Abstract ORIGINAL RESEARCH ARTICLE

Transcription

1 Received: 17 October 2018 Accepted: 23 October 2018 DOI: /hon.2566 ORIGINAL RESEARCH ARTICLE Induction chemotherapy followed by allogeneic HCT versus upfront allogeneic HCT for advanced myelodysplastic syndrome: A propensity score matched analysis Takaaki Konuma 1 Yoshimitsu Shimomura 2 Yukiyasu Ozawa 3 Yasunori Ueda 4 Naoyuki Uchida 5 Makoto Onizuka 6 Megumi Akiyama 7 Takehiko Mori 8 Hirohisa Nakamae 9 Yuju Ohno 10 Souichi Shiratori 11 Yasushi Onishi 12 Yoshinobu Kanda 13 Takahiro Fukuda 14 Yoshiko Atsuta 15,16 Ken Ishiyama 17 Adult Myelodysplastic Syndrome Working Group of the Japan Society for Hematopoietic Cell Transplantation 1 Department of Hematology/Oncology, The Institute of Medical Science, The University of Tokyo, Tokyo, Japan 2 Department of Hematology, Kobe City Hospital Organization, Kobe City Medical Center General Hospital, Kobe, Japan 3 Department of Hematology, Japanese Red Cross Nagoya First Hospital, Nagoya, Japan 4 Department of Hematology and Oncology, Kurashiki Central Hospital, Kurashiki, Japan 5 Department of Hematology, Federation of National Public Service Personnel Mutual Aid Associations Toranomon Hospital, Tokyo, Japan 6 Department of Hematology and Oncology, Tokai University School of Medicine, Isehara, Japan 7 Hematology Division, Tokyo Metropolitan Komagome Hospital, Tokyo, Japan 8 Division of Hematology, Department of Medicine, Keio University School of Medicine, Tokyo, Japan 9 Hematology, Graduate School of Medicine, Osaka City University, Osaka, Japan 10 Department of Internal Medicine, Kitakyushu Municipal Medical Center, Kitakyushu, Japan 11 Department of Hematology, Hokkaido University Hospital, Sapporo, Japan 12 Department of Hematology and Rheumatology, Tohoku University Hospital, Sendai, Japan 13 Division of Hematology, Jichi Medical University Saitama Medical Center, Saitama, Japan Abstract To reduce post transplant relapse, acute myeloid leukemia (AML) type remission induction chemotherapy has been attempted to reduce disease burden before allogeneic hematopoietic cell transplantation (HCT) in patients with advanced myelodysplastic syndrome (MDS). However, the efficacy of induction chemotherapy before HCT is unclear. We retrospectively analyzed the Japanese registration data of 605 adult patients, who had received allogeneic HCT for advanced MDS between 2001 and 2016, to compare the post transplant relapse between patients who received induction chemotherapy followed by allogeneic HCT and those who received upfront HCT. Propensity score matching identified 230 patients from each cohort. There were no significant differences in overall survival and non relapse mortality between the two groups. The cumulative incidence of relapse was significantly higher in patients who received induction chemotherapy than those who received upfront HCT. In the subgroup analyses, upfront HCT had a significantly reduced relapse incidence among patients with poor cytogenetics, those with higher international prognostic scoring system at diagnosis, and those who received reducedintensity conditioning. Our results suggested that AML type remission induction chemotherapy before HCT did not improve post transplant relapse and survival for adult patients with advanced MDS. Upfront HCT is preferable for patients with a poor karyotype. KEYWORDS myelodysplastic syndrome, allogeneic hematopoietic cell transplantation, induction chemotherapy, cytoreductive treatment, relapse, propensity score matched analysis 14 Department of Hematopoietic Stem Cell Transplantation, National Cancer Centre Hospital, Tokyo, Japan Hematological Oncology. 2019;37: wileyonlinelibrary.com/journal/hon 2018 John Wiley & Sons, Ltd. 85

