1. Aetna considers allogeneic bone marrow or peripheral stem cell transplant medically necessary for the treatment of acute lymphocytic leukemia (ALL) when members meet the transplanting institution's selection criteria.  In the absence of an institution's selection criteria, Aetna considers allogeneic bone marrow or peripheral stem cell transplant medically necessary for the treatment of ALL, including primary refractory ALL (i.e., leukemia that does not achieve a complete remission after conventional dose chemotherapy), except for members in refractory relapse, defined as persons in relapse who are unresponsive to 3 or more months of adequate chemotherapy.

2. Aetna considers non-myeloablative allogeneic transplantation, also known as mini-allograft or reduced intensity conditioning transplant, medically necessary for the treatment of ALL for members with no persistent disease who meet all of the selection criteria above.  Note: Persons with persistent disease should not be candidates for a mini-allograft transplant.

3. Aetna considers autologous bone marrow or peripheral stem cell transplant medically necessary for persons with standard risk ALL where no suitable donor is available.

4. Aetna considers tandem (also known as sequential) transplants experimental and investigational for the treatment of ALL because their effectiveness for this indication has not been estanlished.

5. Aetna considers autologous hematopoietic stem cell transplantation (auto-HSCT) medically necessary for the treatment of acute myelogenous leukemia (AML) when members meet the transplanting institution's selection criteria.  In the absence of an institution's selection criteria, Aetna considers auto-HSCT medically necessary for the treatment of AML for any indication (e.g., first or second remission or relapsed AML if responsive to intensified induction chemotherapy) except as first-line treatment.

6. Aetna considers allogeneic hematopoietic stem cell transplantation (allo-HSCT) (ablative or mini-allograft) medically necessary for the treatment of AML when members meet the transplanting institution's selection criteria.  In the absence of an institution's selection criteria, Aetna considers allo-HSCT for the treatment of AML in any one of the following indications:

   1. Members who have relapsed following a prior auto-HSCT and who are medically able to tolerate the procedure; or
   2. Poor-risk to intermediate-risk AML in remission; or

   3. Primary refractory AML (i.e., leukemia that does not achieve a complete remission after conventional dose chemotherapy)

7. Aetna considers a repeat auto-HSCT or allo-HSCT (ablative or mini-allograft) medically necessary for the treatment of AML when members meet the transplanting institution's selection criteria.  In the absence of an institution's selection criteria, Aetna considers a repeat auto-HSCT or allo-HSCT (ablative or mini-allograft) medically necessary when the first auto-HSCT/allo-HSCT was unsuccessful due to primary graft failure or failure to engraft.

8. Aetna considers a repeat auto-HSCT or allo-HSCT (ablative or mini-allograft) experimental and investigational for members with persistent, progressive or relapsed disease because its effectiveness for this indication has not been estanlished.

9. Aetna considers allogeneic (ablative or non-myeloablative) HSCT medically necessary for the treatment of chronic myelo-monocytic leukemia (CMML) and juvenile myelo-monocytic leukemia (JMML) when a human leukocyte antigen (HLA)-matched donor (at least 5 of 6 match) is available.

10. Aetna considers a repeat allogeneic (ablative or non-myeloablative) HSCT due to primary graft failure or failure to engraft medically necessary for the treatment of CMML and JMML.

11. Aetna considers autologous HSCT for the treatment of CMML and JMML experimental and investigational because its effectiveness for these indications has not been established.

12. Aetna considers allogeneic (ablative or non-myeloablative) HSCT medically necessary for T-cell prolymphocytic leukemia.

See also: CPB 0634 - Non-myeloablative Bone Marrow/Peripheral Stem Cell Transplantation (Mini-Allograft/Reduced Intensity Conditioning Transplant), and CPB 0674 - High Dose Chemotherapy and Bone Marrow or Peripheral Stem Cell Transplantation for Chronic Myelogenous Leukemia.

Acute Lymphocytic Leukemia:

Acute lymphocytic leukemia (ALL) is a heterogeneous group of malignancies arising from lymphocytic precursors.  The heterogeneity is related to whether the malignant clone is derived from a T or B cell as evidenced by the expression of different surface antigens.  The different subtypes of ALL are also heterogeneous with respect to response to chemotherapy and to the age distribution (i.e., different subtypes typically occur in either children or adults).  Acute lymphocytic leukemia has a bi-modal age distribution with an initial peak at 2 to 3 years, with the incidence again increasing after the age of 50.  By definition, ALL primarily affects the bone marrow, but in advanced cases, lymph nodes, liver, spleen and central nervous system (CNS) may be involved.  Although the cause of leukemia is not known in most patients, epidemiologic evidence suggests that genetics and environmental factors may play a role in its development. (Franeil et al, 2001)

In the untreated patient, the blood and marrow blast count rises while the granulocyte and platelet count falls accordingly.  The most common symptoms and physical findings result from anemia, thrombocytopenia, and neutropenia and include pallor and fatigue, anorexia, petechiae, purpura, bleeding, and infection.  Treatment must begin as soon as possible to prevent infection and hemorrhage.  It has been proposed that successful treatment of ALL involves the control of bone marrow and systemic disease; it frequently includes the use of systemic combination chemotherapy and CNS preventive therapy.  The 4 phases of ALL treatment are as follows: (i) remission induction, (ii) consolidation, (iii) maintenance therapy, and (iv) CNS prophylaxis.  The standard remission induction protocol for ALL is the combination of vincristine, prednisone, and an anthracycline.  Some regimens also add other drugs, such as asparaginase or cyclophosphamide.  The combination of cranial irradiation and intrathecal methotrexate, high-dose systemic methotrexate and intrathecal methotrexate, or intrathecal chemotherapy alone is commonly used for CNS prophylaxis (Scheinber et al, 2001).  The average length of treatment of ALL varies between 1.5 and 3 years in the effort to eradicate the leukemic cell population.  The greatest number of relapses occurs in the first year after discontinuing chemotherapy (NCI, 2002).

