Acute Leukemia

Risk factors and timing of relapse after allogeneic transplantation in pediatric ALL: for whom and when should interventions be tested?

Article metrics


We previously showed that minimal residual disease (MRD) detection pre-hematopoietic cell transplant (HCT) and acute GvHD (aGvHD) independently predicted risk of relapse in pediatric ALL. In this study we further define risk by assessing timing of relapse and the effects of leukemia risk category and post-HCT MRD. By multivariate analysis, pre-HCT MRD <0.1% and aGvHD by day +55 were associated with decreased relapse and improved event-free survival (EFS). Intermediate leukemia risk status predicted decreased relapse, and improved EFS and overall survival (OS). Patients with pre-HCT MRD 0.1% who did not develop aGvHD compared with those with MRD <0.1% who did develop aGvHD had much worse survival (2 years EFS 18% vs 71%; P=0.001, 2 years OS 46 vs 74%; P=0.04). Patients with pre-HCT MRD <0.1% who did not experience aGvHD had higher rates of relapse than those who did develop aGvHD (40% vs 13%; P= 0.008). Post-HCT MRD led to a substantial increase in relapse risk (HR=4.5, P<0.01). Patients at high risk of relapse can be defined after transplant using leukemia risk category, presence of MRD pre or post HCT, and occurrence of aGvHD. An optimal window to initiate intervention to prevent relapse occurs between day +55 and +200 after HCT.


Published classification schemes for childhood acute lymphoblastic leukemia (ALL) describing when allogeneic hematopoietic cell transplantation (HCT) is recommended are based upon response to therapy and genetic characteristics.1, 2 After a first or subsequent relapse, further delineation of risk of failure of therapy has been made based upon timing and site of relapse, immunophenotype and persistence of minimal residual disease (MRD) after reinduction therapy.3, 4, 5, 6 Although high rates of post-HCT relapse have been noted when persistent MRD is present before the HCT preparative regimen, the effect of MRD on established leukemia risk groups has not been explored. In addition, little is known about whether post-HCT events such as GvHD alter predicted outcomes.

Although often debated, several studies7, 8, 9, 10, 11, 12, 13, 14, 15 have suggested a positive effect of GvHD in decreasing relapse after allogeneic HCT for ALL. We confirmed this finding in a prospective phase III of adding sirolimus for GvHD prophylaxis.16, 17, 18, 19, 20

Past attempts to de-escalate GvHD prophylaxis and encourage GvL have sometimes shown success,21 but can lead to significant morbidity associated with GvHD.1 Any child subjected to either immune withdrawal or a potentially dangerous targeted agent early after HCT should be at very high risk of relapse in order to minimize unnecessary toxicity. Because both aGvHD after transplant and MRD pre-HCT and post HCT may significantly effect relapse risk,22, 23 careful study of the interplay between these factors is needed to define high-risk populations where interventions to reduce relapse could be performed. In this study we look at the effect of established leukemia disease risk-classification groups, pre- and post-HCT MRD detection, and the occurrence of GvHD on relapse and survival. We then identify target populations and timeframes post HCT where intervention to prevent relapse would have the greatest impact.

Patients and methods

Protocol Children's Oncology Group (COG) ASCT0431 (PBMTC ONC051) was conducted from 2007 to 2011 and was approved by the NCI Central Institutional Review Board and local Institutional Review Boards as applicable. The trial was open to ages 1–21 with ALL in CR1 and CR2, utilizing related and unrelated bone marrow or cord blood. Three leukemia risk categories of patients were allowed: high-risk CR1 (Ph+ ALL, extreme hypodiploidy (<44 chromosomes), or induction failure (either M3 bone marrow at day +29 or M2 or MRD >1% at day +43)); high-risk CR2 (B-cell bone marrow relapse <36 m from diagnosis, T cell or Ph+ bone marrow relapse at any time, T-cell isolated extramedullary relapse <18 m from diagnosis); and intermediate-risk (IR) CR2 (B-cell bone marrow relapse 36 m from diagnosis, B-cell isolated extramedullary <18 m from diagnosis). Details regarding transplant approach, GvHD grading, and 6-color multi-channel flow cytometry MRD were published previously.22, 24, 25, 26 MRD was measured on bone marrow pre-HCT and at 1, 3 and 8–12 m post HCT, and was performed at a single COG central flow laboratory (Johns Hopkins). Centers were blinded to pre- and post-HCT MRD results, and modification of the protocol based upon local MRD results was not allowed.

