Red Blood Cell Transfusion Dependence and Outcome after Allogeneic Peripheral Blood Stem Cell Transplantation in Patients with de Novo Myelodysplastic Syndrome (MDS)

The prognosis of patients with de novo myelodysplastic syndrome (MDS) who are red blood cell transfusion–dependent (TD) and receive supportive care is inferior to that of patients who do not require transfusions. Whether TD also affects outcome after allogeneic transplantation is unknown. Consequently, in 172 de novo MDS patients (median age, 51 years), we analyzed the impact of TD on outcome after high-dose conditioning and allogeneic peripheral blood stem cell transplantation (PBSCT). With a median follow-up of 37 months, the probability of 3-year overall survival (OS) did not differ significantly between patients who were TD and those who were not TD before PBSCT (P = .1); however, transfusion burden, as reflected by ferritin levels, was correlated with a greater probability of severe acute graft-versus-host disease (aGVHD; P = .03) and a higher comorbidity index (P = .01), and OS was inferior in those patients with a ferritin level >1000 μg/L before PBSCT (P = .03). In multivariate analysis, only marrow myeloblast count (P = .01) and comorbidity index (P = .001) had a significant impact on OS. Our data do not identify TD as an independent negative prognostic factor for outcome after allogeneic PBSCT' however, iron overload (presumably transfusion-related) may contribute to poor transplantation success by adding to the overall comorbidities. Whether clinical intervention in the form of iron chelation can improve the outcome of allogeneic PBSCT in TD patients with MDS remains to be determined. The prognosis of patients with de novo myelodysplastic syndrome (MDS) who are red blood cell transfusion–dependent (TD) and receive supportive care is inferior to that of patients who do not require transfusions. Whether TD also affects outcome after allogeneic transplantation is unknown. Consequently, in 172 de novo MDS patients (median age, 51 years), we analyzed the impact of TD on outcome after high-dose conditioning and allogeneic peripheral blood stem cell transplantation (PBSCT). With a median follow-up of 37 months, the probability of 3-year overall survival (OS) did not differ significantly between patients who were TD and those who were not TD before PBSCT (P = .1); however, transfusion burden, as reflected by ferritin levels, was correlated with a greater probability of severe acute graft-versus-host disease (aGVHD; P = .03) and a higher comorbidity index (P = .01), and OS was inferior in those patients with a ferritin level >1000 μg/L before PBSCT (P = .03). In multivariate analysis, only marrow myeloblast count (P = .01) and comorbidity index (P = .001) had a significant impact on OS. Our data do not identify TD as an independent negative prognostic factor for outcome after allogeneic PBSCT' however, iron overload (presumably transfusion-related) may contribute to poor transplantation success by adding to the overall comorbidities. Whether clinical intervention in the form of iron chelation can improve the outcome of allogeneic PBSCT in TD patients with MDS remains to be determined. IntroductionSeveral therapeutic modalities for patients with myelodysplastic syndrome (MDS) have been developed over the past few years, some of which may change the natural course of the disease 1Fenaux P. Mufti G.J. Santini V. et al.Azacitidine (AZA) treatment prolongs overall survival (OS) in higher-risk MDS patients compared with conventional care regimens (CCR): results of the AZA-001 phase III study.Blood. 2007; 110: 250A-251AGoogle Scholar. With the rare exception of patients who achieve long-lasting remission with chemotherapy, allogeneic hematopoietic stem cell transplantation (HSCT) is currently the only modality with proved curative potential 2Deeg H.J. Appelbaum F.R. Hemopoietic stem cell transplantation for myelodysplastic syndrome.Curr Opin Oncol. 2000; 12: 116-120Crossref PubMed Scopus (12) Google Scholar. But given the heterogeneity of MDS and the potential complications associated with HSCT, deciding when and in whom to perform allogeneic HSCT often is difficult. The decision model presented by Cutler et al. 3Cutler C.S. Lee S.J. Greenberg P. et al.A decision analysis of allogeneic bone marrow transplantation for the myelodysplastic syndromes: delayed transplantation for low-risk myelodysplasia is associated with improved outcome.Blood. 