Many studies have attempted to identify ALL patients at high-risk for relapse.  The curability of ALL is related to those prognostic factors identified by the peer-reviewed medical literature as follows: (Martin and Gajewski, 2001; NCI, 2002)

1. Age.  In childhood ALL, conventional chemotherapy has been reported to achieve complete remission rates of about 95 % and 80 % can expect to survive 5 years.  The remission and long-term survival rates are reported to decline with age.  Several studies have correlated age over 35 or 50 years with shorter remission duration and decreased survival rates.  It has been reported that approximately 80 to 94 % of adult patients can expect to achieve complete remission rates after conventional chemotherapy and 20 to 40 % can expect to survive 2 years (NCI, 2002; Martin and Gajewski, 2001).

2. Immunologic subtype.  Based on cell surface antigens, ALLs are subdivided according to the corresponding lymphocyte precursor.  About 80 % of cases of ALL in children and adults fall into the category of early pre-B cell ALL. Pre-B cell ALL, which lacks the expression of CD10 (also known as the common ALL antigen that is expressed in both B and T cell subtypes), is associated with chromosomal abnormalities, high white count and a poor prognosis.  Mature B cell ALL is histologically identical to Burkitt's lymphoma, the only distinction being that ALL involves primarily the bone marrow, while Burkitt's lymphoma involves primarily the lymph nodes.  While these entities are recognized as aggressive, their prognosis has improved recently with multi-agent chemotherapeutic regimens.  T-cell ALL is seen in about 10 to 15 % of ALL cases in children and adults.  T cell ALL is associated with male sex, older age, high white counts, CNS involvement and a mediastinal mass.  The lack of expression of CD10 in T-cell ALL is also associated with a poorer prognosis (NCI, 2002).

3. Cytogenetics.  Certain cytogenetic abnormalities are associated with a poor prognosis.  The presence of the Philadelphia (Ph) chromosome or its molecular counterpart, the bcr-abl oncogene, is associated with a particularly poor prognosis.  The bcr-abl oncogene may be detectable only by pulse-field gel electrophoresis or reverse transcriptase polymerase chain reaction.  The presence of the Ph chromosome is more common in adult ALL, occurring in up to 30 % of cases, and thus may be partially responsible for the overall poorer prognosis in adult patients.  Two other chromosomal abnormalities with poor prognoses are t(4;11), and t(9;22).  In addition, patients with deletion of chromosome 7 or trisomy 8 have been reported to have a lower probability of survival at 5 years compared to patients with a normal karyotype (NCI, 2002).  In contrast, hyperdiploidy (i.e., more than 50 chromosomes) is associated with a favorable prognosis in children, but in adults the effect of this abnormality is less clear (Martin and Gajewski, 2001).

4. Response to induction therapy.  Initial chemotherapy promptly achieves a complete remission in most cases.  However, a prolonged time to reach complete remission is associated with a poor prognosis in all age groups.  Prolonged time to remission can be defined as either requiring 2 cycles of induction therapy, or greater than 4 weeks until complete remission of 5 % leukemic blasts in the bone marrow after 7 to 14 days of induction therapy (Martin and Gajewski, 2001).

5. Elevated white blood cell count.  An elevated white blood count (WBC) is associated with a steadily worse prognosis.  Most studies define a WBC of more than 25,000 to 35,000 cells/uL as high-risk (Martin and Gajewski, 2001).

6. Miscellaneous factors.  Other reported poor prognostic features include CNS involvement or extramedullary leukemia, hepatosplenomegaly, and lymphadenopathy (Martin and Gajewski, 2001).

It is considered an important goal in the management of ALL to identify patients that would preferentially benefit from either conventional chemotherapy or intensified therapy such as bone marrow transplant (BMT).  Patients with low-risk features have an excellent prognosis and are routinely treated with conventional chemotherapy.  In contrast, patients with one or more high-risk features have a poor response to standard chemotherapy.  In a large prospective study from Germany, the reported 5-year disease-free survival for adults with one or more high-risk features ranged from 11 % to 33 % (Hoelzer et al, 1988).  In many studies, these high-risk patients have been considered for BMT.  Bone marrow support can be derived from marrow stem cells harvested from either the bone marrow or peripheral blood from either an allogeneic, autologous or syngeneic (i.e., identical twin) donor.

Retrospective studies evaluating allogeneic transplantation for patients with ALL in first complete remission report higher treatment-related mortality and decreased disease relapse (Horowitz et al, 1991; Oh et al, 1998).  These effects appear to offset one another and to counteract any benefit from allogeneic transplantation in first complete remission (Martin and Gajewski, 2001).  In the largest prospective (n = 257) multi-center, randomized controlled trial to date (Sebban et al, 1994), adults with ALL in remission and who were younger than age 40 years received allogeneic BMT if a sibling donor was available or were randomly assigned to either ongoing chemotherapy or autologous BMT.  There was no advantage to allogeneic BMT for the group of patients with standard-risk ALL.  There was significant survival benefit, however, for patients with high-risk ALL (44 % versus 20 %).  The long-term survival of patients who received chemotherapy and autologous transplant was identical (NCI, 2002).