Statistical approach

Patients who did not experience an event were censored at the time of last contact. Event free survival (EFS) and overall survival (OS) were estimated by the Kaplan–Meier method.27 The log-rank statistic was used to compare EFS and OS between standard and experimental treatment groups. The hazard ratios, confidence intervals (CIs) and tests of significance associated with patient characteristics were assessed using proportional hazards regression modeling.28 Time-dependent coding of GvHD occurrence and post-HCT MRD status was used to appropriately account for timing of these factors and number of post-HCT MRD assessments was included in all regression models evaluating post-HCT MRD status. Cause-specific outcome probabilities of relapse or transplant-related mortality (TRM) over time were estimated using non-parametric methods of Aalen and Johannsen.29

To provide graphical comparisons of outcome probabilities over time by aGvHD status, the landmark analysis curve28 approach was used in which outcome probabilities were computed from day +55 forward with aGvHD status determined at day +55. Day +55 was chosen because by then 93% of those who developed aGvHD had experienced peak grade and determination of risk classification at this time would allow sufficient time for implementing further treatment strategies. Landmark analysis curves starting anywhere between days +50 and 80 were essentially equivalent. Eight patients had died of TRM by day +55 (5.5%), with two of the eight having developed aGvHD before death. Two relapses had occurred, meaning 10 out of the 143 patients (7%) did not contribute to the landmark analysis. The use of landmark curves avoids bias that would occur comparing Kaplan–Meier estimates starting at HCT based on aGvHD status, a condition that occurs after HCT.28 P-values given in landmark figures are from the Cox regression analysis with aGvHD as a time-dependent covariate and uses all the available data. HRs for post-HCT MRD and relapse were computed using Cox regression with MRD status and the number of MRD assessments included as time-dependent covariates.

Table 1 shows the patient population involved in the analysis. This phase III trial compared tacrolimus/methotrexate with sirolimus/tacrolimus/methotrexate for GvHD prophylaxis. The populations in both arms were balanced for key factors associated with outcome.22 Although there was less grade I and II aGvHD noted on the sirolimus arm, rates of relapse, survival and TRM were similar between the two arms. In addition, the effect of aGvHD in decreasing relapse was similar between the treatment arms (P=0.65).22 Rates of MRD and the effect of MRD on outcomes were also noted to be similar in both arms (P=0.21). With this in mind, we combined both arms for this analysis of the effect of GvHD and MRD. The main data analysis of risk groups was performed on patients with pre-HCT MRD data present. Twenty seven percent of patients did not provide pre-HCT bone marrow MRD samples, most often because they had already undergone pre-HCT bone marrow sampling before consent and refused a second procedure. Table 1 shows no differences betwee the population tested for MRD pre-HCT and the full cohort.

Table 1 Patient characteristics (percent) in all patients and in patients with pre-HCT MRD assessment


Individual and combined effects of aGvHD and pre-HCT MRD on relapse and survival

As part of the primary manuscript reporting outcomes of COG ASCT0431/PBMTC ONC051, a multivariate analysis-defined independent benefits for relapse and survival from (1) low pre-HCT MRD (absent or <0.1% by flow cytometry) and (2) the occurrence of grades I–III aGvHD (grade IV aGvHD decreased relapse but decreased survival due to TRM).22 In this analysis we sought to determine in more detail how these clinical factors combine to identify risk of poor outcomes. We chose to perform a landmark analysis of patients alive with or without aGvHD at day +55 as outlined in the methods section. Figure 1a shows a dramatic difference in risk of relapse between patients with pre-HCT MRD 0.1% who did not develop aGvHD compared with those with MRD <0.1% who did develop aGvHD (CI of relapse 73% vs 13%; P=<0.0001). It is also notable that there is a significant difference in relapse between those previously considered to be ‘good risk’ patients (pre-HCT MRD <0.1%) who did and did not experience aGvHD (13% vs 40%; P= 0.008).