2004; 104: 579-585Crossref PubMed Scopus (541) Google Scholar showed a survival advantage with early transplantation in patients with MDS classified as intermediate-2 or high risk based on International Prognostic Scoring System (IPSS) criteria 4Greenberg P. Cox C. LeBeau M.M. et al.International scoring system for evaluating prognosis in myelodysplastic syndromes.Blood. 1997; 89: 2079-2088Crossref PubMed Google Scholar; however, that analysis was based largely on observations antedating the availability of new drugs and did not incorporate more recent information on factors that might affect the disease course, such as comorbidity, transfusion dependence (TD), or iron overload. The identification of additional clinical markers that might help in selecting patients likely to benefit from allogeneic HSCT and in determining the optimum time point for HSCT is desirable.The need for red blood cell transfusion support regardless of IPSS score has been shown to have a negative impact on overall survival (OS) in patients who did not undergo transplantation 5Malcovati L. Porta M.G. Pascutto C. et al.Prognostic factors and life expectancy in myelodysplastic syndromes classified according to WHO criteria: a basis for clinical decision making.J Clin Oncol. 2005; 23: 7594-7603Crossref PubMed Scopus (725) Google Scholar. The reason for this is not entirely clear. Patients who become TD may have a more aggressive disease biology. They also may experience adverse effects related to transfusion-induced allosensitization or may suffer toxic organ effects related to iron overload. The finding of a negative impact of TD on prognosis has led to the development of a new proposed scoring system for patients with de novo MDS: the World Health Organization International Prognostic Scoring System (WPSS), which merges the parameters of the World Health Organization's (WHO) classification of MDS with the IPSS, essentially by having TD replace the “cytopenia” category of the IPSS 6Malcovati L. Germing U. Kuendgen A. et al.Time-dependent prognostic scoring system for predicting survival and leukemic evolution in myelodysplastic syndromes.J Clin Oncol. 2007; 25: 3503-3510Crossref PubMed Scopus (880) Google Scholar.Whether TD also is an independent risk factor in patients undergoing allogeneic HSCT remains unknown. A potential negative effect of TD on transplantation outcome has been derived indirectly from a study by Armand et al. 7Armand P. Kim H.T. Cutler C.S. et al.Prognostic impact of elevated pretransplantation serum ferritin in patients undergoing myeloablative stem cell transplantation.Blood. 2007; 109: 4586-4588Crossref PubMed Scopus (370) Google Scholar in 103 patients with MDS (including therapy-related MDS, for which the scoring systems have not been validated) that suggested a possible link between pretransplantation ferritin level and outcome after allogeneic HSCT using conventional conditioning regimens. In that study, the negative impact of iron overload appeared to be independent of the effects of graft-versus-host disease (GVHD). Other possibilities are that transfusion needs and ferritin levels were related to disease stage or disease duration and, associated with this, a long history of transfusion support.We performed a retrospective study involving 6 transplantation centers in Germany, Austria, and the United States to analyze results in 172 patients with de novo MDS undergoing allogeneic HSCT using granulocyte colony-stimulating factor (G-CSF)–mobilized peripheral blood stem cells (PBSCs) after high-intensity conditioning.Materials and MethodsPatient and disease characteristicsThe study cohort comprised 172 patients with de novo MDS (Table 1). The patients were classified according to French-American-British (FAB), WHO, IPSS, and WPSS criteria at the time of transplantation. According to Malcovati et al. 6Malcovati L. Germing U. Kuendgen A. et al.Time-dependent prognostic scoring system for predicting survival and leukemic evolution in myelodysplastic syndromes.J Clin Oncol. 2007; 25: 3503-3510Crossref PubMed Scopus (880) Google Scholar, TD was defined as having received at least one unit of red blood cells within the 8-week period before PBSCT. The precise number of units of red blood cells transfused was not consistently available. Only 4 patients underwent transplantation within 2 months of diagnosis. The most recent ferritin level, obtained not more than 6 weeks before PBSCT, was considered for analysis. Pretransplantation comorbidities were assessed using the hematopoietic cell transplantation comorbidity index (HCT-CI) 8Sorror M.L. Maris M.B. Storb R. et al.Hematopoietic cell transplantation (HCT)-specific comorbidity index: a new tool for risk assessment before allogeneic HCT.Blood. 