- Email: [email protected]
Are matched unrelated donor transplants justified for AML in CR1?
Best Practice & Research Clinical Haematology Vol. 19, No. 2, pp. 321–328, 2006 doi:10.1016/j.beha.2005.12.002 available online at http://www.sciencedirect.com
7 Are matched unrelated donor transplants justified for AML in CR1? Claudio Anasetti* Janelle Perkins
Pherm D
Michael L. Nieder Teresa Field
MD
MD
MD PhD
Blood and Marrow Transplantation Program, University of South Florida; H. Lee Moffitt Cancer Center and Research Institute, 12901 Magnolia Drive, Tampa, FL 33612 USA; and All Children’s Hospital 801 Sixth Street South, St. Petersburg, FL 33701, USA.
There has been controversy about the optimal consolidation therapy for patients with acute myeloid leukemia (AML) in first remission. Hematopoietic stem cell transplantation from human leukocyte antigen (HLA)-identical siblings has improved the survival of patients with unfavorable cytogenetics, but has not improved the survival of intermediate- or favorable-risk patients. If an HLA-identical sibling donor is not available, alternative sources of stem cells may be sought. HLA mismatched transplants are associated with an increased risk of graft rejection and graft-versushost disease, and lower survival. Since large registries of HLA-typed volunteer donors have been established and the newer and more sensitive tissue typing technology can more clearly differentiate between matched and unmatched donors, some AML patients without an HLAmatched sibling have received transplants from an HLA-matched unrelated donor while in first remission. Data from the Center for International Blood and Marrow Transplant Research indicate that survival of patients with unfavorable cytogenetics appears at least as good with unrelated donor grafts as previously reported for matched sibling grafts, and better than consolidation chemotherapy. AML patients in first remission with unfavorable cytogenetics without a matched family donor should be offered an unrelated donor transplant. Key words: cell transplantation; cytogenetics; myeloid.
Hematopoietic stem cell transplantation (HCT) from an HLA-identical sibling donor improves survival of acute myeloid leukemia (AML) patients in first complete remission * Corresponding author. Tel.: C1 813 745 2557; Fax: C1 813 745 3071. E-mail address: [email protected] (C. Anasetti).
1521-6926/$ - see front matter Q 2006 Elsevier Ltd. All rights reserved.
322 C. Anasetti et al
(AML-CR1) with high-risk cytogenetics; the 5-year survival of these patients is 20–30% using HCT compared to 5–15% using chemotherapy alone as consolidation. The data for intermediate-risk patients are controversial, and data for low-risk AML-CR1 indicate worse survival with an allograft, as higher treatment-related mortality exceeds the benefits of the allograft. A meta-analysis of five randomized clinical trials comparing HLA-identical sibling grafts to autografts or chemotherapy for AML-CR1 supports these conclusions.1 Much of the benefit of the allografts derives from the graft-versus-leukemia effect.2 Unfortunately, only about 30% of the patients in the US and Europe have an HLAidentical sibling, but most find an HLA-compatible unrelated volunteer among the 10 million donors registered worldwide. Transplants from unrelated donors, however, have been associated with greater risk of rejection, graft-versus-host disease (GVHD), and mortality than those from a sibling allograft because of either overt or unrecognized HLA disparities between the donor and the recipient. Whether or not an unrelated donor transplant can improve survival in high-risk AML is uncertain. A single study conducted by the Center for International Blood and Marrow Transplant Research (CIBMTR) and the National Marrow Donor Program (NMDP) analyzed the outcome of unrelated donor transplants for subgroups of AML patients according to cytogenetic risk. This study, carried out by Tallman and colleagues in collaboration with investigators for the AML Cytogenetics Study Group of the CIBMTR, evaluated transplants conducted between 1988 and 2002, and was presented at ASH 2004.3
TRANSPLANT OUTCOME BY CYTOGENETIC RISK The prognosis of AML patients depends on the type of cytogenetic abnormalities at diagnosis. According to the SWOG/ECOG criteria, favorable patterns are t(8;21), t(15;17), inv(16), and t(16;16), intermediate are normal karyotype, C6,C8, del(9q), del(12p), and KY, and unfavorable are del(5q)/K5, t(6;9), del(7q)/K7, t(9;22), inv(3q), 9q, 11q, 17p, 20q, 21q, and complex karyotype with at least 3 unrelated abnormalities.4 Tallman et al studied 358 AML patients transplanted in CR1.3 These patients were mostly young with a median age of 31 years (range 1–58). Four percent of the patients had favorable cytogenetics (a group too small for meaningful analysis), 71% (nZ249) had intermediate, and 25% (nZ89) had unfavorable cytogenetics. All patients received myeloablative transplants. Unrelated matches, defined at the antigen level for HLA-A, -B and -DR, accounted for 78% of the group and mismatches for the rest. Transplants in the early years of the study utilized technology far apart from the current state of the art. The median follow-up was 4.7 (0.5–12.1) years. The 5-year overall survival of the AML-CR1 patients with unfavorable cytogenetics was 28%, leukemia-free survival 26%, and the relapse rate 25%. While the nonrelapse mortality was unacceptably high (47%), the 5-year overall survival and leukemia-free survival compare well with the 5–15% reported for high-risk AML-CR1 patients receiving postremission chemotherapy alone. Results of unrelated donor transplantation were similar in the intermediate and unfavorable cytogenetic groups (Table 1).3 Some AML-CR1 patients receive transplants from unrelated donors because of high risk from variables other than cytogenetics that can be assessed by multivariate analysis. The relative risk of relapse after transplantation was 1.74 for the unfavorable compared to the intermediate cytogenetics group (pZ0.036), indicating that the antileukemia activity of an unrelated donor graft does not overcome the risk contributed by
Are matched unrelated donor transplants justified for AML in CR1? 323
Table 1. Outcome of unrelated donor transplants for AML patients in CR1.
