) (43) (22) (11) (2)

8 6 2 2 1

0.36a 0.38a

(42) (32) (11) (11) (5) 0.40

36 (55) 29 (45)

27 (59) 19 (41)

18 14 14 15 4

13 10 7 12 4

(28) (22) (22) (23) (6)

(28) (22) (15) (26) (9)

9 (47) 10 (53) 5 4 7 3

0.38a

(26) (21) (37) (16) 0.94

40 (62) 24 (37) 1 (1)

28 (61) 17 (37) 1 (2)

28 16 20 1

19 13 13 1

12 (63) 7 (37) 0.53

(43) (25) (31) (1)

(42) (28) (28) (2)

9 (47) 3 (16) 7 (37)

10 (15) 17 (26) 38 (58)

8 (17) 17 (37) 21 (46)

2 (11) 0 (0) 17 (89)

17 (26) 31 (48) 17 (26)

16 (35) 22 (48) 8 (17)

1 (5) 9 (47) 9 (47)

4 28 15 17 1

4 17 11 13 1

(6) (44) (23) (26) (1)

(9) (37) (24) (28) (2)

0.001a

0.01 0.42a

0 11 (58) 4 (21) 4 (21)

Abbreviations: CIBMTR = center for international bone marrow transplant research; CSA = cyclosporine; D = donor; EBMT = European bone marrow transplant; HCT = hematopoietic cell transplant; HCT-CI = hematopoietic cell transplant-comorbidity index; MDS = myelodysplastic syndrome; MK = monosomal karyotype; MMF = mycophenolate mofetil; MTX = methotrexate; NA = not available; R = recipient. aKruskal–Wallis Test for continuous and Fisher’s exact test for categorical.

Bone Marrow Transplantation (2015) 1180 – 1186

© 2015 Macmillan Publishers Limited

Patients with t-AML/MDS after HCT N Alam et al

1183 were found to have an OS of 35% (95% CI 21–47) at 2 years (P-value = 0.04; Figure 2c). On multivariate analysis, unrelated donor remained a significant risk factor for OS with a hazard ratio (HR) of 2.714 (95% CI 1.477–4.989), P-value 0.0013. Abnormal adverse cytogenetics also remained significant with HR of 2.708 (95% CI 1.023–7.169), whereas abnormal non-adverse cytogentics had a HR of 1.100 (95% CI 0.368–3.286) and unknown cytogenetics had a HR of 3.438 (95% CI 1.032–11.458), P-value 0.02 (Table 3). Graft source and CIBMTR were taken to the multivariate analysis along with donor type and cytogenetics and did not turn out to be significant. 100

KM curve Lower 95% Point estimate Upper 95%

Overall survival (%)

80

Donor type was further investigated to evaluate the impact of mismatch on OS. Out of 30 unrelated donors, 1 had 2/8 mismatch, 3 had 1/8 mismatch and 2 had 1/0 mismatch. Mismatch did make a difference in outcome as OS at 1 year was 25% (95% CI 10–43) for patients who received HCT from 10/10 matched unrelated donors (n = 24), whereas among unrelated donors who had at least 1 mismatch (n = 6) OS was 33% (95%CI 5–68), P-value 0.662. Separate Ag vs allele typing was not analyzed due to low numbers. NRM In the current study, NRM at 2 years was 31% (95% CI 15–47). The only risk factor that was found to be significant for NRM on univariate analysis was donor type. Unrelated donors had NRM of 60% (95% CI 40–80) at 2 years, whereas matched related donors had NRM of 20% (95% CI 0–412%; P-value 0.0007; Table 2).

60

Relapse risk When risk of relapse was evaluated using Fine and Gray’s method for competing risk, the only risk factor that remained significant was age at transplant. In patients o 60 years a relapse rate of 30% (95% CI 10–49) was observed, whereas in patients ⩾ 60 years a relapse rate of 36% (95% CI 0–73) was seen, P-value 0.03 (Table 2).

40

20

0 0

5

10

15

20

25

30

35

40

Patient and transplant characteristics for MK patients Nineteen patients (29%) among the above 65 patients with t-AML/MDS had MK. Median age of patients with MK was 52.6

Months since transplant

Figure 1. Table 2.