2 86 KONUMA ET AL. 15 Japanese Data Center for Hematopoietic Cell Transplantation, Nagoya, Japan 16 Department of Healthcare Administration, Nagoya University Graduate School of Medicine, Nagoya, Japan 17 Department of Hematology, Kanazawa University Hospital, Kanazawa, Japan Correspondence Takaaki Konuma, MD, PhD., Department of Hematology/Oncology, The Institute of Medical Science, The University of Tokyo, 4 6 1, Shirokanedai, Minato ku, Tokyo , Japan. tkonuma@ims.u tokyo.ac.jp Funding information the Practical Research Project for Allergic Diseases and Immunology (Research Technology of Medical Transplantation) from Japan Agency for Medical Research and Development, AMED, Grant/Award Number: 18ek h INTRODUCTION Allogeneic hematopoietic cell transplantation (HCT) is the only potential curative therapeutic modality for myelodysplastic syndrome (MDS). The higher proportion of marrow blasts at HCT has been shown to be associated with post transplant relapse in patients with advanced MDS. 1,2 Therefore, acute myeloid leukemia (AML) type remission induction chemotherapy has been used to reduce disease burden before allogeneic HCT in some patients with advanced MDS. However, whether the decreased disease burden by induction chemotherapy reflects a selection bias of patients with a more favorable disease profile is unclear. Indeed, because the efficacy of induction chemotherapy before allogeneic HCT using myeloablative conditioning has not been demonstrated for patients with advanced MDS, some patients receive upfront HCT as an initial treatment for MDS. 3-9 Recent advances in the use of reduced intensity conditioning (RIC) and/or alternative donors, such as unrelated cord blood, may have changed the role of induction chemotherapy before HCT or upfront HCT as an initial treatment for patients with advanced MDS. However, there have been no randomized clinical trials published to date comparing the post transplant relapse of patients who received induction chemotherapy before HCT with those who received upfront HCT as an initial treatment for advanced MDS. Therefore, we performed a propensity score matched analysis to clarify the efficacy of AML type remission induction chemotherapy before HCT for advanced MDS using data from a nationwide Japanese database. received the first allogeneic HCT between 2001 and 2016 in Japan. Advanced MDS was defined as refractory anemia with an excess of blasts (RAEB) according to the French American British (FAB) classification 13 for patients undergoing HCT before 2007 or as RAEB 1 and RAEB 2 according to the World Health Organization (WHO) classification of for patients undergoing HCT after Patients with a diagnosis of RAEB in transformation, chronic myelomonocytic leukemia, or therapy related myeloid neoplasm were excluded from this study. Human leukocyte antigen (HLA; HLA A, B, DR) mismatched bone marrow transplant/peripheral blood stem cell transplant (BMT/ PBSCT) was also excluded. We excluded patients who received allogeneic HCT at relapse or a second or subsequent complete remission. Patients with an interval between diagnosis and HCT of more than 12 months were also excluded. We also excluded patients with missing data for the International Prognostic Scoring System (IPSS) score at diagnosis and karyotype. To clarify the impact of AML type induction chemotherapy before HCT on post transplant relapse, patients who received pretransplant treatment other than AML type remission induction chemotherapy, such as low dose chemotherapy and azacitidine (Aza), were also excluded. Finally, 605 patients were eligible for this study (Figure 1). This retrospective study was approved by the institutional review board of the Institute of Medical Science, The University of Tokyo (30 24 B0702). 2.2 Definitions 2 PATIENTS AND METHODS 2.1 Data collection Data were obtained from the Transplant Registry Unified Management Program (TRUMP) of the Japanese Data Center for Hematopoietic Cell Transplantation (JDCHCT) This retrospective study included patients aged 16 to 70 years with de novo advanced MDS who The primary objective of this retrospective study was to compare the post transplant relapse incidence of patients who received AML type induction chemotherapy followed by HCT with those who received upfront HCT. Overall survival (OS) was defined as death from any cause. Relapse was defined as morphological evidence of MDS. Patients who never achieved remission following HCT were considered to have had a relapse on day 1 after HCT. Non relapse mortality (NRM) was defined as death without relapse. The karyotype risk was classified according to the IPSS criteria. 15 Monosomal karyotype

