In the early 1990s, the BFM study group decided to selectively intensify chemotherapy of high-risk (HR) patients at an early point of treatment. As a consequence the formerly used post-induction element Phase IB (6-mercaptopurine 60 mg/m2/d on days 1–28; cyclophosphamide 1000 mg/m2/d on days 1 and 29; cytarabine 75 mg/m2/d on days 3–6, 10–13, 17–20, and 24–27; methotrexate 12 mg intrathecal on days 45 and 59) was no longer used for HR patients in trials ALL-BFM 90 and ALL-BFM 95, but was replaced by intensive rotational chemotherapy blocks. In the subsequent trial ALL-BFM 2000, minimal residual disease (MRD) was introduced into the risk stratification. Therefore, Phase IB was reintroduced for the HR group to have comparable chemotherapy for all risk groups during the phase of MRD measurements. In the present analysis we assessed the impact of reintroduction of Phase IB on treatment outcome through a historical comparison of ALL-BFM 2000 and ALL-BFM 95.
In both trials ALL-BFM 95 and ALL-BFM 2000 prednisone poor-response (≥1000/μl blasts in blood on day 8 of the prednisone pre-phase), non-remission after induction treatment (NRd33) and/or presence of BCR-ABL or MLL-AF4 qualified for HR treatment. In ALL-BFM 2000, MRD was analyzed after induction Phase IA (day 33, TP1) and after Phase IB (week 12, TP2); MRD load of ≥10E-3 at TP2 was an additional HR criterion. Since no MRD was analyzed in ALL-BFM 95, the two trials were compared including only patients being HR according to the ALL-BFM 95 criteria. Apart from Phase IB HR treatment was very similar in the two trials.
From 08/1999 to 06/2006, 455 HR patients from Austria, Germany and Switzerland were enrolled into trial ALL-BFM 2000. Nine patients were excluded due to post-induction treatment in the EsPhALL trial for BCR-ABL -positive ALL. Of the remaining 446 patients, 371 were HR by ALL-BFM 95 risk criteria (“HR-95”) and were used for the historical comparison with the 238 HR patients of ALL-BFM 95. Comparing the characteristics of “HR-95” patients derived from ALL-BFM 95 and ALL-BFM 2000 revealed differences in the distribution of age, immunophenotype, MLL-AF4 and BCR-ABL (ALL-BFM 95 vs. ALL-BFM 2000: age ≥10 years 34.5% vs. 42.6%, p=0.045; T-ALL 36.1% vs. 43.4%, p=0.075; MLL-AF4 8.0% vs. 3.1%, p=0.011; BCR-ABL 18.8% vs. 13.4%, p=0.078). The percentage of patients who underwent allogeneic stem cell transplantation (alloSCT) was significantly higher in ALL-BFM 2000 (41.8% vs. 23.1%, p<0.001).
Probability of event-free survival at 5 years (5y-pEFS) of “HR-95” patients was 53% (SE 3%) in ALL-BFM 95 and 66% (SE 2%) in ALL-BFM 2000 (p=0.002). Analyzing biological subgroups showed a significantly better 5y-pEFS in ALL-BFM 2000 in “HR-95” patients with T-ALL (67% [SE 4%] vs. 50% [SE 5%], p=0.0097) and BCR-ABL -positive ALL (48% [SE 7%] vs. 26% [SE 7%], p=0,028), while no significant difference could be demonstrated in patients with BCR-ABL -negative precursor-B (pB-)ALL (70% [SE 4%] vs. 64% [SE 5%], p=0.34). In a Cox regression analysis including the variables age, immunophenotype, MLL-AF4, BCR-ABL, NRd33, alloSCT (as time-dependent variable), and trial, treatment in ALL-BFM 2000 (i.e. with additional administration of Phase IB) remained a favorable factor with a Hazard ratio of 0.64 (p=0.002).
MRD results in HR patients in ALL-BFM 2000 (n=446) substantiated the good efficacy of Phase IB: Of the patients with a very high MRD load at TP1 (≥10E-2 before Phase IB) 38% could reduce their MRD load by ≥2 log-steps until TP2 (after Phase IB). In particular patients with T-ALL benefited from Phase IB with an MRD reduction by ≥2 log-steps until TP2 in 52% of the patients in comparison to 23% in pB-ALL (p<0.001).
Phase IB was well tolerated in ALL-BFM 2000 with 3 (0.7%) documented life-threatening serious adverse events (SAE) in the HR group in comparison to 7 (1.6%) life-threatening SAE in association with the first HR course.
The ALL-BFM 95 treatment is currently used as regular chemotherapy protocol for childhood ALL in several countries. These data demonstrate that by the addition of Phase IB to the HR treatment arm a significantly better outcome could be achieved in this critical patient group without major additional toxicity.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.