The International Bone Marrow Transplant Registry (IBMTR) reported a 38 % disease-free survival in Ph-positive patients in first complete remission receiving human leukocyte antigen (HLA)-matched transplants from sibling donors.  These results were similar to those of patients who were refractory to initial induction therapy or who were in second or subsequent remission (Barrett et al, 1992).  The Seattle Transplant Group has reported a 49 % disease-free survival at 2 years among 18 Ph-positive patients who received matched unrelated donor transplants (Sierra et al, 1997).

The majority of published studies consider allogeneic BMT a reasonable consideration for all high-risk (e.g., Ph-positive ALL, t(9;22), t(4;11), failure to respond to induction therapy, B-cell lineage with a white blood cell count greater than 30,000/uL) patients of suitable age (Martin and Gajewski, 2001).

Studies report higher relapse rates in recipients of autologous compared with allogeneic transplant (Attal et al, 1995; Blaise et al, 1990 and 1997).  Two likely explanations are that (i) allogeneic transplantation is associated with a graft-versus-leukemia effect, and (ii) contamination of the autologous marrow with residual leukemia cells may result in increased disease relapse rates and decrease survival status after autologous BMT (Martin and Gajewski, 2001).  The largest prospective study performed to date has found that Ph+ ALL patients receiving allogeneic BMT in first complete remission have better event-free survival and overall survival than similar patients receiving autologous BMT (Goldstone et al, 2001).  Therefore, the role of autologous BMT for the treatment of ALL remains uncertain.

Gaynon et al (2006) compared conventional sibling bone marrow transplantation (CBMT), BMT with alternative donor (ABMT), and chemotherapy (CT) for children with ALL and an early first marrow relapse.  After informed consent, 214 patients with ALL and early marrow relapse began multi-agent induction therapy.  A total of 163 patients with fewer than 25 % marrow blasts and count recovery at the end of induction (second remission [CR2]) were allocated by donor availability; and 50 patients with sibling donors were allocated to CBMT.  Seventy-two patients were randomly allocated between ABMT and CT while 41 patients refused allocation.  Overall, 3-year event free survival from entry is 19 % +/- 3 %.  Thirty-two of 50 CBMT patients (64 %) and 19 of 37 ABMT patients (51 %) underwent transplantation in CR2 with 3-year disease-free survival (DFS) of 42 % +/- 7 % and 29 % +/- 7 %.  The 3-year DFS is 29 % +/- 7 %, 21 % +/- 7 %, and 27 % +/- 8 % for patients allocated to CBMT, ABMT, and CT, respectively.  Contrary to protocol, 12 of 35 patients allocated to CT underwent BMT in CR2.  Of these, 5 patients died after BMT and 5 patients relapsed.  The authors concluded that over 50 % of patients died, failed re-induction, or relapsed again before 3 months after CR2 (median time to BMT).  Intent-to-treat pair-wise comparison of ABMT with CT, CT with CBMT, and CBMT with ABMT yields hazards of 1.2, 1.1, and 0.8 with p values of 0.56, 0.80, and 0.36, respectively.  Outcomes remain similar and poor for children with ALL and early marrow relapse.  Bone marrow transplantation is not a complete answer to the challenge of ALL and early marrow relapse.

Goldstone and colleagues (2008) prospectively evaluated the role of allogeneic transplantation for adults with ALL and compared autologous transplantation with standard chemotherapy.  Patients received 2 phases of induction and, if in remission, were assigned to allogeneic transplantation if they had a compatible sibling donor.  Other patients were randomized to chemotherapy for 2.5 years versus an autologous transplantation.  A donor versus no-donor analysis showed that Philadelphia chromosome-negative patients with a donor had a 5-year improved overall survival (OS), 53 % versus 45 % (p = 0.01), and the relapse rate was significantly lower (p < or = 0.001).  The survival difference was significant in standard-risk patients, but not in high-risk patients with a high non-relapse mortality rate in the high-risk donor group.  Patients randomized to chemotherapy had a higher 5-year OS (46 %) than those randomized to autologous transplantation (37 %; p = 0.03).  Matched related allogeneic transplantations for ALL in first complete remission provide the most potent anti-leukemic therapy and considerable survival benefit for standard-risk patients.  However, the transplantation-related mortality for high-risk older patients was unacceptably high and abrogated the reduction in relapse risk.  There is no evidence that a single autologous transplantation can replace consolidation/maintenance in any risk group.

A technology assessment of stem cell transplantation for ALL (IQWiG, 2007) found indirect evidence of prolonged survival from reduced-intensity stem cell transplantation in patients with refractory ALL.  The report found, however, that these results are limited by the small number of evaluable patients in clinical studies.  The report stated that the relevance of the type of donor remains unclear.  The report found no reliable evidence for superiority of non-myeloablative allogeneic stem cell transplantation (compared to myeloablative allogeneic stem cell transplant), or evidence of benefit from in-vitro manipulation of the graft in allogeneic or autologous stem cell transplantation (compared with transplantation without manipulation of the graft).  The report also found no additional benefit over chemotherapy of non-myeloablative therapy or autologous transplantation; however, available studies were not designed to evaluate non-inferiority (equivalence) of these types of stem cell transplantation to chemotherapy.  The report found, in patients with ALL and their subgroups, no evidence from direct comparative studies of a benefit over chemotherapy of allogeneic stem cell transplantation with an unrelated donor.  However, the available literature shows potential benefit, but also harm, from allogeneic stem cell transplantation with an unrelated donor versus chemotherapy for patients with ALL.