3 KONUMA ET AL. 87 FIGURE 1 Flow chart of patient selection strategy (MK) was classified according to the original definition for patients with AML, 16 which were either two or more autosomal monosomies or one autosomal monosomy with other structural abnormalities. The conditioning regimens were classified as a myeloablative conditioning (MAC) regimen if the following were included in the regiment: oral busulfan doses of 9 mg/kg, intravenous busulfan doses of 7.2 mg/kg, total body irradiation (TBI) single doses of 5 Gy or fractionated doses totaling 8 Gy, or melphalan doses of 140 mg/m 2. Other regimens were classified as reduced intensity conditioning (RIC), according to a previous report of the Center for International Blood and Marrow Transplant Research (CIBMTR) Propensity score matching The baseline characteristics of patients and disease could have influenced the decision whether to perform induction chemotherapy before HCT and the post transplant relapse. Therefore, we performed propensity score matched analyses to minimize potential treatment selection bias and eliminate confounding bias in the overall cohort. Logistic regression was used for the propensity score calculation from the following variables: age, sex, karyotype, IPSS at diagnosis, donor source, intensity of conditioning regimen, and year of HCT. A 1:1 matching by propensity score was performed using the nearest neighbor matching method with a caliper width fixed at 0.2. Propensity score matching was carried out using JMP Pro (SAS Institute Japan, Co., Ltd., Tokyo, Japan). 2.4 Statistical analysis To compare differences of variables between the two groups, a chisquare test or Fisher's exact test was used for categorical variables, and the Kruskal Wallis test was used for continuous variables. The probability of OS was estimated according to the Kaplan Meier method, and the log rank test was used to compare the groups. The probability of relapse and NRM were estimated according to cumulative incidence curves, taking into account competing risks, and Gray's test was used to compare the groups. For relapse, NRM was the competing risk, whereas for NRM, relapse was the competing risk. The

4 88 KONUMA ET AL. Cox proportional hazards regression model was used to estimate hazard ratios for overall mortality (inverse of OS), relapse, and NRM in univariate and multivariate analysis. The following factors were used for the multivariate analysis: pretransplant treatment (upfront HCT vs pretransplant chemotherapy), age (16 49 vs vs years), sex (male vs female), Karyotype (good vs intermediate vs poor), IPSS at TABLE 1 Characteristics of patients and transplantations before and after propensity score matching Overall cohort Propensity score matched cohort All Upfront Chemotherapy P value Upfront Chemotherapy P value Number of patients Age at HCT < years 176 (29.1) 110 (36.4) 66 (21.8) 60 (26.1) 60 (26.1) years 203 (33.6) 103 (34.1) 100 (33.0) 88 (38.3) 85 (37.0) years 226 (37.4) 89 (29.5) 137 (45.2) 82 (35.7) 85 (37.0) Sex Male 431 (71.2) 213 (70.5) 218 (71.9) 164 (71.3) 161 (70.0) Female 174 (28.8) 89 (29.5) 85 (28.1) 66 (28.7) 69 (30.0) Karyotype Good 250 (41.3) 119 (39.4) 131 (43.2) 88 (38.3) 83 (36.1) Intermediate 119 (19.7) 60 (19.9) 59 (19.5) 46 (20.0) 47 (20.4) Poor 236 (39.0) 123 (40.7) 113 (37.3) 96 (41.7) 100 (43.5) Monosomal karyotype 104 (17.1) 48 (15.9) 56 (18.5) (16.1) 50 (21.7) IPSS < Low, intermediate (21.5) 84 (27.8) 46 (15.2) 39 (17.0) 39 (17.0) Intermediate 2, high 475 (78.5) 218 (72.2) 257 (84.8) 191 (83.0) 191 (83.0) Donor source RBMT/PBSCT 163 (26.9) 92 (30.5) 71 (23.4) 63 (27.4) 54 (23.5) UBMT/PBSCT 226 (37.4) 99 (32.8) 127 (41.9) 84 (36.5) 88 (38.3) UCBT 216 (35.7) 111 (36.8) 105 (34.7) 83 (36.1) 88 (38.3) Conditioning regimen MAC 381 (63.0) 199 (65.9) 182 (60.1) 145 (63.0) 146 (63.5) RIC 224 (37.0) 103 (34.1) 121 (39.9) 85 (37.0) 84 (36.5) Year of HCT (7.1) 26 (8.6) 17 (5.6) 19 (8.3) 17 (7.4) (41.0) 142 (47.0) 106 (35.0) 91 (39.6) 94 (40.9) (51.9) 134 (44.4) 180 (59.4) 120 (52.2) 119 (51.7) Median months from diagnosis to HCT 5 (<1 12) 4 (<1 12) 5 (<1 12) < (<1 12) 5 (1 12) Abbreviations: HCT, hematopoietic cell transplantation; IPSS, international prognostic scoring system; MAC, myeloablative conditioning; RBMT/PBSCT, related bone marrow transplantation/peripheral blood stem cell transplantation; RIC, reduced intensity conditioning; UBMT/PBSCT, unrelated bone marrow transplantation/peripheral blood stem cell transplantation; UCBT, unrelated cord blood transplantation. FIGURE 2 Histogram of propensity score distribution between the upfront hematopoietic cell transplantation (HCT) group and the induction chemotherapy group