OS of all patients with t-AML/MDS.

Univariate analysis for probabilities of outcomes of relapse, non-relapse mortality and overall survival at 2 years after HCT

Risk factors

Relapse Cumulative incidence (95% CI)

Cytogenetics Normal Abnormal adverse Abnormal non-adverse Unknown Age at transplant (as continuous) o60 years ⩾ 60 years Donor type Matched related donor Unrelated donor Graft source Bone marrow PBSC HCT-CI score 0 1 ⩾2 CIBMTR score 0–1 2 3–4 EBMT score 0–1 2 3–4 Comorbidity-age index score 0 1–2 3–4 ⩾5

Non-relapse mortality P-value

Cumulative incidence (95% CI)

0.08 0.11 0.29 0.26 0.67

(0.00–0.46) (0.00–0.59) (0–0.54) (0.15–1)

0.03

0.30 (0.10–0.49) 0.36 (0.00–0.73)

(0.10–0.79) (0.27–0.73) (0–0.54) (0–1)

0.31

0.30

0.96

0.13

0.73

0.92 0.50 (0.18–0.75) 0.35 (0.14–0.57) 0.29 (0.16–0.44)

0.94 0 0.43 0.40 0.35

(0) (0.19–0.67) (0.10–0.70) (0.02–0.68)

0.03

0.40 (0.18–0.62) 0.42 (0.25–0.58) 0.12 (0.02–0.31)

0.50 (0.07–0.93) 0.41 (0.08–0.75) 0.39 (0.17–0.61) 0.25

0.04 0.54

0.43 (0.25–0.60) 0.19 (0.05–0.40) 0.35 (0.16–0.55)

0.19 (0–0.52) 0.35 (0.13–0.58) 0.63 (0.31–0.94) 0.54

0.003

0.31 (0.11–0.54) 0.35 (0.22–0.48)

0.29 (0.04–0.53) 0.56 (0.24–0.89) 0.35 (0.03–0.67)

0 (0–0) 0.29 (0–0.67) 0.33 (0.10–0.56)

0.65 0.33 (0.21–0.46) 0.36 (0.13–0.59)

0.46

0.54

0.38 (0.06–0.71) 0.29 (0.05–0.53) 0.25 (0–0.82)

(0.14–0.72) (0.09–0.38) (0.29–0.72) (0.03–0.51)

0.43 (0.26–0.58) 0.23 (0.10–0.39)

0.44 (0–0.88) 0.37 (0.19–0.54)

0.32 (0.08–0.57) 0.25 (0–0.74) 0.35 (0.04–0.66)

(0–0.12) (0.01–0.56) (0.20–0.73) (0–0.64)

0.0007

P-value 0.04

0.44 0.21 0.53 0.22

0.20 (0.00–0.42) 0.60 (0.40–0.80)

0.40 (0–0.88) 0.29 (0.10–0.47)

Probability (95% CI)

0.23

0.24

0.04 0.29 0.47 0.29

P-value 0.17

0.44 0.50 0.26 0.22

0.41 (0.24–0.58) 0.29 (0–0.68)

0.41 (0.20–0.61) 0.20 (0.00–0.53)

Overall survival

0.18 1.00 0.32 0.20 0.35

(1.00–1.00) (0.16–0.49) (0.05–0.42) (0.14–0.57)

Abbreviations: CI = confidence interval; CIBMTR = center for international bone marrow transplant research; EBMT = European bone marrow transplant; HCT = hematopoietic cell transplant; HCT-CI = hematopoietic cell transplant-comorbidity index. All significant P-values are in bold.

© 2015 Macmillan Publishers Limited

Bone Marrow Transplantation (2015) 1180 – 1186

Patients with t-AML/MDS after HCT N Alam et al

1184

a

Table 3.

100

Multivariate analysis for overall survival after HCT

Monosomal karyotype NEG POS

Variable

Risk

Overall survival (%)

80

Donor type Unrelated donor Cytogenetics Abnormal adverse Abnormal non-adverse Unknown

60

40

Abbreviations: transplant.