2005; 106: 2912-2919Crossref PubMed Scopus (1926) Google Scholar. Comorbidity scores could be assigned to 164 patients; the breakdown by score was as follows: 0, 74 patients; 1, 31 patients; 2, 19 patients; 3, 20 patients; 4, 13 patients; and ≥ 5, 7 patients.Table 1Disease and Patient CharacteristicsAll Patients (n = 172), nNon-TD (n = 41), n (%)TD (n = 131), n (%)PFAB classificationNS RA/RARS8518 (44)67 (51) RAEB5218 (44)34 (26) RAEB-t162 (5)14 (11) MDS/AML193 (7)16 (12)WHO classification.004 5q syndrome50 (0)5 (4) RA/RARS140 (0)14 (11) RCMD/RCMD-RS6419 (46)45 (34) RAEB-1247 (17)17 (13) RAEB-22710 (24)17 (13) AML355 (12)30 (23) MDS-U30 (0)3 (2)IPSS classification.01 LOW92 (5)7 (6) INT-17718 (47)58 (53) INT-24216 (43)26 (24) HIGH212 (5)19 (17)WPSS classificationNS Very low00 (0)0 (0) Low218 (22)13 (12) Intermediate3810 (27)27 (25) High6617 (47)48 (45) Very high101 (4)9 (8)Median age at treatment, years (range)51 (10-68)53 (21-64)50 (10-68)NSMedian time from diagnosis to treatment, months (range)8.5 (0.1-243)8.7 (0.4-35)8.5 (0.1-243)NSRelated/unrelated donor, n90/8218/2372/59NSHLA mismatch, n101 (2)9 (7)NSMedian CD34 cells/kg (range)6.76 (1.3-32.2)7.49 (2.07-18.7)6.67 (1.3-32.2)NSCMV seropositivity, n541539NSMedian follow-up, months (range)37 (0.5-135)49.1 (0.5-97)37.1 (0.8-135)NSCMV indicates cytomegalovirus.Note. The patients were grouped according to red blood cell transfusion dependence and the current classification and scoring systems used in MDS, including FAB, WHO, IPSS, and WPSS. Open table in a new tab DNA-based HLA-typing of donor and recipient was performed for and and by for patients received patients had received before PBSCT, of whom remission regimens were over 4 or In was with over 2 n = of n = or over n = patients received as of the conditioning as H.J. et of acute and graft-versus-host disease with the of to a 12: PubMed Scopus Google Scholar]. were on with a n = or = with n = (n = or n = was using the and were and using criteria P. et on acute Google et graft-versus-host syndrome in a study of 20 PubMed Scopus Google Scholar]. after was defined as a to regardless of the was classified as a was as the of when a or was the of in patients who had not was considered a to was at the time of from in the of patients and were according to at the retrospective study was performed within the MDS with the in in with the of of survival and survival were obtained by the patients were at follow-up P. from Scopus Google Scholar]. The of and were using for in 1997; PubMed Scopus Google Scholar]; the was used to in of Cox were to the effects of on were analyzed as of 1, outcome by FAB, WHO, and IPSS a median follow-up of 37 months, the 3-year OS was with an for patients patients were grouped by outcome as most in with patients OS was in patients with more including with P = and in P = and for patients MDS had to acute P = not significant The outcome in the was related to the that most of patients received before undergoing by WHO for a of with compared with (Table however, some only of patients and are not in patients with with and the few patients with MDS the 3-year OS was and The 3-year OS was for patients with with with for those with for those with and for those with The between patients with and (P = .01) or (P = were with which showed that the IPSS risk outcome in patients with MDS H.J. et with and for stem cell transplantation from related and in patients with myelodysplastic PubMed Scopus Google Scholar], was a significant between patients in risk category and those in risk category (P = and P = impact of TD on survival after shown in 1, of the 172 patients were TD before PBSCT. the were by did differ by WHO classification (P = was related to the that the TD patients based on had a such as patients with or who not have transplantation were not In were significantly more patients (P = .01) in the cohort that was not Patient and transplantation including 8Sorror M.L. Maris M.B. Storb R. et al.Hematopoietic cell transplantation (HCT)-specific comorbidity index: a new tool for risk assessment before allogeneic HCT.Blood. 