5 KONUMA ET AL. 89 diagnosis (low, intermediate 1 vs intermediate 2, high), donor source (bone marrow transplantation/peripheral blood stem cell transplantation from a related donor [RBMT/PBSCT] vs BMT/PBSCT from an unrelated donor [UBMT/PBSCT] vs cord blood transplantation from an unrelated donor [UCBT]), conditioning regimen (MAC vs RIC), and year of HCT ( vs vs ). All P values were two sided, and the data were analyzed by EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan), 18 a graphical user interface for the R software program (R Foundation for Statistical Computing, Vienna, Austria). 3 RESULTS 3.1 Propensity score matching A total of 605 patients with de novo advanced MDS, who received the first allogeneic HCT between 2001 and 2016, met the criteria for inclusion. Of these, 302 patients received upfront HCT, whereas 303 were treated with AML type remission induction chemotherapy before HCT. Among the overall cohort, the characteristics of sex, FIGURE 3 The probability of overall survival and the cumulative incidences of relapse and non relapse mortality (NRM) in the overall cohort and in the propensity score matched cohort

6 90 KONUMA ET AL. karyotype, and conditioning regimen were comparable between the two groups, but the proportions of age at HCT, IPSS at diagnosis, donor source, and year of HCT were different. The median time from diagnosis to HCT was 4 months for the upfront HCT group and 5 months for the induction chemotherapy group (P<0.001). In the propensity score matched cohort, there were no significant differences in variables between the two groups, except that the median time from diagnosis to HCT in the upfront HCT group was also shorter than that in the pre HCT chemotherapy group (P=0.001) (Table 1). The distribution of the propensity score of the upfront HCT group and the induction chemotherapy group before and after matching were shown in Figure 2. The C statistic of the propensity score model was 0.67, indicating good discrimination. 3.2 OS, relapse, and NRM In the overall cohort, with a median follow up of 41 months (range, 1 month to 188 months) for survivors, the probability of OS and the cumulative incidences of relapse and NRM were comparable between the two groups (Figure 3A C). In multivariate analysis, the pretransplant chemotherapy did not affect the overall mortality, relapse, and NRM (Table 2). In the propensity score matched cohort, with a median follow up of 40 months (range, 1 month to 188 months) for survivors, the probability of OS at 3 years was 41% (95% confidence interval [CI]: 34 48%) for the induction chemotherapy group and 49% (95% CI: 42 56%) for the upfront HCT group (P=0.224 by log rank test, Figure 3 D). The cumulative incidence of relapse at 3 years was 37% (95% CI: 30 43%) for the induction chemotherapy group and 27% (95% CI: 22 33%) for the upfront HCT group (P=0.017 by Gray's test, Figure 3E). The cumulative incidence of NRM at 3 years was 25% (95% CI: 20 31%) for the induction chemotherapy group and 28% (95% CI: 22 34%) for the upfront HCT group (P=0.175 by Gray's test, Figure 3F). There was also a significant difference in the hazard ratio of relapse between the two groups in univariate analysis (Figure 5B). In multivariate analysis, the pretransplant chemotherapy had a TABLE 2 Multivariate analysis of overall mortality, relapse, and NRM in the entire cohort Overall mortality Relapse NRM HR (95%CI) P value HR (95%CI) P value HR (95%CI) P value Pre transplant treatment Upfront HCT Pre HCT chemotherapy 1.00 ( ) ( ) ( ) Age at HCT years years 1.85 ( ) < ( ) ( ) < years 2.49 ( ) < ( ) ( ) <0.001 Sex Male Female 0.70 ( ) ( ) ( ) Karyotype Good Intermediate 0.79 ( ) ( ) ( ) Poor 1.87 ( ) < ( ) < ( ) IPSS Low, intermediate Intermediate 2, high 1.25 ( ) ( ) ( ) Donor source RBMT/PBSCT UBMT/PBSCT 0.90 ( ) ( ) ( ) UCBT 1.34 ( ) ( ) ( ) Conditioning regimen MAC RIC 0.66 ( ) ( ) ( ) Year of HCT ( ) ( ) ( ) ( ) ( ) ( ) CI, confidence interval; HCT, hematopoietic cell transplantation; HR, hazard ratio; IPSS, international prognostic scoring system; MAC, myeloablative conditioning; NRM, non relapse mortality; RBMT/PBSCT, related bone marrow transplantation/peripheral blood stem cell transplantation; RIC, reduced intensity conditioning; UBMT/PBSCT, unrelated bone marrow transplantation/peripheral blood stem cell transplantation; UCBT, unrelated cord blood transplantation. The P values in bold are statistically significant (<0.05).

