Minimal Residual Disease Status Is the Most Important Preditive Factor in Adults with Acute Lymphoblastic Leukemia. Prospective, Multicenter PALG 4-2002 MRD Study.

Current therapeutic protocols for adult acute lymphoblastic leukemia (ALL) take into account the risk of relapse, in order adjust the treatment intensity to individual patient needs. It is postulated that in addition to “classical” risk criteria the status of minimal residual disease (MRD) should be considered for treatment decisions. The aim of this study was to prospectively evaluate the feasibility and prognostic significance of MRD detected with the use of immunophenotyping for outcome of ALL patients treated according to 4-2002 protocol of the Polish Adult Leukemia Group (PALG). Induction therapy included PDN, Asp and 4x epirubicin+VCR. Consolidation consisted of 2x high-dose AraC+Cy, 2x Mtx+Vep, 6MP, and CNS prophylaxis. Patients stratified to high risk (HR) group according to “classical” criteria based on those formerly developed by GMALL (bcr/abl(+), WBC>30 G/L, prepreB, early or mature T phenotype, age>35y, or prolonged time to achieve CR) were further referred for hematopoietic cell transplantation (HCT), whereas those assigned to standard risk (SR) group were treated with maintenance for 2 years. MRD was tested at the level of 0.1% after completion of induction and consolidation therapy in patients achieving CR, employing multicolor flow-cytometry, including a new “empty spaces” method taking into account an individual pattern of antigen expression on blast cells. 165 ALL pts (B-lineage 79%, T-lineage 21%), aged 29 y (17–60) were included. CR rate equaled 85,5%. 23% of CR pts were assigned to SR, 77%- to HR according to classical criteria. MRD evaluation was possible in all but 8 pts. After induction 37% of CR pts were found MRD(+). Among those who remained in CR, MRD after consolidation was detected in 26% of cases. 64% of patients were MRD(−) at both time-points, whereas in the remaining 36% of cases MRD was detected at least once. MRD status affected both relapse incidence (RI) and leukemia-free survival (LFS). After 3 years the RI was higher for pts with MRD(+) vs. MRD(−) if assessed after induction (82%vs.29%,p=0.00007) and after consolidation (62%vs.41%,p=0.05). For pts with MRD(−) at both study end-points the probability of LFS was 65% whereas for those with MRD(+) after either induction and/or consolidation − 26% (p=0.008). In the respective subgroups RI equaled 28% and 73% (p=0.004). The difference was observed for patients assigned to SR group (20%vs.92%,p=0.01) as well as to HR group (33%vs.70%,p=0.05). In a multivariate analysis including classical risk criteria the MRD status remained the only significant factor predictive for RI (HR: 2.5(1.3–4.8),p=0.006) and LFS (HR: 2.1(1.2–3.9),p=0.01). We conclude that immunophenotyping employing “empty spaces” method is feasible for MRD evaluation in adults with ALL. MRD stzatus after induction and consolidation is the most important predictive factor for RI and LFS. Based on our findings patients with MRD detected after induction and/or consolidation should be offered intensified treatment with the use of HCT irrespective of the absence of other risk factors.