2005; 106: 2912-2919Crossref PubMed Scopus (1926) Google Scholar not were in (Table of and patients were for and the developed was between the (P = data not The was for (P = data not which in of TD significantly more patients with risk WHO and IPSS and to have OS compared with the In as shown in the 3-year OS was in the compared with in the TD (P = were significant between and not P = and was a for survival in of patients in the P = of patients according to transfusion TD (n = TD (n = (P = WPSS TD the classification and grouped patients according to the developed WPSS were patients in WPSS risk the of patients for allogeneic HSCT in clinical shown in the 3-year OS was for and for and for patients by WPSS criteria (P = TD is an of the WPSS the impact of TD was not analyzed 3-year OS for IPSS and patients who were TD or was (P = (P = and (P = The risk few patients to a analysis. of risk IPSS in of TD significant a was in the risk in of patients n = compared with TD patients n = (P = by risk and risk (P = risk (P = .1); poor risk (P = level and transplantation pretransplantation ferritin levels have been shown to be associated with a significantly inferior OS to 7Armand P. Kim H.T. Cutler C.S. et al.Prognostic impact of elevated pretransplantation serum ferritin in patients undergoing myeloablative stem cell transplantation.Blood. 2007; 109: 4586-4588Crossref PubMed Scopus (370) Google Scholar, we as a of was correlated with outcome in MDS patients had received before The median pretransplantation ferritin for patients was μg/L to was a significant in ferritin levels between TD and patients (median ferritin levels P In clinical μg/L often as a to chelation patients were grouped by ferritin levels or Patients with a ferritin μg/L (n = had a significantly 3-year OS than patients with levels P = .03). was a higher of and in the with ferritin the did not (P = The analysis was to patients with ferritin μg/L 7Armand P. Kim H.T. Cutler C.S. et al.Prognostic impact of elevated pretransplantation serum ferritin in patients undergoing myeloablative stem cell transplantation.Blood. 2007; 109: 4586-4588Crossref PubMed Scopus (370) Google Scholar. shown in 3-year survival was significantly inferior patients with ferritin levels this P = .03). The was for the 2 not a a link between higher ferritin levels and which was by the of a of disease stage with ferritin levels not In to be higher in patients with ferritin (P = compared the of in patients with to ferritin The of in patients with ferritin μg/L not >1000 P = was significantly higher than in patients with ferritin levels P = were in the patients with not and of were on pretransplantation ferritin level P = P = iron overload can to in et al. 8Sorror M.L. Maris M.B. Storb R. et al.Hematopoietic cell transplantation (HCT)-specific comorbidity index: a new tool for risk assessment before allogeneic HCT.Blood. 2005; 106: 2912-2919Crossref PubMed Scopus (1926) Google Scholar have shown that pretransplantation including of by the have a impact on transplantation Consequently, we analyzed ferritin levels were correlated with the given in 5, that iron overload was significantly associated with a higher (P = ferritin level and comorbidity index (HCT-CI) (P = A significant (P = .01) was between pretransplantation ferritin level and the shown in 2, only the (P = .001) and the classification (P = .01) significant for outcome in multivariate analysis. Patient age years), IPSS number of cells ferritin level mismatch, or donor had significant of or Open table in a new tab of this retrospective analysis was to as a negative prognostic factor in patients with de novo MDS with supportive also affects prognosis in patients who have allogeneic PBSCT. The data significant overall for TD the of a retrospective analysis, the data that the negative prognostic impact of TD in patients who did not undergo transplantation can be by PBSCT. be however, that in the on TD by Malcovati et al. 5Malcovati L. Porta M.G. Pascutto C. et al.Prognostic factors and life expectancy in myelodysplastic syndromes classified according to WHO criteria: a basis for clinical decision making.J Clin Oncol. 