Figure 1

Probability of (a) relapse (b) EFS and (c) OS in ALL patients from all risk categories who had pre-HCT flow MRD measured and lived to day +55 after transplant (n=105). Patients listed as MRD+ had 0.1% measured by flow cytometry in bone marrow within 2 weeks initiating their preparative regimen. Patients listed as experiencing aGvHD had onset of any grade of aGvHD before day +55. MRD <0.1% and no aGvHD n=51, MRD <0.1% and yes aGvHD n=37, MRD 0.1% and no aGvHD n=12, MRD 0.1% and yes aGvHD n=5.

The changes in relapse risk based on these two clinical factors translated into similar changes in EFS. Figure 1b shows a marked difference in survival between patients who had pre-HCT MRD 0.1% who did not develop aGvHD compared with those with MRD <0.1% who did develop aGvHD (2-year EFS 18% vs 71%; P=0.001, 2-year OS 46% vs 74%; P=0.04), and a trend toward increased EFS in the patients who were MRD <0.1% who did or did not develop aGvHD (71% vs 51%; P=0.07). There were only five patients who had pre-HCT MRD 0.1% who did develop aGvHD so possibly due to low numbers, no statistical differences in relapse risk and EFS between this group and any of the others were found.

The combined effect of leukemia risk group, aGvHD and pre-HCT MRD on relapse and survival

With the noted effects of aGvHD and pre-HCT MRD on outcomes, we asked whether current risk classification groups remain relevant, or if risk could be better predicted using these factors alone. COG ALL risk classifications (high-risk [HR] CR1, HR CR2 and IR CR2) are based upon time to relapse, immune phenotype, cytogenetics and site of relapse. Definitions of these leukemia risk groups are detailed in the methods section and are very similar to European classification approaches.4 Hazard ratios estimated in multivariate analyses looking at leukemia risk group, aGvHD and pre-HCT MRD with aGvHD as a time-dependent covariate, are shown in Table 2. Of note, all three of these factors: leukemia risk group, the presence of high levels of MRD pre-HCT and the occurrence of grade I–III aGvHD after HCT, independently influenced outcome. Within the risk groups, HR CR1 and HR CR2 ALL were very similar in outcome, but patients with IR CR2 experienced less relapse and better EFS and OS independent of aGvHD and MRD effects. With this in mind, the HR CR1/CR2 groups were combined for further analyses.

Table 2 Hazard ratios (95% CIs) for leukemia risk group, pre-HCT MRD and aGvHD in the multivariate model

The first three lines of Table 3 show the 2-year risk of relapse based on the key pre-HCT factors, leukemia risk group and pre-HCT MRD. In HR CR1 and CR2 patients, the EFS difference in children with low MRD (<0.1%) compared with high MRD (0.1%) was more than 30% (47% vs 15%, P<0.0001). Intermediate leukemia-risk patients with pre-HCT MRD <0.1% had an excellent prognosis (>80% EFS). The lower portion of Table 3 shows the modifying effect of aGvHD occurring by day +55 on patient risk of experiencing an event. Patients with grade I–III aGvHD had an ~20% higher 2-year EFS probability than patients with no aGvHD in each leukemia risk category and MRD status (Table 2, P=0.02).

Table 3 Two-year EFS probability, relapse risk, TRM risk and OS probability among patients alive and relapse free at day 55+ by leukemia risk group, pre-HCT MRDa

The population at highest risk of poor outcome was HR CR1/CR2 patients who were MRD+ pre-HCT who then did not experience grade I–III aGvHD (EFS 12%). Patients with the best outcome were those in the IR category who were MRD <0.1% who experienced grades I–III aGvHD, with EFS above 90%. Of note, HR CR1/CR2 patients who had low MRD, a group thought to have good outcome, had a relatively poor EFS of 47% if they did not get grade I–III aGvHD. Further details regarding the interaction of leukemia risk group, MRD, and aGvHD are provided in Supplementary table 1.

Effect of post-transplant detection of MRD on relapse

There was a total of 262 post-HCT MRD assessments done in 112 patients (95, 88 and 68 assessments at 1, 3 and 8–12 m, respectively), among whom 21 had at least one positive MRD test. Of these 21 patients, 20 had one positive MRD test while one had three positive MRD tests. Detection of any level of post-transplant MRD led to a substantial increase in relapse risk (HR=4.5, P<0.01). Multivariate analysis including pre-HCT MRD and the occurrence of aGvHD showed that the presence of detectable MRD after transplant led to an increase in HR of 3.5 (P=0.003) independent of other factors.