Acute Myelogenous Leukemia:

Acute myelogenous leukemia (AML), also known as acute myeloid leukemia and acute non-lymphocytic leukemia, is a clonal disease characterized by the proliferation and accumulation of myeloid progenitor cells in the bone marrow, which ultimately leads to hematopoietic failure.  The incidence of AML increases with age, and older patients typically have worse treatment outcomes than do younger patients.  In patients with AML, there is an accumulation of leukemic blasts or immature forms in the bone marrow, peripheral blood, and other tissues, with a variable reduction in the production of normal platelets, mature red blood cells, and non-lymphocytic white blood cells (granulocytes, monocytes).  The increased production of malignant cells, along with reduction in these mature elements, result in a variety of systemic consequences including anemia, bleeding, as well as increased risks of infection.  The prognosis is poor for the majority of AML patients, based on age and/or adverse biologic features.  Standard therapy for AML is highly toxic and poorly tolerated, especially in the elderly, for whom few useful therapies exist.  Allogeneic hematopoietic stem cell transplantation (allo-HSCT) is an important option for patients with high-risk AML during first remission (CR1), as well as for any patient in second or subsequent remission.  Use of reduced intensity conditioning transplantations (mini-allograft) has made allo-HSCT for a wider group of individuals (Stone et al, 2004; Stone, 2007; Robak and Wierzbowska, 2009).

Wahlin et al (2009) noted allo-HSCT from an HLA-identical sibling is recommended in standard and poor-risk patients, whereas unrelated donor transplant was reserved for poor-risk patients.  Autologous hematopoietic stem cell transplantation (auto-HSCT) is optional for standard or poor-risk patients who are ineligible for allo-HSCT.  Jones and Copelan (2009) stated that allo-HSCT provides the most powerful anti-leukemic effect in the treatment of AML.  Due to its significant morbidity and mortality, it should be used in CR1 patients whose relapse risk is substantial.  Reduced intensity transplantation is safer and extends the application of early transplantation to older patients and those with co-morbidities.  In patients with advanced disease, allo-HSCT provides a lower chance for cure, but is often the only curative treatment available.

Edenfield and Gore (1999) noted that allogeneic bone marrow transplantation (allo-BMT) as well as autologous bone marrow transplantation (auto-BMT) have become standard approaches for the management of adults with AML.  The indications for transplantation remain controversial as parallel improvements in intensive chemotherapy have resulted in excellent outcomes for many patients.  Allo-BMT is the therapy of choice for patients who fail to respond to induction chemotherapy.  For those patients in CR1, a policy of intensive post-remission chemotherapy with transplantation upon relapse appears to be optimal.  There are no data to support transplantation in CR1, allogeneic or autologous, for those patients with leukemia characterized by favorable cytogenetic abnormalities (i.e., core-binding factor type or t[15;17]), as these patients do well with non-myeloablative strategies.  Patients with relapsed disease appear to be best served with allogeneic transplantation from a HLA-matched sibling or one-antigen-mismatched family member, whereas for those patients lacking a related donor, unrelated donor allo-BMT or auto-BMT provides similar long-term overall survival.

Nathan and colleagues (2004) performed a meta-analysis to compare the effectiveness of auto-BMT with that of non-myeloablative chemotherapy alone (or no further treatment) in adults with AML.  Eligible studies were identified by searching electronic databases and by examining the reference lists of relevant studies and review articles.  Eligible studies were those that prospectively enrolled adults with AML and randomly assigned patients who were in CR1 and who did not have a matched sibling donor to one of the 2 consolidation therapies.  Two reviewers independently assessed all studies for relevance and validity.  They used a fixed-effects model to calculate the ratio of probabilities for DFS and OS at 48 months or at the nearest recorded assessment point for each study and for all studies combined.  All statistical tests were 2-sided.  These investigators identified 587 potentially relevant studies, 36 of which were retrieved for detailed evaluation.  In the 6 studies eligible for this meta-analysis, a total of 1,044 patients were randomly assigned to receive auto-BMT or non-myeloablative chemotherapy (5 studies) or auto-BMT or no further treatment (1 study).  Compared with patients who received chemotherapy or no further treatment, patients who received auto-BMT had a better DFS (ratio of disease-free survival probabilities = 1.24, 95 % confidence interval [CI]: 1.06 to 1.44; p = 0.006) but a similar overall survival (ratio of overall survival probabilities = 1.01, 95 % CI: 0.89 to 1.15; p = 0.86).  The authors concluded that these findings do not support the routine use of auto-BMT in adult AML patients in CR1.

Yanada et al (2005) stated that the effectiveness of allo-HSCT from a HLA-identical sibling donor remains controversial for patients with AML in CR1.  Because the karyotype identified at diagnosis is the most relevant prognostic factor for AML, it should be possible to assess the effectiveness more accurately on the basis of cytogenetic risk.  These researchers carried out a meta-analysis of 5 studies, which employed both natural randomization based on donor availability and intention-to-treat analysis, with OS as an outcome of interest.  Meta-regression analysis was then performed to identify the effectiveness for patients stratified into the favorable, intermediate, and poor cytogenetic risk groups.  For the entire cohort, there was a statistically significant advantage with allo-HSCT in terms of OS with a summary hazard ratio (HR) of 1.15 (95 % CI: 1.01 to 1.32, p = 0.037) for the random-effect model.  Meta-regression analysis showed a significant coefficient of +0.24 for the poor cytogenetic risk group, and -0.25 for the favorable cytogenetic risk group, indicating that the benefit of allo-HSCT was further increased for the former, and lost for the latter.  The coefficient for the intermediate cytogenetic risk group was +0.09, and was not statistically significant.  The authors concluded that these findings suggested that the effectiveness of allo-HSCT for patients with AML in CR1 depended on cytogenetic risk.  The beneficial effect of allo-HSCT was yielded for the poor-risk group, and probably for the intermediate-risk groups, but was absent for the favorable-risk group.