7 KONUMA ET AL. 91 significantly higher relapse incidence, but did not affect overall mortality and NRM (Table 3). incidence of NRM was lower in the induction chemotherapy group compared with the upfront HCT group among patients who received RIC in univariate analysis (Figure 5C). 3.3 Subgroup analysis in the propensity score matched cohort In the propensity score matched cohort, we evaluated the overall mortality, relapse, and NRM when stratifying by age, karyotype, IPSS at diagnosis, conditioning regimen, and donor source to identify in which patient group an either better or worse outcome in patients who received induction chemotherapy could be observed. The probability of OS did not significantly differ between the two groups among patients within each subgroup (Figure 5A), but the cumulative incidence of relapse was significantly lower in the upfront HCT group compared with the induction chemotherapy group among patients with a poor karyotype, a higher IPSS at diagnosis, and those who received RIC in univariate analysis (Figure 4, Figure 5B). In contrast, the cumulative 4 DISCUSSION The purpose of the propensity score matched analysis was to clarify the impact of induction chemotherapy before HCT on post transplant relapse in adult patients with advanced MDS. Induction chemotherapy did not have any survival benefit after HCT. Unexpectedly, induction chemotherapy was significantly associated with a higher incidence of relapse after HCT, which might partly due to an indelible bias and/or adverse clonal evolution after induction chemotherapy. 19 In the subgroup analysis, this effect was observed among patients with a poor karyotype, a higher IPSS at diagnosis, and those who received RIC. Among patients who received RIC, induction chemotherapy significantly improved NRM, but it did not eventually improve survival after TABLE 3 Multivariate analysis of overall mortality, relapse, and NRM in the propensity score matched cohort. Overall mortality Relapse NRM HR (95%CI) P value HR (95%CI) P value HR (95%CI) P value Pre transplant treatment Upfront HCT Pre HCT chemotherapy 1.15 ( ) ( ) ( ) Age at HCT years years 1.97 ( ) < ( ) ( ) years 3.06 ( ) < ( ) ( ) <0.001 Sex Male Female 0.77 ( ) ( ) ( ) Karyotype Good Intermediate 0.77 ( ) ( ) ( ) Poor 1.77 ( ) < ( ) < ( ) IPSS Low, intermediate Intermediate 2, high 1.62 ( ) ( ) ( ) Donor source RBMT/PBSCT UBMT/PBSCT 1.11 ( ) ( ) ( ) UCBT 1.34 ( ) ( ) ( ) Conditioning regimen MAC RIC 0.67 ( ) ( ) ( ) Year of HCT ( ) ( ) ( ) ( ) ( ) ( ) CI, confidence interval; HCT, hematopoietic cell transplantation; HR, hazard ratio; IPSS, international prognostic scoring system; MAC, myeloablative conditioning; NRM, non relapse mortality; RBMT/PBSCT, related bone marrow transplantation/peripheral blood stem cell transplantation; RIC, reduced intensity conditioning; UBMT/PBSCT, unrelated bone marrow transplantation/peripheral blood stem cell transplantation; UCBT, unrelated cord blood transplantation. The P values in bold are statistically significant (<0.05).