20

CI = confidence

Hazard ratio

95% CI

P-value

2.71 2.71 1.10

1.48–4.99 1.02–7.17 0.37–3.29

0.001 0.02

3.44

1.03–11.46

interval;

HCT = hematopoietic

cell

0 0

48

24

72

96

120

144

168

192

216

240

t-AML/MDS after prior treatment for solid tumors which comprised sarcoma (n = 1; 5%) endometrial cancer (n = 1; 5%), giant-cell tumor (n = 1; 5%) and testicular cancer (n = 1; 5%). t-AML/MDS with MK resulted in 2 patients (11%) after being treated for Behcet’s disease and psoriatic arthritis.

Months since transplant

b 100

P = 0.0032 Related donor Unrelated donor

40

Evaluation of comorbidity scores in the setting of t-AML/MDS Several risk scores have been evaluated and published for prediction of outcomes after HCT. We evaluated the risk scores of HCT-CI, CIBMTR, EBMT and Comorbidity-age index as described in the Patients and methods section. CIBMTR was the only risk score that had any predictive value on univariate analysis and that also was only for OS. Patients with a risk score of 0 and/or 1 by CIBMTR had an OS of 40% (95% CI 18– 62) at 2 years; whereas patients with a risk score of 2 had an OS of 42% (95% CI 25–58) and patients with risk score of 3 and/or 4 had an OS of 12% (95% CI 2–31), P-value 0.03. CIBMTR score; however, was not predictive of NRM in these patients (Table 3), moreover CIBMTR did not remain predictive for OS when taken into multivariate analysis. HCT-CI, EBMT scores and Comorbidity-age index were also evaluated for OS and NRM and did not turn out to be significantly predictive for any of these clinical end points in these patients (Table 3). The highest HCT-CI score seen was 7, (n = 2). Although HCT-CI score assigns a point of 3 for previous history of solid malignancy, we evaluated HCT-CI with and without scoring for previous malignancy. However, in either case HCT-CI remained non-predictive in these patients.

40

Causes of death Overall, there were 51 deaths. Relapse remained a major cause of death with 21 (41%) patients dying of relapse of their myeloid malignancy. Infection was the second common cause with 16 (31%) and GvHD as the third cause with 7 (14%) patients. Six patients (12%) died of regimen-related toxicity and one (2%) died of relapse of their previous malignancy.

Overall survival (%)

80

60

40

20

0 0

5

10

15

20

25

30

35

Months since transplant

c

100

P = 0.0393 BM PBSC

Overall survival (%)

80

60

40

20

0 0

5

10

15

20

25

30

35

Months since transplant

Figure 2. (a) OS of patients with t-AML/MDS and MK. (b) OS of patients with t-AML/MDS by donor type. (c) OS of patients with t-AML/MDS by graft source. BM = bone marrow; neg = negative; PBSC = peripheral blood stem cells; pos = positive.

(range 33.6–66.2) years. Details of patient characteristics, disease and transplant-related data are shown in Supplementary Table 1. Prior malignancy was hematological in 13 (68%) patients whereas solid tumors were present in 4 (21%) patients, odds ratio 0.133 (95% CI 0.037–0.484), P-value = 0.004. Two patients (11%) had prior treatment for benign disorders. Hematological malignancy in these patients included Hodgkin’s disease (n = 3; 16%), de novo AML (n = 1; 5%), multiple myeloma (n = 1; 5%), B-cell NHL (n = 8; 42%). Four patients developed Bone Marrow Transplantation (2015) 1180 – 1186

Outcomes of patients with MK OS of patients with MK remained poor compared with patients who did not have MK. OS at 1, 2 and 3 years for patients with MK was 26% (95% CI 10–47), 21% (95% CI 7–41) and 16% (95% CI 4–35), respevtively, whereas for patients without MK an OS of 46% (95% CI 31–59), 39% (95% CI 25–53) and 31% (95% CI 18–45) was seen at 1, 2 and 3 years respectively, P-value 0.13. OS of patients with and without MK at 2 years are shown in Figure 2a. Eighteen out of 19 (95%) patients died following HCT. Relapse was the cause of death in 7 (39%) patients while 6 (33%) died of infections and 4 (22%) died of GvHD. One patient (6%) died of regimen-related toxicity. © 2015 Macmillan Publishers Limited