Total Intermediate Unfavorable
n
OS 5 years (%)
DFS 5 years (%)
TRM 100 days (%)
Relapse 5 years (%)
358 254 91
29 30 28
28 29 26
33 32 34
19 17 25
OS, overall survival; DFS, disease-free survival; TRM, transplant-related mortality.
unfavorable cytogenetics. However, cytogenetics pattern did not have a significant influence on the overall survival and leukemia-free survival, because mortality in remission was the dominant cause of treatment failure. Survival estimates for patients with unfavorable cytogenetics in this study of unrelated donor transplantation appeared at least as good as survival previously reported for matched sibling transplant. The investigators concluded that patients with AML-CR1 with unfavorable cytogenetics and without a matched family donor should be offered an unrelated donor transplant.3,5
AML-CR1 IN CHILDREN: BONE MARROW TRANSPLANT OR NOT? In 2001, Woods et al published the results of Children’s Cancer Group (CCG) Study 289. In this study, children with AML who underwent a matched, related donor myeloablative HCT had superior survival compared to patients who had chemotherapy alone or chemotherapy and autologous HCT.6 A combined analysis of several CCG trials showed no benefit in overall survival for HCT patients who had inv(16)/t(16;16). Patients with t(8;21) had a 16% improvement in overall survival compared to chemotherapy alone, but this result was not statistically significant. In 2002, Burnett et al published the results of the Medical Research Council (MRC) AML 10 trial, which demonstrated equivalency in overall survival when HSCT was compared to chemotherapy in the subset of children included in the study.7 These data in a relatively small number of patients have supported the decision that HCT will not be offered on future trials to children with AML-CR1 with the favorable cytogenetic findings of either t(8;21) or inv16. Further justification for this approach is based on evidence that some patients who relapse after consolidation with chemotherapy can be salvaged with allogeneic HCT. In the United States and Canada, where survival of children with intermediate- and high-risk AML treated with chemotherapy appears to be somewhat inferior to that achieved on MRC trials, HCT will continue to be offered to patients in CR1 who have a matched family donor. There is no meaningful data of unrelated donor transplants in children with AML-CR1 and high-risk cytogenetics.
THE ROLE OF BLOOD STEM CELLS There have been improvements in transplant technology over the last 20 years, and more recent data from individual institutions demonstrate a better overall outcome of unrelated donor transplants for AML-CR1. For example, in a study of peripheral blood stem cell transplants at the Fred Hutchinson Cancer Research Center from 2000 to 2004, Kaplan–Meier curves showed a 67% survival beyond 1 year in 27 patients with
324 C. Anasetti et al
AML-CR1 (Anasetti C, unpublished observation). The role of blood versus marrow stem cells from unrelated donors remains the object of an ongoing randomized trial of the Blood and Marrow Transplantation—Collaborative Trial Network.
THE IMPACT OF MORE PRECISE DONOR HLA MATCHING There is a robust body of data demonstrating that more precise donor matching has improved the safety of unrelated transplants. For patients without an HLA-identical sibling, HCT from HLA-compatible unrelated donors has become possible owing to the expansion of registries of HLA-typed volunteer donors that has increased the probability of finding an unrelated match. For an average Caucasian, an HLA-A, -B, or -DR antigen matched donor is available to over 90% of the cases, where HLA antigens are defined by serologic typing. However, when all the polymorphic HLA allele sequences are examined by DNA typing, only half of the donors in one study were identical for HLA-A, -B, -C, and -DRB1 alleles; of the remaining, 25% had a single mismatch and 75% had multiple mismatches.8 Mismatching for any one HLA-A, -B, -C and -DRB1 alleles increased the risk of mortality. Because of these findings, donor selection utilizing high resolution HLA typing, including the HLA-C locus, has become required by the US NMDP. Using such criteria, a closely matched unrelated donor can now be found for 44% and up to 65% of the patients, if one allele mismatch is allowed. Thus, a prospective study of unrelated donor transplantation for AML-CR1 compared to chemotherapy has become feasible. Based on the frequency of HLA allele mismatches that went undetected with old serologic typing, and the effect of HLA allele mismatch on outcome as observed in the Flomenberg study, we estimated that if allele mismatches could be avoided by utilizing high resolution DNA typing, mortality would decrease from 70% to approximately 64% in the baseline Flomenberg study population, that is a 9–10% reduction in risk.8 Obviously, some patients have such rare combinations of HLA alleles that will never find a match, and therefore the potential decrease in the risk of mortality with the introduction of high resolution DNA typing is somewhat below 10%. Reduction in mortality with more precise donor HLA matching should enhance the safety and the risk/benefit ratio of unrelated donor transplantation for patients with high-risk AML-CR1.
A DONOR, QUICKLY! An important element for applying unrelated donor HSCT to high-risk AML-CR1 is minimizing the time interval between achieving CR1 and the time of transplantation. The median interval from remission to transplant in first CR in cooperative group trials of sibling transplants is 60–90 days. Ideally, the transplant should occur within 2 months of achieving CR1. With unrelated donor transplants, the interval from initiation of the search for donors to transplant has been about 4 months nation-wide. The first hurdle in rapidly finding an unrelated donor is getting the patient HLA typed promptly. If the hematologist/oncologist does not send a patient’s blood sample at diagnosis to the typing lab, there is a 1- to 2-month delay in the process. A further delay is sometimes encountered in obtaining approval for treatment reimbursement, which is dependent on the patient’s insurance carrier. There are now high-throughput HLA typing
Are matched unrelated donor transplants justified for AML in CR1? 325
laboratories that have decreased the cost of HLA typing, increased the efficiency, and greatly decreased the turnaround time. Some of these laboratories have contracts with the NMDP, so they can type both the donors and the patients. The NMDP has launched a series of pilot studies to assess whether assisting the transplant centers can facilitate donor selection, and whether speeding up the donor search process to as little as 21 days can be achieved when necessary. Thus, it is feasible nowadays to transplant urgent patients, such AML-CR1, from an unrelated donor within 60–90 days from remission induction.