A positive or negative post-HCT MRD test was predictive of relapse or non-relapse, but over time the predictive power of any given positive MRD measurement diminished. Comparing patients whose last MRD test was in the past 60 days, those who tested MRD+ had much higher relapse rates than patients who tested negative (HR 4.5, P=0.01). However, comparing patients whose last MRD test was over 60 days in the past, relapse rates were similar in those who tested positive or negative (HR 1.1, P=0.87). Among the 88 patients who were MRD negative at 1-month post-HCT, 30 eventually relapsed, but none did so within 60 days of the MRD assessment (minimum time was 69 days). In this cohort, there were only two patients who relapsed within 60 days of a negative MRD test.

Flow MRD detected in bone marrow at 8–12 months after HCT was more strongly predictive of relapse than earlier MRD assessments. Risk of relapse by 2 years for patients with MRD detected at 8–12 months post HCT was 73% vs 33% (P=0.01), whereas relapse rates between those with very early post-HCT MRD positivity and those who were MRD negative were not significantly different. This relapse higher relapse rate with MRD positivity late post-HCT also results in worsened survival (Figure 2). Finally, the level of MRD-measured post HCT may be important. Of the 19 MRD+ post-HCT patients with MRD quantitation available, 10 had MRD <0.1% and 9 had MRD 0.1% with 2 (20%) and 6 (67%) relapsing in each group (P=0.07).

Figure 2

The effect of post-HCT MRD positivity on EFS when tested at specific time points after transplant. (a) EFS from day +55 by post-engraftment MRD status. At day +55 patients on study n=131, patients with MRD assessment before day +55 (planned day +30, range day 18–54) n=95, MRD negative n=88, MRD positive n=7. (b) EFS from day +200 by 3-month MRD status. At day +200 patients on study n=131, patients with MRD assessment between day +55 and day +200 (planned day +100, range day 76–181) n=95, MRD negative n=88, MRD positive n=7. (c) EFS from day +300 by 8-month MRD status. At day +300 patients on study n=93, patients with MRD assessment between day +200 and day +300 (optional assessment day +240, range day 226–290) n=37, MRD negative n=31, MRD positive n=6.

Defining target populations and optimal timing for post-transplant interventions

Figure 3a shows a non-parametric patient status analysis of probabilities of (1) relapse-free survival, (2) relapse and (3) TRM in patients the first 2 years after transplant, and the effect of the presence or absence of aGvHD on these events. The graph shows 100% of the patient population and follows them over the first few years post HCT. All patients started alive, without aGvHD (blue lines=no aGvHD), and without relapse. Over time, a portion of patients developed aGvHD (thus moving to the red lines; red lines=experienced aGvHD). Patients with or without aGvHD either stayed alive (solid lines), relapsed (dot–dash) or experienced TRM (dashed lines). The most striking observation was that the probability of relapse after getting aGvHD was low, resulting in remarkable stability of the ‘alive, yes aGvHD’ probability state over time. In contrast, when patients did not get aGvHD their relapse risk was high (as shown by the upward slope of the ‘relapse, no aGvHD’ curve), especially after day +100, and remained high until just over 400 days after transplant. Of note, 59% of all relapses occurred between days +100 and +400 in patients who did not experience aGvHD, implying that a successful intervention to prevent relapse after transplant in this population could have a significant impact on survival.

Figure 3

Aalen–Johansen non-parametric patient state probabilities: (a) all patients (N=143) are represented through day +700 after HCT. (b) patients with low or absent MRD pre-HCT and (c) patients with pre-HCT MRD 0.1% measured by flow cytometry in bone marrow within 2 weeks of initiating their preparative regimen. Red and blue lines indicate patients who did (red) or did not (blue) experience aGvHD. Solid red and blue lines show patients with relapse-free survival, dashed lines represent TRM, dot–dash lines represent relapse. Because TRM was noted to be low in the overall population represented in a, patients, this curve was not included in b and c to emphasize the effect of aGvHD on relapse.