Patients with t-AML/MDS after HCT N Alam et al

DISCUSSION t-AML/MDS tends to be increasing as therapies for malignancies are getting better, as well as supportive care.5,9 This study presents a summary of 16 years experience of t-AML/MDS patients following HCT. We also evaluated outcomes for MK in t-AML/ MDS as a subgroup which to our knowledge has not been reported previously in this setting. Overall outcomes still remain poor for t-AML/MDS though outcomes for de novo AML seem to have improved.10,26,27 OS at 2 years was 34% in the present study which is substantially lower than de novo AML.28 Relapse remained a major cause of death in these patients with various transplant-related complications such as infections, GvHD and regimen-related toxicities as major contributory causes of NRM. Cytogenetics remained significant in current study for outcomes. Patients with abnormal adverse cytogenetics had significantly inferior outcomes compared with normal cytogenetics indicating the significant impact of poor disease biology on outcomes of HCT. Type of unrelated donor was also a very significant risk factor for poor OS, as well as higher NRM (Table 2). This is in contrast to recent reports of equivalent outcomes of related vs unrelated donors in high-risk leukemias.29 However, our study does include patients from 1996 and it is possible that the molecular low-resolution donor typing that was being done at our center till June of 2011 is partly responsible for these outcomes. Despite this, our findings are very similar to those reported in a large CIBMTR study.12 Patient with MK were also analyzed as a subgroup. Median age of these patients was 52.6 years which was similar to median age for overall population in our study. We found an OS of 16% at 3 years in patients with MK in setting of t-AML/MDS. These results are not statistically significant compared with non-MK karyotype perhaps due to small number of patients. Outcomes of MK in t-AML/MDS remained poor after allogeneic HCT which was what has been shown previously for de novo MK. However, we did not have assessment of cytogenetics before transplantation to see if they persisted and if that also had an impact on outcome. The outcome of t-AML/MDS with MK will need to be tested in a bigger data set, and if these results are validated then would indicate the need for novel approaches in this group of patients.30 Several risk scores have been proposed in the setting of HCT to predict outcomes based on comorbidities.16,17,22,23 In the present study, the only score that predicted outcomes on univariate analysis were CIBMTR score (Table 2). This is the only score that incorporates donor type as a risk factor and because in our study donor type had strong association with outcomes, it is plausible that this is the reason for CIBMTR risk score to be predictive. Other risk scores including the recently published score of comorbidity-age index were not significant, despite the fact that these patients tend to have more comorbidities and hence higher scores compared with de novo patients going for allogeneic HCT. The limitations of this study are that these are retrospective data. Also, the present study is in a single center setting. However, t-AML/MDS is becoming more common as therapies for primary malignancies are getting better and patients have a higher likelihood of living long enough to develop these secondary malignancies. Hence, it is very important to evaluate this in a larger data set and/ or prospective manner. Also as new therapeutic options such as azacitidine and/or decitabine are now available, these may constitute other strategies for these patients and their disease control and perhaps alter/modify disease course. In conclusion, our data indicate modest outcomes in patients with t-AML/MDS after HCT. Unrelated donor type was associated with lower OS and higher NRM. © 2015 Macmillan Publishers Limited

CONFLICT OF INTEREST The authors declare no conflict of interest.

AUTHOR CONTRIBUTIONS NA and VG contributed to the design of study, the supervision of data collection and of data interpretation, data analysis, and writing the manuscript. EGA prepared statistical plan, performed statistical analysis and interpreted results. DK, JU, HAM, JK, JHL and MS were involved in the supervision of the data collection and interpretation of the data. All authors critically reviewed and approved the final draft of the manuscript.