SAFER MYELOABLATIVE CONDITIONING REGIMENS Busulfan/cyclophosphamide is a standard conditioning regimen for myeloid malignancies based on multiple randomized trials in siblings, and is sufficiently immune suppressive for engraftment of unrelated donor transplants. The Seattle group has shown in chronic myeloid leukemia (CML) that busulfan dose targeting based on pharmacokinetics minimizes toxicity, rejection, and relapse.9 The same group reported that aberrant cyclophosphamide metabolism when combined with total body radiation is associated with increased mortality.10 Fludarabine, a purine analogue, inhibits lymphocyte proliferation, promotes lymphocyte apoptosis, and is highly immunosuppressive with low non-hematopoietic toxicity. So there is rationale to test whether fludarabine can substitute for cyclophosphamide with less toxicity in a Flu/Bu regimen. Investigators at the University of Calgary and MD Anderson Cancer Center developed such a conditioning regimen as an effort to reduce toxicity. The inclusion of pharmacokinetics-targeted busulfan dosing as part of the Flu/Bu regimen was first explored at the Fred Hutchinson, in Seattle. In a trial conducted in collaboration with Dresden, Germany, a combination of 120 mg/m2 fludarabine and 16 mg/kg oral busulfan was delivered over 4 days with dose adjustments to plasma busulfan levels of 900G100 ng/mL.11 The steady-state plasma level of busulfan was chosen because previous studies had shown that maintaining the levels at or above this threshold could reduce the risk of relapse in patients with CML.9,12 Peripheral blood stem cells from HLA-compatible related or unrelated donors were used for transplantation, as with unrelated donor transplantation after Flu/Bu (without anti T-cell antibodies), the relatively low number of marrow cells compared to blood cells is thought insufficient to achieve engraftment. Therefore, donors were injected with granulocyte-colony stimulating factor, and apheresis products were collected for transplantation. Methotrexate and cyclosporine were used for acute GVHD prophylaxis. In the study, 42 high-risk patients with only advanced malignancy were included, such as blast phase CML, high-risk myelodysplastic syndromes (MDS), or MDS-transformed AML. Median age was 52 years (range 12–65). The busulfan concentration at steady state (CSS) was measured by the ratio of busulfan area under the curve divided by the dosing interval between doses (6 h). Doses of oral busulfan were linearly adjusted to achieve the target concentration if the CSS achieved was lower than 800 or more than 1000 ng/mL. Twenty-seven of the 42 patients required busulfan dose adjustment to achieve the required plasma value. The median total dose of busulfan administered was 14.4 mg/kg (range, 9.8–18.7 mg/kg) actual body weight and 15.9 mg/kg (range, 11–21.8 mg/kg) ideal body weight. Engraftment occurred in all patients with donor chimerism of the CD3 T–cell lineage at about 75% and the granulocyte lineage approaching 100% at 2.5 months after transplant. When the
326 C. Anasetti et al
chimerism was tested at 1 year, donor type was a median of 99.5% for granulocytes and 98% for T cells. This study had no late graft failure. The mortality in patients with advanced myeloid malignancy was 7% at 100 days and approximately 24% at 2 years, and did not differ for related and unrelated donor transplants.11 The busulfan mode of delivery has also been investigated. Using data on oral busulfan usage collected by CIBMTR from over 550 institutions worldwide, a matched-pair analysis was performed comparing this data set against that including intravenous (IV) busulfan usage from four clinical studies facilitated by ESP Pharma.13 The analysis compared two key clinically important variables in patients receiving IVor oral busulfan: overall survival to day 100 and the incidence of hepatic veno-occlusive disease and mortality through posttransplant day C28. The CIBMTR/ESP Pharma analysis was presented at ASH 2004 and showed that with IV busulfan substituted for the oral formulation, there was a significant decrease in the risk of veno-occlusive disease and transplant-related mortality (Table 2). The overall incidence of hepatic veno-occlusive disease at day 28 was 4.6% (4/83) with IV busulfan and 20.3% (38/149) with oral busulfan (p!0.001). Among the autologous transplant recipients, 100-day mortality was 0% for those receiving IV, and 9.3% for those receiving oral busulfan (pZ0.16), whereas for allotransplant recipients, 100-day mortality was 8.7% with IV and 22.5% with oral busulfan (pZ0.015). Logistic regression analyses by treatment group determined that IV busulfan was associated with a lower probability of 100-day mortality compared to the oral mode for all patients combined (pZ0.005). Thus, there was a lower early mortality with IV busulfan administration as compared to oral for the outcome of HCT.13 A study from MD Anderson examined the outcome of allogeneic HCT patients !65 years treated with Flu/Bu IV for MDS and AML.14 The preparative regimens contained fludarabine 40 mg/m2 and busulfan IV 130 mg/m2 given daily four times. GVHD prophylaxis was tacrolimus and methotrexate. Stem cells were from bone marrow in 47 patients and peripheral blood in 49. Sixty patients received grafts from related donors and 36 had unrelated grafts. The overall survival at 1 year was 65% and relapsefree survival 52% with no differences between sibling and unrelated donor grafts. The event-free survival for patients at any stage with deletions of chromosomes 5 or 7 was 34%. Event-free survival was 78% for patients in any remission. Field et al from Moffitt Cancer Center reported the early results of a targeted IV busulfan pharmacokinetic dosing strategy used in combination with fludarabine before either related (nZ29) or unrelated (nZ27) blood stem cell transplantation.15 The median patient age was 48 (range 22–68) years. Patient diagnoses included AML (24), ALL (6), MF (6), NHL (5), MDS (4), CML (4), CLL (2), MM (2), CMML (1), plasma cell leukemia (1) and PNH (1). The CIBMTR disease risk score was high in 42 patients,
Table 2. Matched pairs analysis of IV versus oral busulfan.
Venous occlusive disease Autologous 100 day TRM Allogeneic 100 day TRM
BU IV ESP (%) nZ101
Oral BU CIBMTR (%) nZ216
5 0 8.7
20 9.3 22.5
p value !0.001 0.16 0.015
BU, busulfan; CIBMTR, Center for International Blood and Marrow Transplant Research; ESP, ESP Pharma; TRM, transplant-related mortality; d, day; IV, intravenous.