The Dutch-Belgian Hemato-Oncology Cooperative Group and the Swiss Group for Clinical Cancer Research (HOVON-SAKK) collaborative study group (Cornelissen et al, 2007) evaluated outcome of patients with AML in CR1 entered in 3 consecutive studies according to a donor versus no-donor comparison.  Between 1987 and 2004, a total of 2,287 patients were entered in these studies of whom 1,032 patients (45 %) without FAB M3 or t(15;17) were in CR1 after 2 cycles of chemotherapy, received consolidation treatment, and were younger than 55 years of age and therefore eligible for allo-HSCT.  An HLA-identical sibling donor was available for 326 patients (32 %), whereas 599 patients (58 %) lacked such a donor, and information was not available in 107 patients.  Compliance with allo-HSCT was 82 % (268 of 326).  Cumulative incidences of relapse were, respectively, 32 % versus 59 % for patients with versus those without a donor (p < 0.001).  Despite more treatment-related mortality (TRM) in the donor group (21 % versus 4 %, p < 0.001), DFS appeared significantly better in the donor group (48 % +/- 3 % versus 37 % +/- 2 % in the no-donor group, p < 0.001).  Following risk-group analysis, DFS was significantly better for patients with a donor and an intermediate- (p = 0.01) or poor-risk profile (p = 0.003) and also better in patients younger than 40 years of age (p <0 .001).  These investigators evaluated their findings and those of the previous MRC, BGMT, and EORTC studies in a meta-analysis, which revealed a significant benefit of 12 % in OS by donor availability for all patients with AML in CR1 without a favorable cytogenetic profile.

Craddock (2008) stated that allo-HSCT represents the most active form of anti-leukemic therapy in AML.  Advances in transplant technology and supportive care have resulted in improved outcomes in patients allografted using a myeloablative conditioning regimen.  At the same time the use of reduced-intensity conditioning regimens has allowed an immunologically mediated graft-versus-leukaemia effect to be exploited in older patients who were previously ineligible for transplantation on the grounds of age or co-morbidity.  This coupled with the increased availability of alternative stem cell sources, in the form of either unrelated or cord blood donations, has established allogeneic transplantation as a key therapeutic strategy in the treatment of both younger and older adults with AML.

Koreth and colleagues (2009) stated that the optimal treatment of AML in CR1 is uncertain.  Current consensus, based on cytogenetic risk, recommends myeloablative allo-HSCT for poor-risk but not for good-risk AML.  Allogeneic-HSCT, autologous transplantation, and consolidation chemotherapy are considered of equivalent benefit for intermediate-risk AML.  These researchers quantified relapse-free survival (RFS) and OS benefit of allo-HSCT for AML in CR1 overall and also for good-, intermediate-, and poor-risk AML.  Systematic review and meta-analysis of prospective trials evaluating allo-HSCT versus non-allo-HSCT therapies for AML in CR1 were carried out.  These researchers identified 1,712 articles.  Prospective trials assigning adult patients with AML in CR1 to undergo allo-HSCT versus non-allo-HSCT treatment(s) based on donor availability and trials reporting RFS and/or OS outcomes on an intention-to-treat, donor versus no-donor basis were identified.  Two reviewers independently extracted study characteristics, interventions, and outcomes.  Hazard ratios with 95 % CIs were determined.  Overall, 24 trials and 6,007 patients were analyzed (5,951 patients in RFS analyses and 5,606 patients in OS analyses); 3,638 patients were analyzed by cytogenetic risk (547, 2,499, and 592 with good-risk, intermediate-risk, and poor-risk AML, respectively).  Inter-study heterogeneity was non-significant.  Fixed-effects meta-analysis was performed.  Compared with non-allo-HSCT, the HR of relapse or death with allo-HSCT for AML in CR1 was 0.80 (95 % CI: 0.74 to 0.86).  Significant RFS benefit of allo-HSCT was documented for poor-risk (HR, 0.69; 95 % CI: 0.57 to 0.84) and intermediate-risk AML (HR, 0.76; 95 % CI: 0.68 to 0.85) but not for good-risk AML (HR, 1.06; 95 % CI: 0.80 to 1.42).  The HR of death with alloSCT for AML in CR1 was 0.90 (95 % CI: 0.82 to 0.97).  Significant OS benefit with allo-HSCT was documented for poor-risk (HR, 0.73; 95 % CI: 0.59 to 0.90) and intermediate-risk AML (HR, 0.83; 95 % CI: 0.74 to 0.93) but not for good-risk AML (HR, 1.07; 95 % CI: 0.83 to 1.38).  The authors concluded that compared with non-allo-HSCT therapies, allo-HSCT has significant RFS and OS benefit for intermediate- and poor-risk AML but not for good-risk AML in CR1.