2005; 23: 7594-7603Crossref PubMed Scopus (725) Google Scholar, the median age was years, compared with 51 years in the patients are more likely to have organ and may be more to organ and related to the additional iron as a of transfusions. the of MDS may be aggressive in patients than in those over age years A. C. et syndromes in patients than age Clin Oncol. PubMed Scopus Google TD cohort significantly more patients with MDS as by conventional In the of TD patients was in the cohort with more MDS, in those patients disease was more and parameters than TD the of is by the impact of risk on PBSCT In patients with the TD did not affect the negative impact of the appeared to be the in with recent et al.Hematopoietic cell transplantation in patients with myelodysplastic syndrome or acute from myelodysplastic in patients with de novo disease and disease or 2007; 110: PubMed Scopus Google P. Kim H.T. et of and on outcome of and MDS after allogeneic transplantation.Blood. Scholar]. The data that patients with a or those not was for the category a of that may have prognostic et al.Hematopoietic cell transplantation in patients with myelodysplastic syndrome or acute from myelodysplastic in patients with de novo disease and disease or 2007; 110: PubMed Scopus Google Scholar], and TD may as an additional an to TD with iron we analyzed the data also on ferritin In with the of Armand et al. 7Armand P. Kim H.T. Cutler C.S. et al.Prognostic impact of elevated pretransplantation serum ferritin in patients undergoing myeloablative stem cell transplantation.Blood. 2007; 109: 4586-4588Crossref PubMed Scopus (370) Google Scholar and A. P. et severe iron overload after stem cell transplantation and possible with 2004; PubMed Scopus Google Scholar], the current data suggested that elevated ferritin levels had a negative impact on related to higher of and more severe Our results that the negative effect of as reflected by elevated ferritin levels, was at ferritin levels of is in chelation is not levels of or higher are and in the of iron 2007; PubMed Scopus Google Scholar]. the study TD as as ferritin level, the data that TD can be a negative risk the retrospective analysis did not a significant data in patients with modalities improved life expectancy with chelation in TD patients C. et impact of iron chelation on survival in transfused MDS a analysis by the 2007; 110: Scholar]. data are for patients undergoing have shown an between iron overload and an of A. P. et severe iron overload after stem cell transplantation and possible with 2004; PubMed Scopus Google et of the in marrow Google Scholar], in et and outcome of in hematopoietic stem cell PubMed Scopus Google Scholar]. TD were a risk factor for transplantation might be considered a and impact could be analyzed in the of the 8Sorror M.L. Maris M.B. Storb R. et al.Hematopoietic cell transplantation (HCT)-specific comorbidity index: a new tool for risk assessment before allogeneic HCT.Blood. 2005; 106: 2912-2919Crossref PubMed Scopus (1926) Google Scholar. of a of iron overload with the may be that iron chelation can in of which may affect after treatment, such as PBSCT. the of transplantation conditioning affects be of to the analysis to patients undergoing allogeneic HSCT after conditioning U. et conditioning by allogeneic hematopoietic cell transplantation in PubMed Scopus Google R. et PBSCT with conditioning in patients with and MDS: results of a et outcome of and myeloablative conditioning in allogeneic stem cell transplantation for patients than years of age with acute a retrospective from the of the for and 2005; PubMed Scopus Google et allogeneic transplantation for patients with myelodysplastic syndrome or acute with a retrospective PubMed Scopus Google A. et hematopoietic stem cell transplantation in and MDS using myeloablative the of PubMed Scopus Google C. et stem cell transplantation using a conditioning has the to and survival in patients with acute and Clin Oncol. 2005; 23: PubMed Scopus Google et and conditioning for patients with undergoing allogeneic cell transplantation.Blood. PubMed Scopus Google A. et allogeneic hematopoietic stem cell transplantation for myelodysplastic syndrome and acute with using and 2004; 104: PubMed Scopus Google et stem cell transplantation after a conditioning in patients with myelodysplastic syndrome or acute PubMed Scopus Google this retrospective study some pretransplantation iron and transplantation TD was not shown to be an independent risk this to iron which to in also to of and, or higher levels be the currently used Patients who are considered for PBSCT may benefit from iron chelation at iron overload. allogeneic PBSCT be considered in the disease course in patients with MDS who are IntroductionSeveral therapeutic modalities for patients with myelodysplastic syndrome (MDS) have been developed over the past few years, some of which may change the natural course of the disease 1Fenaux P. Mufti G.J. Santini V. et al.Azacitidine (AZA) treatment prolongs overall survival (OS) in higher-risk MDS patients compared with conventional care regimens (CCR): results of the AZA-001 phase III study.Blood. 