Are matched unrelated donor transplants justified for AML in CR1? 327
Overall Survival 1.0
Cumulative Proportion Surviving
0.9 0.8 0.7 0.6 0.5 0.4 0.3 Sibling donors (n=29) Matched unrelated donors (n=27) p=0.74
0.2 0.1 0.0
0
2
4
8 10 12 6 Months After Transplant
14
16
18
Figure 1. Overall survival of HLA matched sibling or unrelated donor transplants for malignancy after condition with Flu/Bu IV targeted to 800–1000 ng/ml Css.
intermediate in 3, low in 10, and indeterminate in 1. Fludarabine 40 mg/m2 was given intravenously daily for four days, with each infusion followed immediately by intravenous busulfan. The dose of busulfan for days 1 and 2 was 130 mg/m2. Pharmacokinetic analysis was performed after the first infusion of busulfan; in 54 patients, the goal was to adjust busulfan doses for days 3 and 4 to achieve an average targeted Css level of 800–1000 ng/mL. Levels were either not drawn or drawn incorrectly in 3 patients and doses were not changed. Thirty-one of the remaining 51 (61%) patients had their doses adjusted, increased in 23 and decreased in 8, while 20 patients had Css within the desired range without adjustment. Patients received tacrolimus and standard doses of methotrexate for GVHD prophylaxis, with five exceptions. Engraftment occurred in 54 (97%) patients. Median follow-up is 174 days (range 100–491 days). Six (11%) patients have died of nonrelapse causes. The 6-month Kaplan–Meier estimate of overall survival for the whole cohort is 79%, and event-free survival 62%. Overall survival was similar for sibling and unrelated donor transplants (Figure 1). These data support the concepts that pharmacokinetics-based busulfan targeting is helpful to assure that all patients receive the desirable busulfan dose, despite intrevenous administration. This Flu/Bu IV targeted regimen appears gentle, even in patients aged 40–65 years, and it is likely to improve both the safety and the utilization of unrelated donor transplant for AML-CR1.
CONCLUSION Initial registry data suggest that AML-CR1 patients with high-risk cytogenetics and without an HLA-identical sibling would benefit from an unrelated donor graft over consolidation chemotherapy alone. The technology for donor selection is evolving, and we anticipate will
328 C. Anasetti et al
lead to improve in the outcomes of unrelated donor transplants over time. The development of safer conditioning regimens, such as the combination of fludarabine and intravenous busulfan, has decreased mortality of related and unrelated transplants and we anticipate will enhance the therapeutic benefits of allografting for AML.
REFERENCES 1. Yanada M, Matsuo K, Emi N & Naoe T. Efficacy of allogeneic hematopoietic stem cell transplantation depends on cytogenetic risk for acute myeloid leukemia in first disease remission: a metaanalysis. Cancer 2005; 103: 1652–1658. 2. Appelbaum FR. Haematopoietic cell transplantation as immunotherapy. Nature 2001; 411: 385–389. 3. Tallman M, Dewald G, Lazarus HM et al. Impact of cytogenetics on outcome of unrelated donor hematopoietic stem cell transplantation for adults with AML. Blood 2004; 104: 237a [abstr 830]. 4. Slovak ML, Kopecky KJ, Cassileth PA et al. Karyotypic analysis predicts outcome of preremission and postremission therapy in adult acute myeloid leukemia: a Southwest Oncology Group/Eastern Cooperative Oncology Group Study. Blood 2000; 96: 4075–4083. 5. Kim DH, Sohn SK, Kim JG et al. Parameters for predicting allogeneic PBSCToutcome of acute myeloid leukemia: cytogenetics at presentation versus disease status at transplantation. Annals of Hematology 2005; 84: 25–32. 6. Woods WG, Neudorf S, Gold S et al. A comparison of allogeneic bone marrow transplantation, autologous bone marrow transplantation, and aggressive chemotherapy in children with acute myeloid leukemia in remission. Blood 2001; 97: 56–62. 7. Burnett AK, Wheatley K, Goldstone AH et al. The value of allogeneic bone marrow transplant in patients with acute myeloid leukaemia at differing risk of relapse: results of the UK MRC AML 10 trial. British Journal of Haematology 2002; 118: 385–400. 8. Flomenberg N, Baxter-Lowe LA, Confer D et al. Impact of HLA class I and class II high resolution matching on outcomes of unrelated donor bone marrow transplantation: HLA-C mismatching is associated with a strong adverse effect on transplant outcome. Blood 2004; 104: 1923–1930. 9. Slattery JT, Clift RA, Buckner CD et al. Marrow transplantation for chronic myeloid leukemia: the influence of plasma busulfan levels on the outcome of transplantation. Blood 1997; 89: 3055–3060. 10. McDonald GB, Slattery JT, Bouvier ME et al. Cyclophosphamide metabolism, liver toxicity, and mortality following hematopoietic stem cell transplantation. Blood 2003; 101: 2043–2048. 11. Bornhauser M, Storer B, Slattery JT et al. Conditioning with fludarabine and targeted busulfan for transplantation of allogeneic hematopoietic stem cells. Blood 2003; 102: 820–826. 12. Radich JP, Gooley T, Bensinger W et al. HLA-matched related hematopoietic cell transplantation for chronicphase CML using a targeted busulfan and cyclophosphamide preparative regimen. Blood 2003; 102: 31–35. 13. Sobocinski KA, Thall PF, Bekele N et al. Matched Pairs analysis of IV vs. PO Busulfan as a conditioning agent prior to transplantation. Blood 2004; 104: 103a [abstr 349, Ref Type: Abstract]. 14. de Lima M, Shahjahan M, Alamo A et al. Allogeneic hematopoietic stem cell transplantation (HSCT) for patients aged 65 years or older with AML and MDS. Blood 2004; 104: 632a [abstr 2301]. 15. Field T, Perkins J, Alsina M et al. Targeted intravenous busulfan combined with fludarabine (tBuFlu) as conditioning prior to allogeneic peripheral blood stem cell transplant: early results of low treatment related mortality. Blood 2005; 106: 500a [abstr 1759].