REFERENCES 1 Vardiman JW, Thiele J, Arber DA, Brunning RD, Borowitz MJ, Porwit A et al. The 2008 revision of the World Health Organization (WHO) classification of myeloid neoplasms and acute leukemia: rationale and important changes. Blood 2009; 114: 937–951. 2 Curtis RE, Boice JD Jr, Stovall M, Bernstein L, Holowaty E, Karjalainen S et al. Relationship of leukemia risk to radiation dose following cancer of the uterine corpus. J Natl Cancer Inst 1994; 86: 1315–1324. 3 Ratain MJ, Kaminer LS, Bitran JD, Larson RA, Le Beau MM, Skosey C et al. Acute nonlymphocytic leukemia following etoposide and cisplatin combination chemotherapy for advanced non-small-cell carcinoma of the lung. Blood 1987; 70: 1412–1417. 4 Curtis RE, Boice JD Jr, Stovall M, Bernstein L, Greenberg RS, Flannery JT et al. Risk of leukemia after chemotherapy and radiation treatment for breast cancer. N Engl J Med 1992; 326: 1745–1751. 5 Eichenauer DA, Thielen I, Haverkamp H, Franklin J, Behringer K, Halbsguth T et al. Therapy-related acute myeloid leukemia and myelodysplastic syndromes in patients with Hodgkin lymphoma: a report from the German Hodgkin Study Group. Blood 2014; 123: 1658–1664. 6 Fisher B, Rockette H, Fisher ER, Wickerham DL, Redmond C, Brown A. Leukemia in breast cancer patients following adjuvant chemotherapy or postoperative radiation: the NSABP experience. J Clin Oncol 1985; 3: 1640–1658. 7 Koontz MZ, Horning SJ, Balise R, Greenberg PL, Rosenberg SA, Hoppe RT et al. Risk of therapy-related secondary leukemia in Hodgkin lymphoma: the Stanford University experience over three generations of clinical trials. J Clin Oncol 2013; 31: 592–598. 8 Traweek ST, Slovak ML, Nademanee AP, Brynes RK, Niland JC, Forman SJ. Clonal karyotypic hematopoietic cell abnormalities occurring after autologous bone marrow transplantation for Hodgkin's disease and non-Hodgkin's lymphoma. Blood 1994; 84: 957–963. 9 Mailankody S, Pfeiffer RM, Kristinsson SY, Korde N, Bjorkholm M, Goldin LR et al. Risk of acute myeloid leukemia and myelodysplastic syndromes after multiple myeloma and its precursor disease (MGUS). Blood 2011; 118: 4086–4092. 10 Kayser S, Dohner K, Krauter J, Kohne CH, Horst HA, Held G et al. The impact of therapy-related acute myeloid leukemia (AML) on outcome in 2853 adult patients with newly diagnosed AML. Blood 2011; 117: 2137–2145. 11 Kern W, Haferlach T, Schnittger S, Hiddemann W, Schoch C.. Prognosis in therapyrelated acute myeloid leukemia and impact of karyotype. J Clin Oncol 2004; 22: 2510–2511. 12 Abdelhameed A, Pond GR, Mitsakakis N, Brandwein J, Chun K, Gupta V et al. Outcome of patients who develop acute leukemia or myelodysplasia as a second malignancy after solid tumors treated surgically or with strategies that include chemotherapy and/or radiation. Cancer 2008; 112: 1513–1521. 13 Feng R, Liu H, Chang N, Fan Y, Li J, Zhang Y et al. Monosomal karyotype among adult acute myeloid leukemia: clinical characteristic and prognostic analysis. Zhonghua Xue Ye Xue Za Zhi 2014; 35: 393–396. 14 Kayser S, Zucknick M, Dohner K, Krauter J, Kohne CH, Horst HA et al. Monosomal karyotype in adult acute myeloid leukemia: prognostic impact and outcome after different treatment strategies. Blood 2012; 119: 551–558. 15 Breems DA, Van Putten WL, De Greef GE, Van Zelderen-Bhola SL, Gerssen-Schoorl KB, Mellink CH et al. Monosomal karyotype in acute myeloid leukemia: a better indicator of poor prognosis than a complex karyotype. J Clin Oncol 2008; 26: 4791–4797. 16 Litzow MR, Tarima S, Perez WS, Bolwell BJ, Cairo MS, Camitta BM et al. Allogeneic transplantation for therapy-related myelodysplastic syndrome and acute myeloid leukemia. Blood 2010; 115: 1850–1857. 17 Kroger N, Brand R, van Biezen A, Zander A, Dierlamm J, Niederwieser D et al. Risk factors for therapy-related myelodysplastic syndrome and acute myeloid leukemia treated with allogeneic stem cell transplantation. Haematologica 2009; 94: 542–549.