Figures 3b and c are similar patient state probability curves for those with pre-HCT MRD <0.1% or MRD 0.1%, respectively. These figures show the clear pattern of increased risk of relapse in patients not developing aGvHD and low risk in those who do develop aGvHD, regardless of the pre-HCT MRD. This method of analysis demonstrates a window of opportunity for intervention to prevent relapse between day +55 and day +200, when those at high risk of relapse have been defined, and most patients who will relapse have yet to do so.


With promising immunological and cellular therapeutic approaches coming into more general use, it is important to define timing and target populations for intervention to prevent relapse after HCT. This analysis clarifies that both pre-HCT and post-HCT factors (disease risk, pre- and post-HCT MRD and aGvHD) can be used to define populations with poor outcome. Weaknesses of this study derive from the fact that subdividing our cohort created some small clinical subgroups (for example, pre-HCT MRD+ with aGvHD), follow up may not be sufficient to detect differences in some of our subgroups, and one must always approach unplanned comparisons with caution. That said the clinical factors we describe (pre-HCT MRD, post-HCT aGvHD, IR vs HR CR1/CR2) define populations with marked and highly significant differences in outcome, allowing for consideration of interventional trials for these patients.

This is the first multicenter study correlating post-HCT flow MRD with relapse adding data to other studies correlating PCR-based post-HCT MRD with outcomes.30 Measurement of MRD by flow cytometry is the standard technique across pediatric ALL trials in the United States, as it is well suited for rapid turnover. Negative BM flow MRD after HCT provides reassurance regarding relapse for at least a few months, however, high level (0.1%) or late (>8 months after HCT) detection of flow MRD is associated with poor outcome. Although the 6-color flow cytometry used in this study detected disease uniformly to a level of 0.01% or lower,31 we do not know if these results are directly comparable to PCR methods. We also do not know whether higher channel flow cytometry or next-generation sequencing MRD approaches32 would modify our conclusions regarding risk, but we anticipate that more sensitive approaches will only strengthen the predictive ability of pre- and post- HCT MRD.

Because aGvHD alters a given patient’s risk profile, use of the absence of aGvHD as a criterion to choose a population for post-HCT intervention is an attractive approach. In this study, the patients most likely to benefit from intervention were those in HR CR1/CR2 with MRD 0.1% pre-HCT who did not develop aGvHD (OS<20%). Of note, the majority of relapses (58%) occurred in patients who had low (<0.1%) or negative MRD pre-HCT. Because patients with low/negative MRD pre-HCT and no aGvHD have a relapse rate >40%, including this population in relapse reduction studies may be appropriate. If a strategy of treating post-HCT MRD positivity is adopted for patient selection, our data suggest that monitoring would need to be performed every 2 months until at least day +450. Further study using more frequent post-HCT MRD monitoring of later timepoints (from day +100 to +450) is needed to better define the predictive power of this assay after HCT.

Two approaches could be taken for post-HCT intervention. The first would be a ‘maintenance’ strategy, employing agents that do not require intact adaptive immunity (for example, targeted immunotoxins, moxetumomab33 or inotuzumab34). A second approach includes two steps: (1) speeding up the rate of immune taper, and if successfully off immune suppression without significant GvHD, (2) introduction of an agent that uses adaptive immunity to stimulate a GvL effect or targets cancer (for example, blinatumomab35 or chimeric antigen receptor-modified T cells36, 37, 38). Either approach should occur early post HCT in order to avoid the high risk of relapse between day +100 and +450 (Figure 3). Whether such post-transplant approaches will improve survival requires further study. Although some studies have suggested improved survival with early immunological interventions for worsening mixed chimerism,39, 40 others using similar interventions for pre-HCT MRD have shown delay in relapse, but no change in survival.41

In conclusion, we have identified strong interactions between leukemia risk groups, the occurrence of aGvHD and MRD before and after HCT in determining relapse risk and survival in children with ALL after HCT. By day +55, specific populations at very high risk for relapse can be defined. There is a window between day +55 and day +100–200 when the large majority of at-risk patients have not relapsed. Study of interventions with maintenance courses of immunotoxins or weaning immune suppression followed by agents designed to enhance anti-tumor effect may prevent relapse in appropriately defined high-risk populations.