7 Are matched unrelated donor transplants justified for AML in CR1? Claudio Anasetti* Janelle Perkins
Pherm D
Michael L. Nieder Teresa Field
MD
MD
MD PhD
Blood and Marrow Transplantation Program, University of South Florida; H. Lee Moffitt Cancer Center and Research Institute, 12901 Magnolia Drive, Tampa, FL 33612 USA; and All Children’s Hospital 801 Sixth Street South, St. Petersburg, FL 33701, USA.
There has been controversy about the optimal consolidation therapy for patients with acute myeloid leukemia (AML) in first remission. Hematopoietic stem cell transplantation from human leukocyte antigen (HLA)-identical siblings has improved the survival of patients with unfavorable cytogenetics, but has not improved the survival of intermediate- or favorable-risk patients. If an HLA-identical sibling donor is not available, alternative sources of stem cells may be sought. HLA mismatched transplants are associated with an increased risk of graft rejection and graft-versushost disease, and lower survival. Since large registries of HLA-typed volunteer donors have been established and the newer and more sensitive tissue typing technology can more clearly differentiate between matched and unmatched donors, some AML patients without an HLAmatched sibling have received transplants from an HLA-matched unrelated donor while in first remission. Data from the Center for International Blood and Marrow Transplant Research indicate that survival of patients with unfavorable cytogenetics appears at least as good with unrelated donor grafts as previously reported for matched sibling grafts, and better than consolidation chemotherapy. AML patients in first remission with unfavorable cytogenetics without a matched family donor should be offered an unrelated donor transplant. Key words: cell transplantation; cytogenetics; myeloid.
Hematopoietic stem cell transplantation (HCT) from an HLA-identical sibling donor improves survival of acute myeloid leukemia (AML) patients in first complete remission * Corresponding author. Tel.: C1 813 745 2557; Fax: C1 813 745 3071. E-mail address: [email protected] (C. Anasetti).
1521-6926/$ - see front matter Q 2006 Elsevier Ltd. All rights reserved.
322 C. Anasetti et al
(AML-CR1) with high-risk cytogenetics; the 5-year survival of these patients is 20–30% using HCT compared to 5–15% using chemotherapy alone as consolidation. The data for intermediate-risk patients are controversial, and data for low-risk AML-CR1 indicate worse survival with an allograft, as higher treatment-related mortality exceeds the benefits of the allograft. A meta-analysis of five randomized clinical trials comparing HLA-identical sibling grafts to autografts or chemotherapy for AML-CR1 supports these conclusions.1 Much of the benefit of the allografts derives from the graft-versus-leukemia effect.2 Unfortunately, only about 30% of the patients in the US and Europe have an HLAidentical sibling, but most find an HLA-compatible unrelated volunteer among the 10 million donors registered worldwide. Transplants from unrelated donors, however, have been associated with greater risk of rejection, graft-versus-host disease (GVHD), and mortality than those from a sibling allograft because of either overt or unrecognized HLA disparities between the donor and the recipient. Whether or not an unrelated donor transplant can improve survival in high-risk AML is uncertain. A single study conducted by the Center for International Blood and Marrow Transplant Research (CIBMTR) and the National Marrow Donor Program (NMDP) analyzed the outcome of unrelated donor transplants for subgroups of AML patients according to cytogenetic risk. This study, carried out by Tallman and colleagues in collaboration with investigators for the AML Cytogenetics Study Group of the CIBMTR, evaluated transplants conducted between 1988 and 2002, and was presented at ASH 2004.3
TRANSPLANT OUTCOME BY CYTOGENETIC RISK The prognosis of AML patients depends on the type of cytogenetic abnormalities at diagnosis. According to the SWOG/ECOG criteria, favorable patterns are t(8;21), t(15;17), inv(16), and t(16;16), intermediate are normal karyotype, C6,C8, del(9q), del(12p), and KY, and unfavorable are del(5q)/K5, t(6;9), del(7q)/K7, t(9;22), inv(3q), 9q, 11q, 17p, 20q, 21q, and complex karyotype with at least 3 unrelated abnormalities.4 Tallman et al studied 358 AML patients transplanted in CR1.3 These patients were mostly young with a median age of 31 years (range 1–58). Four percent of the patients had favorable cytogenetics (a group too small for meaningful analysis), 71% (nZ249) had intermediate, and 25% (nZ89) had unfavorable cytogenetics. All patients received myeloablative transplants. Unrelated matches, defined at the antigen level for HLA-A, -B and -DR, accounted for 78% of the group and mismatches for the rest. Transplants in the early years of the study utilized technology far apart from the current state of the art. The median follow-up was 4.7 (0.5–12.1) years. The 5-year overall survival of the AML-CR1 patients with unfavorable cytogenetics was 28%, leukemia-free survival 26%, and the relapse rate 25%. While the nonrelapse mortality was unacceptably high (47%), the 5-year overall survival and leukemia-free survival compare well with the 5–15% reported for high-risk AML-CR1 patients receiving postremission chemotherapy alone. Results of unrelated donor transplantation were similar in the intermediate and unfavorable cytogenetic groups (Table 1).3 Some AML-CR1 patients receive transplants from unrelated donors because of high risk from variables other than cytogenetics that can be assessed by multivariate analysis. The relative risk of relapse after transplantation was 1.74 for the unfavorable compared to the intermediate cytogenetics group (pZ0.036), indicating that the antileukemia activity of an unrelated donor graft does not overcome the risk contributed by
Are matched unrelated donor transplants justified for AML in CR1? 323
Table 1. Outcome of unrelated donor transplants for AML patients in CR1.