The Italian Society of Hematology and 2 affiliated societies (the Italian Society of Experimental Hematology and the Italian Group for Bone Marrow Transplantation) commissioned project to an expert panel aimed at developing clinical practice guidelines for the treatment of AML (Morra et al, 2009).  After systematic comprehensive literature review, the expert panel formulated recommendations for the management of primary AML (with the exception of acute promyelocytic leukemia) and graded them according to the supporting evidence.  When evidence was lacking, consensus-based statements have been added.  First-line therapy for all newly diagnosed patients eligible for intensive treatment should include 1 cycle of induction with standard dose cytarabine and an anthracycline.  After achieving CR, patients aged less than 60 years should receive consolidation therapy including high-dose cytarabine. Myeloablative allo-HSCT from an HLA-compatible sibling should be performed in CR1: (i) in children with intermediate- to high-risk cytogenetics or who achieved CR1 after the second course of therapy; (ii) in adults less than 40 years with an intermediate-risk; in those aged less than 55 years with either high-risk cytogenetics or who achieved CR1 after the second course of therapy.  Stem cell transplantation from an unrelated donor is recommended in CR1 in adults 30 years old or younger, and in children with very high-risk disease lacking a sibling donor.  Alternative donor stem cell transplantation is an option in high-risk patients without a matched donor who urgently need transplantation.  Patients aged less than 60 years, who either are not candidate for allo-HSCT or lack a donor, are candidates for auto-SCT.

Hartwig and colleagues (2009) noted that for patients with myeloid malignancies who relapse after allo-HSCT, one salvage option is a second allo-HSCT.  These researchers retrospectively analyzed outcomes of the second allo-HSCT in 25 patients who received at least 2 allografts from related/unrelated donors due to relapse of acute AML, myelodysplastic syndrome or myelofibrosis after the first allo-HSCT.  A minority of the AML/myelodysplastic syndrome patients had reached complete hematological remission before the second allo-HSCT (6/25, 24 %).  Reduced conditioning strategies were performed in the majority (n = 23).  Complete remission was achieved in all 21 cases with available data after the second allo-HSCT, but relapse was seen in 11/25 patients (44 %).  After a median follow-up of 18 months (range of 6 to 47), 8/25 patients (32 %) were still alive, and of those, 6 (24 %) were in stable remission.  In 9 cases mortality was associated to relapse, and in 8 cases to transplant-related causes (treatment-related mortality; 8/25, 32 %).  The authors concluded that a second allo-HSCT offers the chance of stable remission for some patients relapsing with a myeloid malignancy after a first allo-HSCT, although high treatment-related mortality and relapse rates remain a problem.

A technology assessment of stem cell transplantation for AML (IQWiG, 2007) found evidence from direct comparative studies demonstrated improved survival with non-myeloablative allogeneic stem cell transplantation with a related donor compared to conventional chemotherapy.  Indirect comparisons suggest improved OS in both non-myeloablative allogeneic stem cell transplantation with an unrelated donor over conventional chemotherapy in persons with refractory AML.  There was also evidence from indirect comparisons suggesting benefit of myeloablative allogeneic stem cell transplantation with an unrelated donor in persons with AML.  The assessment found insufficient evidence to draw conclusions about the comparative efficacy of transplants with related donors versus unrelated donors for AML.  The assessment found a lack of evidence comparing nonmyeloablative versus myeloablative stem cell transplant in AML.  There was also insufficient evidence that in-vitro manipulation of the graft improves outcomes of allogeneic or autologous transplantation in AML.

Chronic Myelo-Monocytic Leukemia and Juvenile Myelo-Monocytic Leukemia:

Chronic myelo-monocytic leukemia (CMML) and juvenile myelo-monocytic leukemia (JMML) are malignancies as a consequence of over-production of monocytes and myelocytes (immature leukocytes) by the bone marrow.  Some of these blood stem cells never become mature leukocytes; and these immature leukocytes are known as blasts.  Over time, the monocytes, myelocytes, and blasts crowd out the reticulocytes and platelets in the bone marrow leading to infection, anemia, or easy bleeding.

Kroger et al (2002) reported the results of 50 allogeneic transplantations from related (n = 43) or unrelated (n = 7) donors, performed for CMML in 43 European centers.  The median age at transplant was 44 years (range of 19 to 61).  Eighteen patients had excess blasts ranging from 5 % to 30 % at the time of transplantation.  Two graft failures were observed (4 %).  Neutrophil (greater than 0.5 x 109/L) and platelet engraftment (greater than 50 x 109/L) was reached after a median of 17 days (range of 11 to 38) and 27 days (range of 11 to 48), respectively.  Acute graft-versus-host disease (GVHD grade II to IV was seen in 35 % of patients, while 20 % developed severe-acute GVHD grade III/IV.  Twenty-six patients (52 %) died of treatment-related causes.  After a median follow-up of 40 months (range of 11 to 110), the 5-year-estimated OS was 21 % (95 % CI: 15 to 27 %) and the 5-year estimated DFS was 18 % (95 % CI: 13 to 23 %).  Earlier transplantation in the course of disease, male donor, use of unmanipulated grafts, allogeneic transplantation and occurrence of acute GVHD favored better DFS, but did not reach statistical significance.  The 5-year estimated probability of relapse was 49 %.  The data showed a trend for a lower relapse probability of acute GVHD grade II to IV (24 % versus 54 %; p = 0.07), and for a higher relapse rate in patients with T cell-depleted grafts (62 % versus 45 %), suggesting a "graft-versus-CMML effect".