2007; 110: 250A-251AGoogle Scholar. With the rare exception of patients who achieve long-lasting remission with chemotherapy, allogeneic hematopoietic stem cell transplantation (HSCT) is currently the only modality with proved curative potential 2Deeg H.J. Appelbaum F.R. Hemopoietic stem cell transplantation for myelodysplastic syndrome.Curr Opin Oncol. 2000; 12: 116-120Crossref PubMed Scopus (12) Google Scholar. But given the heterogeneity of MDS and the potential complications associated with HSCT, deciding when and in whom to perform allogeneic HSCT often is difficult. The decision model presented by Cutler et al. 3Cutler C.S. Lee S.J. Greenberg P. et al.A decision analysis of allogeneic bone marrow transplantation for the myelodysplastic syndromes: delayed transplantation for low-risk myelodysplasia is associated with improved outcome.Blood. 2004; 104: 579-585Crossref PubMed Scopus (541) Google Scholar showed a survival advantage with early transplantation in patients with MDS classified as intermediate-2 or high risk based on International Prognostic Scoring System (IPSS) criteria 4Greenberg P. Cox C. LeBeau M.M. et al.International scoring system for evaluating prognosis in myelodysplastic syndromes.Blood. 1997; 89: 2079-2088Crossref PubMed Google Scholar; however, that analysis was based largely on observations antedating the availability of new drugs and did not incorporate more recent information on factors that might affect the disease course, such as comorbidity, transfusion dependence (TD), or iron overload. The identification of additional clinical markers that might help in selecting patients likely to benefit from allogeneic HSCT and in determining the optimum time point for HSCT is desirable.The need for red blood cell transfusion support regardless of IPSS score has been shown to have a negative impact on overall survival (OS) in patients who did not undergo transplantation 5Malcovati L. Porta M.G. Pascutto C. et al.Prognostic factors and life expectancy in myelodysplastic syndromes classified according to WHO criteria: a basis for clinical decision making.J Clin Oncol. 2005; 23: 7594-7603Crossref PubMed Scopus (725) Google Scholar. The reason for this is not entirely clear. Patients who become TD may have a more aggressive disease biology. They also may experience adverse effects related to transfusion-induced allosensitization or may suffer toxic organ effects related to iron overload. The finding of a negative impact of TD on prognosis has led to the development of a new proposed scoring system for patients with de novo MDS: the World Health Organization International Prognostic Scoring System (WPSS), which merges the parameters of the World Health Organization's (WHO) classification of MDS with the IPSS, essentially by having TD replace the “cytopenia” category of the IPSS 6Malcovati L. Germing U. Kuendgen A. et al.Time-dependent prognostic scoring system for predicting survival and leukemic evolution in myelodysplastic syndromes.J Clin Oncol. 2007; 25: 3503-3510Crossref PubMed Scopus (880) Google Scholar.Whether TD also is an independent risk factor in patients undergoing allogeneic HSCT remains unknown. A potential negative effect of TD on transplantation outcome has been derived indirectly from a study by Armand et al. 7Armand P. Kim H.T. Cutler C.S. et al.Prognostic impact of elevated pretransplantation serum ferritin in patients undergoing myeloablative stem cell transplantation.Blood. 2007; 109: 4586-4588Crossref PubMed Scopus (370) Google Scholar in 103 patients with MDS (including therapy-related MDS, for which the scoring systems have not been validated) that suggested a possible link between pretransplantation ferritin level and outcome after allogeneic HSCT using conventional conditioning regimens. In that study, the negative impact of iron overload appeared to be independent of the effects of graft-versus-host disease (GVHD). Other possibilities are that transfusion needs and ferritin levels were related to disease stage or disease duration and, associated with this, a long history of transfusion support.We performed a retrospective study involving 6 transplantation centers in Germany, Austria, and the United States to analyze results in 172 patients with de novo MDS undergoing allogeneic HSCT using granulocyte colony-stimulating factor (G-CSF)–mobilized peripheral blood stem cells (PBSCs) after high-intensity