Impact of minimal residual disease (MRD) on prognosis in children with acute lymphoblastic leukemia (ALL) according to WBC count, age and TEL/AML1 status at diagnosis. results of the AIEOP-BFM ALL 2000 study.

Minimal residual disease (MRD), the most sensitive method to evaluate treatment response, has been adopted to stratify patients in study AIEOP-BFM ALL 2000. To assess whether PCR-MRD levels discriminate outcome in patients classified by WBC count, age at diagnosis, NCI criteria (Standard Risk, SR: WBC < 50,000/cmm and age 1–9 years; High Risk, HR: all others) and TEL/AML1 status. Between 07–2000 and 07–2006, 4,730 Ph-negative patients were enrolled in AIEOP-BFM ALL 2000 study. They were treated with BFM Induction (protocol IA) consolidation (protocol IB), extra-compartment/intensified consolidation (HD-MTX in non-HR patients, blocks in HR patients), reinduction therapy (one or more Protocols II or III), followed by maintainance. BM samples obtained at weeks 5 (Time Point 1, TP1) and 12 (TP2) of induction/consolidation therapy were used for PCR-based MRD analysis of patient specific gene targets. At least 2 sensitive markers ( 1 x 10–4) could be determined in 3,707 (78.4%) patients. SR was defined by MRD– at both TP1 and TP2; HR by MRD 1x10–3 at TP2; Intermediate Risk (IR): all others. Median follow-up was 3 years; 5-year percent EFS (SE) estimates are given.Patients at MRD-SR, IR or HR had, respectively, an EFS of 93.0 (1.0), 80.5 (1.5) and 43.4 (6.0) in patients with WBC <50,000/cmm vs 90.4 (2.6), 72.4 (3.0) and 47.0 (5.1) in patients with WBC 50,000/cmm. Patients at MRD SR, IR or HR had, respectively, EFS of 93.6 (1.0), 80.3 (1.5) and 44.1 (5.4) if aged 1–9 years vs 87.2 (3.4), 73.9 (3.0) and 49.3 (5.2) if aged 10 years. Patients at SR by NCI criteria [N= 2,355, EFS of 85.3 (1.0)] were stratified by PCR-MRD as SR (N=1046; 44.4%), IR (N=1198; 50.9%), or HR (N=111; 4.7%). EFS in these subgroups was 93.9 (1.0), 81.3 (1.6) and 43.9 (7.2), respectively (p<0.001). In patients at HR by NCI criteria [N=1,352, EFS of 75.6 (1.6)], 403 (29.8%), 774 (57.3%) and 175 (12.9%) respectively were at SR IR and HR by MRD. EFS was 89.4 (2.2) in MRD SR, 74.7 (2.3) in MRD IR and 47.9 (4.2) in MRD HR patients (p<0.001). Of 3,707 study patients, 3,410 were investigated for TEL/AML1 status: 771 (22.6%) were positive and 2,639 were negative. TEL/AML1+ patients were at SR (N=444; 57.6%) or IR (N=317; 41.1%) or HR (N=10; 1.3%) by PCR-MRD; EFS in this subgroup was 94.4% (1.5), 80% (3.7) and 60% (18.4), respectively (p<0.001). TEL/AML1– patients at SR (N=887; 33.6%) or IR (N=1497; 56.7%) or HR (N=255; 9.7%) had an EFS of 91.6% (1.3), 78.5% (1.4) and 45.7% (4.5), respectively (p<0.001). PCR-MRD in patients treated with BFM-oriented therapy overcomes the prognostic value of "historical" factors such as WBC count, age, NCI criteria or TEL/AML1 status, as it markedly discriminates prognosis within each subgroup defined by these variables. Study design for contemporary risk-directed therapy of childhood ALL should incorporate a technique for MRD determination.