Kerbauy and co-workers (2005) evaluated the outcomes of allogeneic HSCT in 43 patients with CMML.  Patients were classified according to the French-American-British and World Health Organization classifications, as well as the International Prognostic Scoring System and the M.D.  Anderson prognostic score.  Co-morbidity scores were assessed by using an HSCT-specific co-morbidity index.  Patients were aged 1 to 66 years (median of 48 years).  Twenty-one patients received transplants from related donors (18 HLA-identical siblings and 3 HLA-non-identical family members), and 22 received transplants from unrelated donors (18 HLA matched and 4 HLA non-identical).  Several busulfan or total body irradiation (TBI)-based conditioning regimens were used.  Sustained engraftment was achieved in 41 patients.  Eighteen are alive at 1.9 to 14.1 years, for an estimated relapse-free survival of 41 % at 4 years.  Ten patients have relapsed, thus leading to a cumulative incidence of 23 % at 4 years.  Risk category by International Prognostic Scoring System, World Health Organization, M.D. Anderson prognostic score, or proliferative/dysplastic status had no statistically significant association with outcomes.  However, patients with higher co-morbidity scores had worse OS than patients with lower scores (p = 0.01).  There was a trend for a higher relapse incidence among patients at higher risk by the M.D. Anderson prognostic score.  These findings suggested that patients with few or no co-morbidities and those who undergo transplantation earlier in the disease course have the highest probability of successful outcome after allogeneic HSCT.

Elliott and colleagues (2006) reviewed their experience of allogeneic HSCT and donor lymphocyte infusions (DLI) for adults with CMML.  A total of 17 consecutive adults underwent allogeneic HSCT from related (n = 14) or unrelated (n = 3) donors.  Median age was 50 years (range of 26 to 60).  Seven patients (41 %) demonstrated relapse or persistent disease at a median of 6 months (range of 3 to 55.5).  Five patients underwent DLI for morphologic relapse and 1 for mixed donor chimerism.  Two patients achieved durable complete remissions of 15 months each.  The overall transplant-related mortality was 41 % (n = 7).  With a median follow-up of 34.5 months, 3 patients (18 %) currently remain alive and in continuous CR.  These findings demonstrated a graft-versus-leukemia effect in CMML, both for allogeneicHSCT and for DLI.  However, consistent with reported experience of others, overall outcomes remain less than optimal and unpredictable.

Krishnamurthy et al (2010) reported on single-center results of 18 patients with CMML who have undergone allogeneic HSCT.  The median age of patients was 54 years.  Seven patients had AML, which had transformed from CMML.  Overall, 11 patients received stem cells from an unrelated donor.  A total of 15 patients received a T-cell-depleted fludarabine/BU-based reduced-intensity conditioning HSCT.  The actuarial 3-year OS, non-relapse mortality (NRM) and relapse incidence for the cohort was 31 +/- 11 %, 31 +/- 14 % and 47 +/- 13 %, respectively.  Patients with favorable cytogenetics had a 3-year DFS of 65 +/- 17 %, whereas none of the 7 patients with intermediate-risk or poor-risk cytogenetics survived beyond 2 years (p < 0.01).  No patients with favorable risk cytogenetics died from NRM causes, while the 2-year NRM for the intermediate-risk/poor-risk cytogenetics subgroup was 71 +/- 22 % (p < 0.02).  In terms of disease status at transplantation, patients who had less than 5 % BM blasts had a 3-year DFS of 46.9 +/- 19 % compared with those with greater than 5% blasts at the time of transplantation (i.e., 20.0 +/- 13 %).  Recipient age, type of conditioning regimen or stem cell dose did not have a significant impact on overall outcomes.  These findings supported existing evidence that allogeneic HSCT is a feasible therapeutic option for CMML, with the ability to attain long-term remission among patient subgroups.

Cheng et al (2012) performed a literature search and reviewed available data for adult CMML patients undergoing HSCT.  The dearth of data that span 2 decades with changing transplant practices prohibited these researchers from performing a formal meta-analysis.  However, these investigators elected to present the current status of HSCT in adult CMML patients.  The authors concluded that carefully selected CMML patients may have the most benefit from this curative approach.

Woods and associates (2002) reported the first large prospective study of children with myelodysplastic syndrome (MDS) and JMML treated in a uniform fashion on Children's Cancer Group protocol 2891.  A total of 90 with JMML, various forms of MDS, or AML with antecedent MDS were treated with a 5-drug induction regimen (standard or intensive timing).  Patients achieving remission were allocated to allogeneic BMT if a matched family donor was available.  All other patients were randomized between autologous BMT and aggressive non-myeloablative chemotherapy.  Results were compared with patients with de novo AML.  Patients with JMML and refractory anemia (RA) or RA-excess blasts (RAEB) exhibited high induction failure rates and overall remission of 58 % and 48 %, respectively.  Remission rates for patients with RAEB in transformation (RAEB-T) (69 %) or antecedent MDS (81 %) were similar to de novo AML (77 %).  Actuarial survival rates at 6 years were as follows: JMML, 31 % +/- 26 %; RA and RAEB, 29 % +/- 16 %; RAEB-T, 30 % +/- 18 %; antecedent MDS, 50 % +/- 25 %; and de novo AML, 45 % +/- 3 %.  For patients achieving remission, long-term survivors were found in those receiving either allogeneic BMT or chemotherapy.  The presence of monosomy 7 had no additional adverse effect on MDS and JMML.  The authors concluded that childhood subtypes of MDS and JMML represent distinct entities with distinct clinical outcomes.  Children with a history of MDS who present with AML do well with AML-type therapy.  Patients with RA or RAEB respond poorly to AML induction therapy.  The optimum treatment for JMML remains unknown.