Total Intermediate Unfavorable
n
OS 5 years (%)
DFS 5 years (%)
TRM 100 days (%)
Relapse 5 years (%)
358 254 91
29 30 28
28 29 26
33 32 34
19 17 25
OS, overall survival; DFS, disease-free survival; TRM, transplant-related mortality.
unfavorable cytogenetics. However, cytogenetics pattern did not have a significant influence on the overall survival and leukemia-free survival, because mortality in remission was the dominant cause of treatment failure. Survival estimates for patients with unfavorable cytogenetics in this study of unrelated donor transplantation appeared at least as good as survival previously reported for matched sibling transplant. The investigators concluded that patients with AML-CR1 with unfavorable cytogenetics and without a matched family donor should be offered an unrelated donor transplant.3,5
AML-CR1 IN CHILDREN: BONE MARROW TRANSPLANT OR NOT? In 2001, Woods et al published the results of Children’s Cancer Group (CCG) Study 289. In this study, children with AML who underwent a matched, related donor myeloablative HCT had superior survival compared to patients who had chemotherapy alone or chemotherapy and autologous HCT.6 A combined analysis of several CCG trials showed no benefit in overall survival for HCT patients who had inv(16)/t(16;16). Patients with t(8;21) had a 16% improvement in overall survival compared to chemotherapy alone, but this result was not statistically significant. In 2002, Burnett et al published the results of the Medical Research Council (MRC) AML 10 trial, which demonstrated equivalency in overall survival when HSCT was compared to chemotherapy in the subset of children included in the study.7 These data in a relatively small number of patients have supported the decision that HCT will not be offered on future trials to children with AML-CR1 with the favorable cytogenetic findings of either t(8;21) or inv16. Further justification for this approach is based on evidence that some patients who relapse after consolidation with chemotherapy can be salvaged with allogeneic HCT. In the United States and Canada, where survival of children with intermediate- and high-risk AML treated with chemotherapy appears to be somewhat inferior to that achieved on MRC trials, HCT will continue to be offered to patients in CR1 who have a matched family donor. There is no meaningful data of unrelated donor transplants in children with AML-CR1 and high-risk cytogenetics.
THE ROLE OF BLOOD STEM CELLS There have been improvements in transplant technology over the last 20 years, and more recent data from individual institutions demonstrate a better overall outcome of unrelated donor transplants for AML-CR1. For example, in a study of peripheral blood stem cell transplants at the Fred Hutchinson Cancer Research Center from 2000 to 2004, Kaplan–Meier curves showed a 67% survival beyond 1 year in 27 patients with
324 C. Anasetti et al
AML-CR1 (Anasetti C, unpublished observation). The role of blood versus marrow stem cells from unrelated donors remains the object of an ongoing randomized trial of the Blood and Marrow Transplantation—Collaborative Trial Network.
THE IMPACT OF MORE PRECISE DONOR HLA MATCHING There is a robust body of data demonstrating that more precise donor matching has improved the safety of unrelated transplants. For patients without an HLA-identical sibling, HCT from HLA-compatible unrelated donors has become possible owing to the expansion of registries of HLA-typed volunteer donors that has increased the probability of finding an unrelated match. For an average Caucasian, an HLA-A, -B, or -DR antigen matched donor is available to over 90% of the cases, where HLA antigens are defined by serologic typing. However, when all the polymorphic HLA allele sequences are examined by DNA typing, only half of the donors in one study were identical for HLA-A, -B, -C, and -DRB1 alleles; of the remaining, 25% had a single mismatch and 75% had multiple mismatches.8 Mismatching for any one HLA-A, -B, -C and -DRB1 alleles increased the risk of mortality. Because of these findings, donor selection utilizing high resolution HLA typing, including the HLA-C locus, has become required by the US NMDP. Using such criteria, a closely matched unrelated donor can now be found for 44% and up to 65% of the patients, if one allele mismatch is allowed. Thus, a prospective study of unrelated donor transplantation for AML-CR1 compared to chemotherapy has become feasible. Based on the frequency of HLA allele mismatches that went undetected with old serologic typing, and the effect of HLA allele mismatch on outcome as observed in the Flomenberg study, we estimated that if allele mismatches could be avoided by utilizing high resolution DNA typing, mortality would decrease from 70% to approximately 64% in the baseline Flomenberg study population, that is a 9–10% reduction in risk.8 Obviously, some patients have such rare combinations of HLA alleles that will never find a match, and therefore the potential decrease in the risk of mortality with the introduction of high resolution DNA typing is somewhat below 10%. Reduction in mortality with more precise donor HLA matching should enhance the safety and the risk/benefit ratio of unrelated donor transplantation for patients with high-risk AML-CR1.
A DONOR, QUICKLY! An important element for applying unrelated donor HSCT to high-risk AML-CR1 is minimizing the time interval between achieving CR1 and the time of transplantation. The median interval from remission to transplant in first CR in cooperative group trials of sibling transplants is 60–90 days. Ideally, the transplant should occur within 2 months of achieving CR1. With unrelated donor transplants, the interval from initiation of the search for donors to transplant has been about 4 months nation-wide. The first hurdle in rapidly finding an unrelated donor is getting the patient HLA typed promptly. If the hematologist/oncologist does not send a patient’s blood sample at diagnosis to the typing lab, there is a 1- to 2-month delay in the process. A further delay is sometimes encountered in obtaining approval for treatment reimbursement, which is dependent on the patient’s insurance carrier. There are now high-throughput HLA typing
Are matched unrelated donor transplants justified for AML in CR1? 325
laboratories that have decreased the cost of HLA typing, increased the efficiency, and greatly decreased the turnaround time. Some of these laboratories have contracts with the NMDP, so they can type both the donors and the patients. The NMDP has launched a series of pilot studies to assess whether assisting the transplant centers can facilitate donor selection, and whether speeding up the donor search process to as little as 21 days can be achieved when necessary. Thus, it is feasible nowadays to transplant urgent patients, such AML-CR1, from an unrelated donor within 60–90 days from remission induction.