  1. 1

    Pulsipher MA, Peters C, Pui CH . High-risk pediatric acute lymphoblastic leukemia: to transplant or not to transplant? Biol Blood Marrow Transplant 2011; 17: S137–S148.

  2. 2

    Locatelli F, Schrappe M, Bernardo ME, Rutella S . How I treat relapsed childhood acute lymphoblastic leukemia. Blood 2012; 120: 2807–2816.

  3. 3

    Borgmann A, von Stackelberg A, Hartmann R, Ebell W, Klingebiel T, Peters C et al. Unrelated donor stem cell transplantation compared with chemotherapy for children with acute lymphoblastic leukemia in a second remission: a matched-pair analysis. Blood 2003; 101: 3835–3839.

  4. 4

    Schrauder A, von Stackelberg A, Schrappe M, Cornish J, Peters C . Allogeneic hematopoietic SCT in children with ALL: current concepts of ongoing prospective SCT trials. Bone Marrow Transplant 2008; 41: S71–S74.

  5. 5

    Parker C, Waters R, Leighton C, Hancock J, Sutton R, Moorman AV et al. Effect of mitoxantrone on outcome of children with first relapse of acute lymphoblastic leukaemia (ALL R3): an open-label randomised trial. Lancet 2010; 376: 2009–2017.

  6. 6

    Raetz EA, Borowitz MJ, Devidas M, Linda SB, Hunger SP, Winick NJ et al. Reinduction platform for children with first marrow relapse in acute lymphoblastic lymphoma. J Clin Oncol 2008; 26: 3971–3978.

  7. 7

    Horowitz MM, Gale RP, Sondel PM, Goldman JM, Kersey J, Kolb HJ et al. Graft-versus-leukemia reactions after bone marrow transplantation. Blood 1990; 75: 555–562.

  8. 8

    Cornelissen JJ, Carston M, Kollman C, King R, Dekker AW, Lowenberg B et al. Unrelated marrow transplantation for adult patients with poor-risk acute lymphoblastic leukemia: strong graft-versus-leukemia effect and risk factors determining outcome. Blood 2001; 97: 1572–1577.

  9. 9

    Passweg JR, Tiberghien P, Cahn JY, Vowels MR, Camitta BM, Gale RP et al. Graft-versus-leukemia effects in T lineage and B lineage acute lymphoblastic leukemia. Bone Marrow Transplant 1998; 21: 153–158.

  10. 10

    Lee S, Cho BS, Kim SY, Choi SM, Lee DG, Eom KS et al. Allogeneic stem cell transplantation in first complete remission enhances graft-versus-leukemia effect in adults with acute lymphoblastic leukemia: antileukemic activity of chronic graft-versus-host disease. Biol Blood Marrow Transplant 2007; 13: 1083–1094.

  11. 11

    Nordlander A, Mattsson J, Ringden O, Leblanc K, Gustafsson B, Ljungman P et al. Graft-versus-host disease is associated with a lower relapse incidence after hematopoietic stem cell transplantation in patients with acute lymphoblastic leukemia. Biol Blood Marrow Transplant 2004; 10: 195–203.

  12. 12

    Zikos P, Van Lint MT, Lamparelli T, Gualandi F, Occhini D, Bregante S et al. Allogeneic hemopoietic stem cell transplantation for patients with high risk acute lymphoblastic leukemia: favorable impact of chronic graft-versus-host disease on survival and relapse. Haematologica 1998; 83: 896–903.

  13. 13

    Gustafsson Jernberg A, Remberger M, Ringden O, Winiarski J . Graft-versus-leukaemia effect in children: chronic GVHD has a significant impact on relapse and survival. Bone Marrow Transplant 2003; 31: 175–181.

  14. 14

    Locatelli F, Zecca M, Messina C, Rondelli R, Lanino E, Sacchi N et al. Improvement over time in outcome for children with acute lymphoblastic leukemia in second remission given hematopoietic stem cell transplantation from unrelated donors. Leukemia 2002; 16: 2228–2237.

  15. 15

    Dini G, Zecca M, Balduzzi A, Messina C, Masetti R, Fagioli F et al. No difference in outcome between children and adolescents transplanted for acute lymphoblastic leukemia in second remission. Blood 2011; 118: 6683–6690.

