Allogeneic Hematopoietic Cell Transplantation Provides No Benefit for Patients With Hypodiploid Acute Lymphoblastic Leukemia

Michael J. Burke, MD


J Clin Oncol. 2019;37(10):763-764. 

Treating pediatric patients with B-cell acute lymphoblastic leukemia (B-ALL), on the basis of risk stratification, currently provides a long-term survival rate of more than 90%.[1] Risk stratification for ALL includes a number of criteria, one of which is the underlying leukemia biology. The biology of B-ALL can identify patients with favorable outcomes (ie, hyperdiploidy of chromosomes 4 and 10 or translocation of chromosomes 12 and 21 involving genes ETV6 and RUNX1) or those having a worse prognosis (ie, KMT2A rearrangement, intrachromosomal amplification of chromosome 21, severe hypodiploidy). On the basis of the disease biology alone, treatment decisions may consist of chemotherapy only or may indicate the need for more intensive or alternative therapies such as allogeneic hematopoietic cell transplantation (HCT). HCT in first complete remission (CR1) has traditionally been reserved for pediatric patients with very poor prognosis (predicted event-free survival [EFS] < 50%) such as those with primary induction failure (> 25% blasts in the bone marrow at the end of induction therapy) or severe hypodiploidy (< 44 chromosomes or a DNA index < 0.81) given their reportedly poor outcomes.[2–4] However, the question of whether HCT is still indicated for patients with hypodiploidy is raised in the articles by McNeer et al[5] and Pui et al[6] that accompany this editorial.

Nachman and colleagues[2] previously reported a large cohort of pediatric patients with non-Philadelphia chromosome–positive hypodiploid ALL registered on clinical trials across 10 cooperative groups between 1986 and 1996. The 8-year EFS for the entire cohort (n = 130) was 38.5% ± 5.7%. On the basis of modal chromosome number, patients with fewer than 30 chromosomes (n = 46) had an 8-year EFS of 28.3% ± 6.6% compared with 36.9% ± 9.8% for 33 to 39 chromosomes (n = 26), 18.7% ± 15.8% for 40 to 43 chromosomes (n = 8), and 52.2% ± 7.4% for patients with 44 chromosomes (n 5 50). There were too few patients who proceeded to HCT in this analysis to comment on the efficacy of HCT for this high-risk subgroup, but outcomes reported for those with fewer than 44 chromosomes were dismal enough to raise the question of whether to proceed to HCT. Recently the Center for International Blood and Marrow Transplant Research performed an analysis of HCT outcomes for 78 pediatric patients (age 18 years or younger) with hypodiploid ALL (defined as ≤ 45 chromosomes) collected between 1999 and 2010.[7] This report included patients in CR1 and second complete remission or greater (39 patients had 43 or fewer chromosomes, 12 had 44, and 27 had 45) who reported 5-year leukemia-free survival of 51% and overall survival (OS) of 56%. Although these outcomes seem reasonably favorable with HCT and better than those reported by Nachman et al,[2] half the cohort had 44 or 45 chromosomes, conditions that are known to have better survival, and thus the role for HCT in hypodiploid ALL remains unclear.

McNeer et al[5] report on 131 pediatric patients with hypodiploid B-ALL treated on Children's Oncology Group studies between 2003 and 2011 in which 61 patients (45.6%) underwent HCT in CR1.[6] Patients who received HCT had a 5-year EFS of 56.4% ± 7.3% and a 5-year OS of 65.6% ± 6.9%, similar to results in the Center for International Blood and Marrow Transplant Research report. However, these outcomes were not significantly better than outcomes for those who did not receive HCT who had an EFS of 48.8% ± 7.8% (P=.62) and an OS of 53.8%±7.8% (P=.32). When evaluating outcomes on the basis of the presence of end of induction (EOI) minimal residual disease (MRD), those who achieved EOI MRD of less than 0.01% had significantly greater 5-year EFS (63.7% ± 5.8%) and OS (73.2% ± 5.3%) compared with patients who had EOI MRD of 0.01% or more and who had dismal outcomes (EFS, 26.2% ± 8.5%; OS, 27.7% ± 8.9%). Of note, HCT did not improve outcomes for patients who had EOI MRD of 0.01% or more (EFS, P = .67; OS, P = .86) or less than 0.01% (EFS, P = .77; OS, P = .39). Thus, the authors concluded that patients with hypodiploid B-ALL who achieve EOI MRD of less than 0.01% have a reasonably good outcome compared with what has previously been reported and, more importantly, who did not benefit from HCT.

These conclusions were also made by Pui et al[6] who provide the largest analysis to date of children with newly diagnosed hypodiploid ALL, with a total of 306 patients treated by 16 cooperative groups between 1997 and 2013. For this cohort, the 5-year EFS was 55.1% and the 5-year OS was 61.2%, with those who were EOI MRD negative reporting an even greater EFS of 75%. Similar to the Children's Oncology Group findings surrounding HCT reported earlier, Pui et al[6] identified no advantage in either EFS (P = .47) or OS (P = .21) with HCT compared with chemotherapy alone. This finding was independent of MRD, in which HCT offered no advantage over chemotherapy alone whether the EOI MRD was positive or negative (EFS, P = .81; OS, P = .29). The authors concluded that EOI MRD identified patients with hypodiploid ALL who had favorable outcomes warranting chemotherapy only and those with very poor outcomes who remained MRD positive and for whom novel therapies such as PI3K and PI3K/mTOR inhibitors (both of which have demonstrated preclinical efficacy in hypodiploid ALL[8]) should be investigated.

In summary, the articles that accompany this editorial have called into question the role of HCT for pediatric patients with hypodiploid ALL.[5,6] HCT may have been considered the standard treatment approach for these patients, but the reports by McNeer et al[5] and Pui et al[6] using MRD analysis at EOI to identify patients who had relatively good outcomes when they were treated with chemotherapy alone and not benefiting from HCT argue against this. Just as chemotherapy strategies must be tested in high-risk subsets of patients before they can be declared successful and be used as a standard approach to treatment, HCT for highrisk subgroups must also prove beneficial before being adopted as standard practice. Such is the case for pediatric patients with ALL and primary induction failure in whom outcomes seem to be improved with matched, related HCT compared with chemotherapy alone.[9] In the case of severe hypodiploid ALL (< 44 chromosomes), McNeer et al[5] and Pui et al[6] identify no role for HCT for patients who are MRD negative after induction, given their good outcomes (EFS > 70%). They also report no benefit when HCT is used in patients who remain MRD positive and for whom survival remains poor (EFS is approximately 30%-50%). Thus, severe hypodiploidy seems to be a marker for poor response to current therapies particularly when MRD persists at EOI. Whether integrating new agents such as PI3K inhibitors or chimeric antigen receptor T-cell therapy into the treatment for these patients can improve these outcomes is yet to be seen but is clearly worth investigating.

In conclusion, good treatment can sometimes overcome the prognostic importance of poor prognostic factors. Unfortunately, it seems that HCT is not capable of achieving this when it comes to hypodiploid ALL and persistent MRD.