SAFER MYELOABLATIVE CONDITIONING REGIMENS Busulfan/cyclophosphamide is a standard conditioning regimen for myeloid malignancies based on multiple randomized trials in siblings, and is sufficiently immune suppressive for engraftment of unrelated donor transplants. The Seattle group has shown in chronic myeloid leukemia (CML) that busulfan dose targeting based on pharmacokinetics minimizes toxicity, rejection, and relapse.9 The same group reported that aberrant cyclophosphamide metabolism when combined with total body radiation is associated with increased mortality.10 Fludarabine, a purine analogue, inhibits lymphocyte proliferation, promotes lymphocyte apoptosis, and is highly immunosuppressive with low non-hematopoietic toxicity. So there is rationale to test whether fludarabine can substitute for cyclophosphamide with less toxicity in a Flu/Bu regimen. Investigators at the University of Calgary and MD Anderson Cancer Center developed such a conditioning regimen as an effort to reduce toxicity. The inclusion of pharmacokinetics-targeted busulfan dosing as part of the Flu/Bu regimen was first explored at the Fred Hutchinson, in Seattle. In a trial conducted in collaboration with Dresden, Germany, a combination of 120 mg/m2 fludarabine and 16 mg/kg oral busulfan was delivered over 4 days with dose adjustments to plasma busulfan levels of 900G100 ng/mL.11 The steady-state plasma level of busulfan was chosen because previous studies had shown that maintaining the levels at or above this threshold could reduce the risk of relapse in patients with CML.9,12 Peripheral blood stem cells from HLA-compatible related or unrelated donors were used for transplantation, as with unrelated donor transplantation after Flu/Bu (without anti T-cell antibodies), the relatively low number of marrow cells compared to blood cells is thought insufficient to achieve engraftment. Therefore, donors were injected with granulocyte-colony stimulating factor, and apheresis products were collected for transplantation. Methotrexate and cyclosporine were used for acute GVHD prophylaxis. In the study, 42 high-risk patients with only advanced malignancy were included, such as blast phase CML, high-risk myelodysplastic syndromes (MDS), or MDS-transformed AML. Median age was 52 years (range 12–65). The busulfan concentration at steady state (CSS) was measured by the ratio of busulfan area under the curve divided by the dosing interval between doses (6 h). Doses of oral busulfan were linearly adjusted to achieve the target concentration if the CSS achieved was lower than 800 or more than 1000 ng/mL. Twenty-seven of the 42 patients required busulfan dose adjustment to achieve the required plasma value. The median total dose of busulfan administered was 14.4 mg/kg (range, 9.8–18.7 mg/kg) actual body weight and 15.9 mg/kg (range, 11–21.8 mg/kg) ideal body weight. Engraftment occurred in all patients with donor chimerism of the CD3 T–cell lineage at about 75% and the granulocyte lineage approaching 100% at 2.5 months after transplant. When the
326 C. Anasetti et al
chimerism was tested at 1 year, donor type was a median of 99.5% for granulocytes and 98% for T cells. This study had no late graft failure. The mortality in patients with advanced myeloid malignancy was 7% at 100 days and approximately 24% at 2 years, and did not differ for related and unrelated donor transplants.11 The busulfan mode of delivery has also been investigated. Using data on oral busulfan usage collected by CIBMTR from over 550 institutions worldwide, a matched-pair analysis was performed comparing this data set against that including intravenous (IV) busulfan usage from four clinical studies facilitated by ESP Pharma.13 The analysis compared two key clinically important variables in patients receiving IVor oral busulfan: overall survival to day 100 and the incidence of hepatic veno-occlusive disease and mortality through posttransplant day C28. The CIBMTR/ESP Pharma analysis was presented at ASH 2004 and showed that with IV busulfan substituted for the oral formulation, there was a significant decrease in the risk of veno-occlusive disease and transplant-related mortality (Table 2). The overall incidence of hepatic veno-occlusive disease at day 28 was 4.6% (4/83) with IV busulfan and 20.3% (38/149) with oral busulfan (p!0.001). Among the autologous transplant recipients, 100-day mortality was 0% for those receiving IV, and 9.3% for those receiving oral busulfan (pZ0.16), whereas for allotransplant recipients, 100-day mortality was 8.7% with IV and 22.5% with oral busulfan (pZ0.015). Logistic regression analyses by treatment group determined that IV busulfan was associated with a lower probability of 100-day mortality compared to the oral mode for all patients combined (pZ0.005). Thus, there was a lower early mortality with IV busulfan administration as compared to oral for the outcome of HCT.13 A study from MD Anderson examined the outcome of allogeneic HCT patients !65 years treated with Flu/Bu IV for MDS and AML.14 The preparative regimens contained fludarabine 40 mg/m2 and busulfan IV 130 mg/m2 given daily four times. GVHD prophylaxis was tacrolimus and methotrexate. Stem cells were from bone marrow in 47 patients and peripheral blood in 49. Sixty patients received grafts from related donors and 36 had unrelated grafts. The overall survival at 1 year was 65% and relapsefree survival 52% with no differences between sibling and unrelated donor grafts. The event-free survival for patients at any stage with deletions of chromosomes 5 or 7 was 34%. Event-free survival was 78% for patients in any remission. Field et al from Moffitt Cancer Center reported the early results of a targeted IV busulfan pharmacokinetic dosing strategy used in combination with fludarabine before either related (nZ29) or unrelated (nZ27) blood stem cell transplantation.15 The median patient age was 48 (range 22–68) years. Patient diagnoses included AML (24), ALL (6), MF (6), NHL (5), MDS (4), CML (4), CLL (2), MM (2), CMML (1), plasma cell leukemia (1) and PNH (1). The CIBMTR disease risk score was high in 42 patients,
Table 2. Matched pairs analysis of IV versus oral busulfan.
Venous occlusive disease Autologous 100 day TRM Allogeneic 100 day TRM
BU IV ESP (%) nZ101
Oral BU CIBMTR (%) nZ216
5 0 8.7
20 9.3 22.5
p value !0.001 0.16 0.015
BU, busulfan; CIBMTR, Center for International Blood and Marrow Transplant Research; ESP, ESP Pharma; TRM, transplant-related mortality; d, day; IV, intravenous.