  16. 16

    Teachey DT, Sheen C, Hall J, Ryan T, Brown VI, Fish J et al. mTOR inhibitors are synergistic with methotrexate: an effective combination to treat acute lymphoblastic leukemia. Blood 2008; 112: 2020–2023.

  17. 17

    Brown VI, Fang J, Alcorn K, Barr R, Kim JM, Wasserman R et al. Rapamycin is active against B-precursor leukemia in vitro and in vivo, an effect that is modulated by IL-7-mediated signaling. Proc Natl Acad Sci USA 2003; 100: 15113–15118.

  18. 18

    Teachey DT, Obzut DA, Cooperman J, Fang J, Carroll M, Choi JK et al. The mTOR inhibitor CCI-779 induces apoptosis and inhibits growth in preclinical models of primary adult human ALL. Blood 2006; 107: 1149–1155.

  19. 19

    Pulsipher MA, Wall DA, Grimley M, Goyal RK, Boucher KM, Hankins P et al. A phase I/II study of the safety and efficacy of the addition of sirolimus to tacrolimus/methotrexate graft versus host disease prophylaxis after allogeneic haematopoietic cell transplantation in paediatric acute lymphoblastic leukaemia (ALL). Br J Haematol 2009; 147: 691–699.

  20. 20

    Pulsipher MA, Langholz B, Wall DA, Schultz KR, Bunin N, Carroll WL et al. The addition of sirolimus to tacrolimus/methotrexate GVHD prophylaxis in children with ALL: a phase 3 Children's Oncology Group/Pediatric Blood and Marrow Transplant Consortium trial. Blood 2014; 123: 2017–2025.

  21. 21

    Locatelli F, Zecca M, Rondelli R, Bonetti F, Dini G, Prete A et al. Graft versus host disease prophylaxis with low-dose cyclosporine-A reduces the risk of relapse in children with acute leukemia given HLA-identical sibling bone marrow transplantation: results of a randomized trial. Blood 2000; 95: 1572–1579.

  22. 22

    Pulsipher MA, Langholz B, Wall DA, Schultz KR, Bunin N, Carroll WL et al. The addition of sirolimus to tacrolimus/methotrexate GVHD prophylaxis in children with ALL: a phase III COG/PBMTC trial. Blood 2014; 123: 2017–2025.

  23. 23

    Bader P, Kreyenberg H, Henze GH, Eckert C, Reising M, Willasch A et al. Prognostic value of minimal residual disease quantification before allogeneic stem-cell transplantation in relapsed childhood acute lymphoblastic leukemia: the ALL-REZ BFM Study Group. J Clin Oncol 2009; 27: 377–384.

  24. 24

    Borowitz MJ, Wood BL, Devidas M, Loh M, Raetz E, Nachman JB et al. Improved Post-Induction Chemotherapy Does Not Abrogate Prognostic Significance of Minimal Residual Disease (MRD) for Children and Young Adults with High Risk Acute ymphoblastic Leukemia (ALL). A Report From Children's Oncology Group (COG) Study AALL0232. Blood 2011; 18: 625.

  25. 25

    Dworzak MN, Froschl G, Printz D, Mann G, Potschger U, Muhlegger N et al. Prognostic significance and modalities of flow cytometric minimal residual disease detection in childhood acute lymphoblastic leukemia. Blood 2002; 99: 1952–1958.

  26. 26

    Borowitz MJ, Pullen DJ, Shuster JJ, Viswanatha D, Montgomery K, Willman CL et al. Minimal residual disease detection in childhood precursor-B-cell acute lymphoblastic leukemia: relation to other risk factors. A Children's Oncology Group study. Leukemia 2003; 17: 1566–1572.

  27. 27

    Kaplan EL, Meier P . Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958; 53: 457–481.

  28. 28

    Cox D . Regression models and life tables. J R Stat Soc B 1972; 34: 187–202.

  29. 29

    Aalen O, Johansen S . An empirical transition matrix for nonhomogeneous Markov chains based on censored observations. Scand J Statist 1978; 5: 141–150.

  30. 30

    Bader P, Kreyenberg H, von Stackelberg A, Eckert C, Salzmann-Manrique E, Meisel R et al. Monitoring of minimal residual disease after allogeneic stem-cell transplantation in relapsed childhood acute lymphoblastic leukemia allows for the identification of impending relapse: results of the ALL-BFM-SCT 2003 Trial. J Clin Oncol 2015; 33: 1275–1284.