Bone Marrow Transplantation (2015) 1180 – 1186

1185

Patients with t-AML/MDS after HCT N Alam et al

1186 18 Alam N, Marras TK, Atenafu EG, Gupta V, Kuruvilla J, Lipton JH et al. Allogeneic peripheral blood stem cell transplantation significantly increases risk of chronic graft-versus-host disease of lung compared with bone marrow transplantation. Biol Blood Marrow Transplant 2012; 18: 1905–1910. 19 Alam N, Atenafu EG, Tse G, Viswabandya A, Gupta V, Kim D et al. Limited benefit of pentostatin salvage therapy for steroid-refractory grade III-IV acute graft-versus-host disease. Clin Transplant 2013; 27: 930–937. 20 Grimwade D, Hills RK, Moorman AV, Walker H, Chatters S, Goldstone AH et al. Refinement of cytogenetic classification in acute myeloid leukemia: determination of prognostic significance of rare recurring chromosomal abnormalities among 5876 younger adult patients treated in the United Kingdom Medical Research Council trials. Blood 2010; 116: 354–365. 21 Greenberg PL, Tuechler H, Schanz J, Sanz G, Garcia-Manero G, Sole F et al. Revised international prognostic scoring system for myelodysplastic syndromes. Blood 2012; 120: 2454–2465. 22 Sorror ML, Maris MB, Storb R, Baron F, Sandmaier BM, Maloney DG et al. Hematopoietic cell transplantation (HCT)-specific comorbidity index: a new tool for risk assessment before allogeneic HCT. Blood 2005; 106: 2912–2919. 23 Sorror ML, Storb RF, Sandmaier BM, Maziarz RT, Pulsipher MA, Maris MB et al. Comorbidity-age index: a clinical measure of biologic age before allogeneic hematopoietic cell transplantation. J Clin Oncol 2014; 32: 3249–3256. 24 Przepiorka D, Weisdorf D, Martin P, Klingemann HG, Beatty P, Hows J et al. 1994 Consensus conference on acute GVHD grading. Bone Marrow Transplant 1995; 15: 825–828.

25 Filipovich AH, Weisdorf D, Pavletic S, Socie G, Wingard JR, Lee SJ et al. National Institutes of Health consensus development project on criteria for clinical trials in chronic graft-versus-host disease: I. Diagnosis and staging working group report. Biol Blood Marrow Transplant 2005; 11: 945–956. 26 Smith SM, Le Beau MM, Huo D, Karrison T, Sobecks RM, Anastasi J et al. Clinical-cytogenetic associations in 306 patients with therapy-related myelodysplasia and myeloid leukemia: the University of Chicago series. Blood 2003; 102: 43–52. 27 Lekakis LJ, Cooper BW, de Lima MG. Allogeneic stem cell transplantation for acute myeloid leukemia in first complete remission: are we closer to knowing who needs it? Curr Hematol Malig Rep 2014; 9: 128–137. 28 Devillier R, Harbi S, Furst S, Crocchiolo R, El-Cheikh J, Castagna L et al. Poor outcome with nonmyeloablative conditioning regimen before cord blood transplantation for patients with high-risk acute myeloid leukemia compared with matched related or unrelated donor transplantation. Biol Blood Marrow Transplant 2014; 20: 1560–1565. 29 Gupta V, Tallman MS, He W, Logan BR, Copelan E, Gale RP et al. Comparable survival after HLA-well-matched unrelated or matched sibling donor transplantation for acute myeloid leukemia in first remission with unfavorable cytogenetics at diagnosis. Blood 2010; 116: 1839–1848. 30 Orozco JJ, Appelbaum FR. Unfavorable, complex, and monosomal karyotypes: the most challenging forms of acute myeloid leukemia. Oncology 2012; 26: 706–712.

Supplementary Information accompanies this paper on Bone Marrow Transplantation website (http://www.nature.com/bmt)

Bone Marrow Transplantation (2015) 1180 – 1186

© 2015 Macmillan Publishers Limited