Baker and colleagues (2004) examined the effectiveness of allogeneic BMT without a TBI conditioning regimen in children with JMML.  A total of 8 patients with JMML (n = 6) or monosomy 7 (n = 2) underwent BMT at a median age of 20 months.  Donor source included fully matched related (n = 3), mis-matched related (n = 2), or fully matched unrelated (n = 3).  The conditioning regimen included busulfan (BU), cyclophosphamide (CY), and etoposide (VP16) (melphalan was substituted for VP16 in 1 patient).  The first patient in the series underwent TBI.  Graft-versus-host disease prophylaxis was with cyclosporin and methotrexate and in-vivo T-cell depletion (Campath 1 g) for mis-matched and unrelated transplants.  Seven and 2 patients, respectively, received chemotherapy and splenectomy before BMT.  At a median follow-up of 48 months after BMT, 5 patients remained in remission.  The OS rate was 63 % at 5 years.  All deaths occurred in patients with refractory disease at the time of BMT.  Allogeneic BMT without TBI appears to be effective therapy for JMML and avoids some of the potential late sequelae of TBI in pre-school children.

Korthof et al (2005) noted that JMML is a childhood leukemia for which allogeneic BMT is the only curative therapy.  These investigators performed 26 BMTs in 23 children (age of 0.5 to 12.7 years).  Conditioning was CY/TBI based (1980 to 1996, n = 14) or BU/CY/melphalan based (1996 to 2001, n = 9).  Donors were HLA-identical siblings (n = 11), unrelated volunteers (n = 9) or mis-matched family members (n = 3).  A total of 10 patients survive in CR (median follow-up of 6.8 years, range of 3.1 to 22.2 years).  Relapse or persistent disease was observed in 8 and 2 patients, respectively.  Nine of these patients died, 1 achieved a second remission following acute non-lymphatic leukemia chemotherapy (duration to date 5.3 years).  Transplant-related mortality occurred in 4 patients.  Overall survival at 5 and 10 years was 43.5 %.  Using T-cell-depleted, one-antigen mis-matched unrelated donors was the only significant adverse factor associated with relapse in multi-variate analysis (p = 0.039, HR 4.9).  Together with a trend towards less relapse in patients with GVHD and in patients transplanted with matched unrelated donors, this suggested a graft-versus-leukemia effect of allogeneic BMT in JMML.

Locatelli and associates (2005) stated that allogeneic HSCT is the only proven curative therapy for JMML.  The European Working Group on Childhood MDS (EWOG-MDS) and the European Blood and Marrow Transplantation (EBMT) Group reported the outcome of 100 children (67 boys and 33 girls) with JMML given unmanipulated HSCT after a preparative regimen including busulfan, cyclophosphamide, and melphalan.  Forty-eight and 52 children received transplants from an HLA-identical relative or an unrelated donor (UD), respectively.  The source of hematopoietic stem cells was bone marrow, peripheral blood, and cord blood in 79, 14, and 7 children, respectively.  Splenectomy had been performed before HSCT in 24 children.  The 5-year cumulative incidence of transplantation-related mortality and leukemia recurrence was 13 % and 35 %, respectively.  Age older than 4 years predicted an increased risk of disease recurrence.  The 5-year probability of event-free survival for children given HSCT from either a relative or a UD was 55 % and 49 %, respectively (p = NS), with median observation time of patients alive being 40 months (range of 6 to 144).  In multi-variate analysis, age older than 4 years and female sex predicted poorer outcome.  Results of this study compare favorably with previously published reports.  Disease recurrence remains the major cause of treatment failure.  Outcome of UD-HSCT recipients is comparable to that of children receiving transplants from an HLA-identical sibling.

Loh (2011) stated that JMML is an aggressive myeloid neoplasm of childhood that is clinically characterized by over-production of monocytic cells that can infiltrate organs, including the spleen, liver, gastrointestinal tract, and lung.  Juvenile myelo-monocytic leukemia is categorized as an overlap myeloproliferative neoplasms/myelodysplastic syndromes by the World Health Organization and also shares some clinical and molecular features with CMML, a similar disease in adults.  While the current standard of care for patients with JMML relies on allogeneic HSCT, relapse is the most frequent cause of treatment failure.

The Lymphoma and Leukemia Society (2009) noted that "Allogeneic stem cell transplantation is the only known curative option for JMML patients.  This treatment has been noted to achieve long-term survival in up to 50 % of patients but relapses are known to occur in up to 30 % to 40 % of patients after transplantation.  Second transplants have been beneficial for some patients".  The Lymphoma and Leukemia Society also noted that "Allogeneic stem cell transplantation has been used to treat and sometimes cure CMML patients".

The National Cancer Institute (2010) noted that "Bone marrow/stem-cell transplantation appears to be the only current treatment that alters the natural history of CMML".  Regarding JMML, the NCI (2010) noted that "No consistently effective therapy is available for JMML .... Bone marrow transplantation seems to offer the best chance for a cure".

The National Comprehensive Cancer Network (2010) clinical practice guidelines in myelodysplastic syndromes noted that "Allogeneic HSCT from an HLA-matched sibling donor is a preferred approach for treating a portion of patients with MDS.  Standard conditioning is used for relatively younger patients, while the approach using non-myeloablative conditioning is preferable in older individuals".