8 92 KONUMA ET AL. FIGURE 4 The cumulative incidence of relapse according to the karyotype, International Prognostic Scoring System (IPSS) count at diagnosis, and conditioning regimen in the propensity score matched cohort HCT. These data suggested that induction chemotherapy before HCT did not have any benefit of post transplant relapse irrespective of the intensity of conditioning regimen. In advanced MDS, induction chemotherapy has been recommended for younger patients with a higher proportion of marrow blasts. Although about half of patients achieved remission, the duration of remission was short. 20 Therefore, for patients who are a candidate for allogeneic HCT, the role of induction chemotherapy for advanced MDS might be to reduce the tumor burden before HCT. However, in the setting of HCT following MAC, previous retrospective studies have demonstrated no significant differences in posttransplant relapse and survival when induction chemotherapy before HCT was compared with upfront HCT in patients with advanced MDS or AML following MDS, 3-6 which is consistent with one of our results. Moreover, induction chemotherapy may not be effective in patients with a poor karyotype. 20 Interestingly, Onida et al reported that intensive chemotherapy did not provide any benefit of posttransplant outcome in patients with a poor karyotype, even if the patients had been transplanted in remission. 2 Indeed, our results also showed that upfront HCT was significantly associated with a decreased incidence of post transplant relapse compared with induction chemotherapy in patients with a poor karyotype and in those with a higher IPSS at diagnosis, which mostly included poor karyotype. Therefore, upfront HCT should be considered for advanced MDS patients with a poor karyotype. The necessity of pretransplant chemotherapy might be dependent of the intensity of conditioning regimen. Recently, prospective randomized studies comparing MAC with RIC for MDS showed different results. 21,22 A European Group for Blood and Marrow Transplantation (EBMT) study by Kröger et al showed no significant difference in relapse incidence, NRM, and OS between RIC and MAC in MDS patients with <20% of blasts at the time of HCT. 21 In EBMT study, approximately half of patients received pretransplant induction chemotherapy. By contrast, the Blood and Marrow Transplant Clinical Trials Network (BMT CTN) study by Scott et al showed that RIC resulted in higher relapse incidence compared with MAC, resulting in the better relapse free survival with MAC in AML and MDS patients with <5% blasts at the time of HCT. 22 However, the impact of pretransplant chemotherapy on post transplant outcomes according to the intensity of conditioning regimen have yet to be clarified. Unexpectedly, our

9 KONUMA ET AL. 93 FIGURE 5 Forest plot for hazard ratios of overall mortality, relapse, and non relapse mortality (NRM) in the subgroup analyses. PSM, propensity score matched; HR, hazard ratio data showed that pretransplant induction chemotherapy was significantly associated with a higher incidence of relapse after HCT among patients received RIC, but not MAC. Therefore, our data suggested that MAC, but not RIC, could overcome an indelible bias and/or adverse clonal evolution after induction chemotherapy. With regard to the achievement of complete remission, Aza seems to be less effective than induction chemotherapy in patients with intermediate 2 and high risk MDS. However, Aza had several advantages over chemotherapy, including mild adverse profiles and extended time to AML progression. Indeed, Aza has recently been

10 94 KONUMA ET AL. used as a bridging therapy before HCT Nevertheless, the necessity of a bridging treatment to allow time for patients to reach HCT might be dependent on donor availability. The rapid availability is the most important benefit for UCBT. Therefore, UCBT was relatively common in our study because most patients needed urgent HCT. Thus, the rapid identification of an alternative donor, such as an unrelated cord blood, could favor the applicability of upfront HCT for advanced MDS. Our study had several limitations. First, our study was certainly subject to the selection bias of patients who actually received allogeneic HCT. In fact, induction chemotherapy related early mortality is not low for patients with advanced MDS, 20 and a considerable proportion of patients could not receive HCT in part due to adverse events and/or disease progression during induction chemotherapy. 