Are matched unrelated donor transplants justified for AML in CR1? 327
Overall Survival 1.0
Cumulative Proportion Surviving
0.9 0.8 0.7 0.6 0.5 0.4 0.3 Sibling donors (n=29) Matched unrelated donors (n=27) p=0.74
0.2 0.1 0.0
0
2
4
8 10 12 6 Months After Transplant
14
16
18
Figure 1. Overall survival of HLA matched sibling or unrelated donor transplants for malignancy after condition with Flu/Bu IV targeted to 800–1000 ng/ml Css.
intermediate in 3, low in 10, and indeterminate in 1. Fludarabine 40 mg/m2 was given intravenously daily for four days, with each infusion followed immediately by intravenous busulfan. The dose of busulfan for days 1 and 2 was 130 mg/m2. Pharmacokinetic analysis was performed after the first infusion of busulfan; in 54 patients, the goal was to adjust busulfan doses for days 3 and 4 to achieve an average targeted Css level of 800–1000 ng/mL. Levels were either not drawn or drawn incorrectly in 3 patients and doses were not changed. Thirty-one of the remaining 51 (61%) patients had their doses adjusted, increased in 23 and decreased in 8, while 20 patients had Css within the desired range without adjustment. Patients received tacrolimus and standard doses of methotrexate for GVHD prophylaxis, with five exceptions. Engraftment occurred in 54 (97%) patients. Median follow-up is 174 days (range 100–491 days). Six (11%) patients have died of nonrelapse causes. The 6-month Kaplan–Meier estimate of overall survival for the whole cohort is 79%, and event-free survival 62%. Overall survival was similar for sibling and unrelated donor transplants (Figure 1). These data support the concepts that pharmacokinetics-based busulfan targeting is helpful to assure that all patients receive the desirable busulfan dose, despite intrevenous administration. This Flu/Bu IV targeted regimen appears gentle, even in patients aged 40–65 years, and it is likely to improve both the safety and the utilization of unrelated donor transplant for AML-CR1.
CONCLUSION Initial registry data suggest that AML-CR1 patients with high-risk cytogenetics and without an HLA-identical sibling would benefit from an unrelated donor graft over consolidation chemotherapy alone. The technology for donor selection is evolving, and we anticipate will
328 C. Anasetti et al
lead to improve in the outcomes of unrelated donor transplants over time. The development of safer conditioning regimens, such as the combination of fludarabine and intravenous busulfan, has decreased mortality of related and unrelated transplants and we anticipate will enhance the therapeutic benefits of allografting for AML.
REFERENCES 1. Yanada M, Matsuo K, Emi N & Naoe T. Efficacy of allogeneic hematopoietic stem cell transplantation depends on cytogenetic risk for acute myeloid leukemia in first disease remission: a metaanalysis. Cancer 2005; 103: 1652–1658. 2. Appelbaum FR. Haematopoietic cell transplantation as immunotherapy. Nature 2001; 411: 385–389. 3. Tallman M, Dewald G, Lazarus HM et al. Impact of cytogenetics on outcome of unrelated donor hematopoietic stem cell transplantation for adults with AML. Blood 2004; 104: 237a [abstr 830]. 4. Slovak ML, Kopecky KJ, Cassileth PA et al. Karyotypic analysis predicts outcome of preremission and postremission therapy in adult acute myeloid leukemia: a Southwest Oncology Group/Eastern Cooperative Oncology Group Study. Blood 2000; 96: 4075–4083. 5. Kim DH, Sohn SK, Kim JG et al. Parameters for predicting allogeneic PBSCToutcome of acute myeloid leukemia: cytogenetics at presentation versus disease status at transplantation. Annals of Hematology 2005; 84: 25–32. 6. Woods WG, Neudorf S, Gold S et al. A comparison of allogeneic bone marrow transplantation, autologous bone marrow transplantation, and aggressive chemotherapy in children with acute myeloid leukemia in remission. Blood 2001; 97: 56–62. 7. Burnett AK, Wheatley K, Goldstone AH et al. The value of allogeneic bone marrow transplant in patients with acute myeloid leukaemia at differing risk of relapse: results of the UK MRC AML 10 trial. British Journal of Haematology 2002; 118: 385–400. 8. Flomenberg N, Baxter-Lowe LA, Confer D et al. Impact of HLA class I and class II high resolution matching on outcomes of unrelated donor bone marrow transplantation: HLA-C mismatching is associated with a strong adverse effect on transplant outcome. Blood 2004; 104: 1923–1930. 9. Slattery JT, Clift RA, Buckner CD et al. Marrow transplantation for chronic myeloid leukemia: the influence of plasma busulfan levels on the outcome of transplantation. Blood 1997; 89: 3055–3060. 10. McDonald GB, Slattery JT, Bouvier ME et al. Cyclophosphamide metabolism, liver toxicity, and mortality following hematopoietic stem cell transplantation. Blood 2003; 101: 2043–2048. 11. Bornhauser M, Storer B, Slattery JT et al. Conditioning with fludarabine and targeted busulfan for transplantation of allogeneic hematopoietic stem cells. Blood 2003; 102: 820–826. 12. Radich JP, Gooley T, Bensinger W et al. HLA-matched related hematopoietic cell transplantation for chronicphase CML using a targeted busulfan and cyclophosphamide preparative regimen. Blood 2003; 102: 31–35. 13. Sobocinski KA, Thall PF, Bekele N et al. Matched Pairs analysis of IV vs. PO Busulfan as a conditioning agent prior to transplantation. Blood 2004; 104: 103a [abstr 349, Ref Type: Abstract]. 14. de Lima M, Shahjahan M, Alamo A et al. Allogeneic hematopoietic stem cell transplantation (HSCT) for patients aged 65 years or older with AML and MDS. Blood 2004; 104: 632a [abstr 2301]. 15. Field T, Perkins J, Alsina M et al. Targeted intravenous busulfan combined with fludarabine (tBuFlu) as conditioning prior to allogeneic peripheral blood stem cell transplant: early results of low treatment related mortality. Blood 2005; 106: 500a [abstr 1759].