  31. 31

    Bruggemann M, Schrauder A, Raff T, Pfeifer H, Dworzak M, Ottmann OG et al. Standardized MRD quantification in European ALL trials: proceedings of the Second International Symposium on MRD assessment in Kiel, Germany, 18-20 September 2008. Leukemia 2010; 24: 521–535.

  32. 32

    Faham M, Zheng J, Moorhead M, Carlton VE, Stow P, Coustan-Smith E et al. Deep-sequencing approach for minimal residual disease detection in acute lymphoblastic leukemia. Blood 2012; 120: 5173–5180.

  33. 33

    Kreitman RJ, Tallman MS, Robak T, Coutre S, Wilson WH, Stetler-Stevenson M et al. Phase I trial of anti-CD22 recombinant immunotoxin moxetumomab pasudotox (CAT-8015 or HA22) in patients with hairy cell leukemia. J Clin Oncol 2012; 30: 1822–1828.

  34. 34

    Dijoseph JF, Dougher MM, Armellino DC, Evans DY, Damle NK . Therapeutic potential of CD22-specific antibody-targeted chemotherapy using inotuzumab ozogamicin (CMC-544) for the treatment of acute lymphoblastic leukemia. Leukemia 2007; 21: 2240–2245.

  35. 35

    Topp MS, Kufer P, Gokbuget N, Goebeler M, Klinger M, Neumann S et al. Targeted therapy with the T-cell-engaging antibody blinatumomab of chemotherapy-refractory minimal residual disease in B-lineage acute lymphoblastic leukemia patients results in high response rate and prolonged leukemia-free survival. J Clin Oncol 2011; 29: 2493–2498.

  36. 36

    Kalos M, Levine BL, Porter DL, Katz S, Grupp SA, Bagg A et al. T cells with chimeric antigen receptors have potent antitumor effects and can establish memory in patients with advanced leukemia. Sci Transl Med 2011; 3: 95ra73.

  37. 37

    Grupp SA, Kalos M, Barrett D, Aplenc R, Porter DL, Rheingold SR et al. Chimeric antigen receptor-modified T cells for acute lymphoid leukemia. N Engl J Med 2013; 368: 1509–1518.

  38. 38

    Grupp SA, June CH . Adoptive cellular therapy. Curr Top Microbiol Immunol 2011; 344: 149–172.

  39. 39

    Bader P, Kreyenberg H, Hoelle W, Dueckers G, Handgretinger R, Lang P et al. Increasing mixed chimerism is an important prognostic factor for unfavorable outcome in children with acute lymphoblastic leukemia after allogeneic stem-cell transplantation: possible role for pre-emptive immunotherapy? J Clin Oncol 2004; 22: 1696–1705.

  40. 40

    Rettinger E, Willasch AM, Kreyenberg H, Borkhardt A, Holter W, Kremens B et al. Preemptive immunotherapy in childhood acute myeloid leukemia for patients showing evidence of mixed chimerism after allogeneic stem cell transplantation. Blood 2011; 118: 5681–5688.

  41. 41

    Lankester AC, Bierings MB, van Wering ER, Wijkhuijs AJ, de Weger RA, Wijnen JT et al. Preemptive alloimmune intervention in high-risk pediatric acute lymphoblastic leukemia patients guided by minimal residual disease level before stem cell transplantation. Leukemia 2010; 24: 1462–1469.

Download references


This study was supported in part by N01 HC-45220/HHSN268200425220C, U10 CA098543 and R01CA1116660. PBMTC activities were supported by 2U01HL069254 and a consortium grant from the St Baldrick’s Foundation.

Author information

Correspondence to M A Pulsipher.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Additional information

This study was presented in part at the American Society of Hematology Meeting, December 2012 and the Second International Workshop: Biology Prevention and Treatment of Relapse After Allogeneic Hematopoietic Stem Cell Transplantation, November 5–6, 2012, Bethesda MD, Plenary Oral Presentation, Best Abstracts Session.

Supplementary Information accompanies this paper on Bone Marrow Transplantation website

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Further reading