28,29 Robin et al prospectively evaluated the impact of donor availability on survival in high risk MDS patients who candidates to HCT. In this study, 81 patients of the 112 (72%) patients with donors actually received HCT, whereas the remaining 31 patients with donors did not receive HCT because of progression disease despite pretransplant treatment, acquisition of a comorbidity contraindicating HCT, death during treatment, and other reasons. Given these results, upfront HCT should be considered for patients with advanced MDS. Second, patients received Aza before HCT were not included in our study. Several retrospective studies showed that similar post transplant outcomes in patients who received Aza before HCT compared with those received induction chemotherapy in patients with advanced MDS, although the proportion of patients, who receive either induction chemotherapy or treatment of Aza before HCT, actually undergoing HCT is unclear in the real world setting. In addition, Damaj et al also showed comparable survival and relapse between patients who received Aza before HCT and those who received upfront HCT in the setting of RIC using a propensity score matched analysis. 7 Therefore, prospective randomized trials will be required to clarify the role of cytoreductive treatment before allogeneic HCT in patients with advanced MDS. Third, we were unable to evaluate the proportion of blast in bone marrow or peripheral blood at diagnosis because the information is not included in the registry data. In fact, although the proportion of blasts in bone marrow at diagnosis is within a range of 5 19% in our study, the proportion of blasts might be one of the most important factor for the selection of induction chemotherapy before HCT for advanced MDS. Fourth, another major limitation of this study is the lack of information about the genetic mutation profiles. Recent studies showed that mutations in TP53 and Ras pathway were associated with relapse and surivival after HCT. 30,31 These mutations might also affect the relapse incidences after HCT. Thus, the proportion of blast and molecular profiles need to be included for analyzing the propensity score matching in the future study. In summary, this propensity score matched analysis demonstrated that AML type remission induction chemotherapy before HCT did not improve post transplant relapse and survival for adult patients with advanced MDS. Upfront HCT is preferable for patients with a poor karyotype. Because the biological and clinical continuum was observed in MDS and older AML, 32,33 upfront HCT might be a valuable option for treatment in older AML. Further studies are required to clarify the role of upfront HCT for MDS and older AML. ACKNOWLDGEMENTS We thank all of the physicians and staff at the centers who provided the clinical data to the Transplant Registry Unified Management Program (TRUMP) of the Japanese Data Center for Hematopoietic Cell Transplantation (JDCHCT). This work was supported in part by the Practical Research Project for Allergic Diseases and Immunology (Research Technology of Medical Transplantation) from Japan Agency for Medical Research and Development, AMED under grant number 18ek h0002. AUTHOR CONTRIBUTIONS TK designed the research, analyzed the data, performed the statistical analysis and wrote the first draft of the manuscript. YS, TM and YK contributed to the critical review of the manuscript. All the other authors contributed to data collection. All authors approved the final version. CONFLICTS OF INTEREST The authors declare no competing financial interests. ORCID Takaaki Konuma Yoshinobu Kanda REFERENCES Warlick ED, Cioc A, Defor T, Dolan M, Weisdorf D. Allogeneic stem cell transplantation for adults with myelodysplastic syndromes: importance of pretransplant disease burden. Biol Blood Marrow Transplant. 2009;15: Onida F, Brand R, van Biezen A, et al. Impact of the International Prognostic Scoring System cytogenetic risk groups on the outcome of patients with primary myelodysplastic syndromes undergoing allogeneic stem cell transplantation from human leukocyte antigen identical siblings: a retrospective analysis of the European Society for Blood and Marrow Transplantation Chronic Malignancies Working Party. Haematologica. 2014;99(10): Anderson JE, Gooley TA, Schoch G, et al. Stem cell transplantation for secondary acute myeloid leukemia: evaluation of transplantation as initial therapy or following induction chemotherapy. Blood. 1997;89(7): Scott BL, Storer B, Loken MR, Storb R, Appelbaum FR, Deeg HJ. Pretransplantation induction chemotherapy and posttransplantation relapse in patients with advanced myelodysplastic syndrome. Biol Blood Marrow Transplant. 2005;11: Nakai K, Kanda Y, Fukuhara S, et al. Value of chemotherapy before allogeneic hematopoietic stem cell transplantation from an HLA identical sibling donor for myelodysplastic syndrome. Leukemia. 2005;19(3): Alessandrino EP, Della Porta MG, Pascutto C, Bacigalupo A, Rambaldi A. Should cytoreductive treatment be performed before transplantation in patients with high risk myelodysplastic syndrome? J Clin Oncol. 2013;31: Damaj G, Mohty M, Robin M, et al. Upfront allogeneic stem cell transplantation after reduced intensity/nonmyeloablative conditioning for patients with myelodysplastic syndrome: a study by the Société Française de Greffe de Moelle et de Thérapie Cellulaire. Biol Blood Marrow Transplant. 2014;20(9): Saure C, Schroeder T, Zohren F, et al. Upfront allogeneic blood stem cell transplantation for patients with high risk myelodysplastic

11 KONUMA ET AL. 95 syndrome or secondary acute myeloid leukemia using a FLAMSAbased high dose sequential conditioning regimen. Biol Blood Marrow Transplant. 2012;18(3): Kako S, Kanda Y, Kato J, et al. The bridge treatment selected at the decision for transplantation did not affect the outcomes in patients with MDS. Hematol Oncol. 2017;35(3): Atsuta Y, Suzuki R, Yoshimi A, et al. Unification of hematopoietic stem cell transplantation registries in Japan and establishment of the TRUMP System. Int J Hematol. 2007;86(3): Atsuta Y. Introduction of Transplant Registry Unified Management Program 2 (TRUMP2): scripts for TRUMP data analyses, part I (variables other than HLA related data). Int J Hematol. 2016;103: Kanda J. Scripts for TRUMP data analyses. Part II (HLA related data): statistical analyses specific for hematopoietic stem cell transplantation. Int J Hematol. 2016;103: Bennett JM, Catovsky D, Daniel MT, et al. Proposals for the classification of the myelodysplastic syndromes. Br J Haematol. 1982;51(2): Vardiman JW, Thiele J, Arber DA, 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(5): Greenberg P, Cox C, LeBeau MM, et al. International scoring system for evaluating prognosis in myelodysplastic syndromes. Blood. 1997;89(6): 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(29): Giralt S, Ballen K, Rizzo D, et al. Reduced intensity conditioning regimen workshop: defining the dose spectrum. Report of a workshop convened by the center for international blood and marrow transplant research. Biol Blood Marrow Transplant. 2009;15(3): Kanda Y. Investigation of the freely available easy to use software 'EZR' for medical statistics. Bone Marrow Transplant. 2013;48: Ding L, Ley TJ, Larson DE, et al. Clonal evolution in relapsed acute myeloid leukaemia revealed by whole genome sequencing. Nature. 2012;481(7382): Beran M. Intensive chemotherapy for patients with high risk myelodysplastic syndrome. Int J Hematol. 2000;72: Kröger N, Iacobelli S, Franke GN, et al. Dose reduced versus standard conditioning followed by allogeneic stem cell transplantation for patients with myelodysplastic syndrome: a prospective randomized Phase III study of the EBMT (RICMAC Trial). J Clin Oncol. 2017;35(19): Scott BL, Pasquini MC, Logan BR, et al. Myeloablative versus reducedintensity hematopoietic cell transplantation for acute myeloid leukemia and myelodysplastic syndromes. J Clin Oncol. 2017;35(11): Gerds AT, Gooley TA, Estey EH, Appelbaum FR, Deeg HJ, Scott BL. Pretransplantation therapy with azacitidine vs induction chemotherapy and posttransplantation outcome in patients with MDS. Biol Blood Marrow Transplant. 2012;18: Damaj G, Duhamel A, Robin M, et al. Impact of azacitidine before allogeneic stem cell transplantation for myelodysplastic syndromes: a study by the Société Française de Greffe de Moelle et de Thérapie Cellulaire and the Groupe Francophone des Myélodysplasies. J Clin Oncol. 2012;30(36): Potter VT, Iacobelli S, van Biezen A, et al. Comparison of intensive chemotherapy and hypomethylating agents before allogeneic stem cell transplantation for advanced myelodysplastic syndromes: a study of the myelodysplastic syndrome subcommittee of the chronic malignancies working party of the european society for blood and marrow transplant research. Biol Blood Marrow Transplant. 2016;22: Festuccia M, Baker K, Gooley TA, Sandmaier BM, Deeg HJ, Scott BL. Hematopoietic cell transplantation in myelodysplastic syndromes after treatment with hypomethylating agents. Biol Blood Marrow Transplant. 2017;23: Voso MT, Leone G, Piciocchi A, et al. Feasibility of allogeneic stem cell transplantation after azacitidine bridge in higher risk myelodysplastic syndromes and low blast count acute myeloid leukemia: results of the BMT AZA prospective study. Ann Oncol. 2017;28(7): Estey E, de Lima M, Tibes R, et al. Prospective feasibility analysis of reduced intensity conditioning (RIC) regimens for hematopoietic stem cell transplantation (HSCT) in elderly patients with acute myeloid leukemia (AML) and high risk myelodysplastic syndrome (MDS). Blood. 2007;109(4): Robin M, Porcher R, Adès L, et al. HLA matched allogeneic stem cell transplantation improves outcome of higher risk myelodysplastic syndrome A prospective study on behalf of SFGM TC and GFM. Leukemia. 2015;29(7): Lindsley RC, Saber W, Mar BG, et al. Prognostic mutations in myelodysplastic syndrome after stem cell transplantation. N Engl J Med. 2017;376(6): Yoshizato T, Nannya Y, Atsuta Y, et al. Genetic abnormalities in myelodysplasia and secondary acute myeloid leukemia: impact on outcome of stem cell transplantation. Blood. 2017;129(17): Deschler B, de Witte T, Mertelsmann R, Lübbert M. Treatment decision making for older patients with high risk myelodysplastic syndrome or acute myeloid leukemia: problems and approaches. Haematologica. 2006;91: Lindsley RC, Mar BG, Mazzola E, et al. Acute myeloid leukemia ontogeny is defined by distinct somatic mutations. Blood. 2015;125(9): How to cite this article: Konuma T, Shimomura Y, Ozawa Y, et al. Induction chemotherapy followed by allogeneic HCT versus upfront allogeneic HCT for advanced myelodysplastic syndrome: A propensity score matched analysis. Hematological Oncology. 2019;37: