Contribution of All-Trans Retinoic Acid to Improved Early Relapse-Free Outcome in Infant Acute Lymphoblastic Leukemia Comparing to the Chemotherapy Alone.

The efficacy of conventional and intensified chemotherapy in infant acute lymphoblastic leukemia (ALL) is limited worldwide particularly because of the high early relapses and treatment related mortality. We hypothesized that all-trans retinoid acid (ATRA) may participate in the differentiation of infant leukemic cells to normal progenitors with a consequent loss of malignant features and enhanced antiproliferative effects of chemotherapy. Our prospective multicenter non-randomized MLL-Baby/ALL-MB 2002 trial evaluated whether the novel recently developed ATRA based regimen, MLL-Baby is safe and more effective in the prevention of very early relapses comparing to ALL-MB 2002 - a well established protocol in Russia and Belarus. The trial was approved by Ethics Committees and parents’ informed consent was signed in all cases. The decision to receive MLL-Baby or ALL-MB 2002 was judged by the treating physician. The major difference between the 2 treatment approaches was the addition of ATRA in the MLL-Baby protocol. Patients were equally distributed of patients (pts.) between the two groups according to risk: only pts. with t(4;11) and/or MLL/AF-4 or without complete remission (CR) achieved on days 36/43 were subjects for the high risk group (HRG). The remaining pts were stratified to an intermediate risk group (IRG). Treatment schedules were also similar in both regimens except for CNS-disease prophylaxis: in MLL-Baby, cranial irradiation 12 Gy was retained only for HRG pts with initial CNS involvement who were older than 12 months and substituted by 5 additional intrathecal treatments, while all pts. allocated to ALL-MB 2002 1year of age underwent cranial irradiation 12 Gy, regardless of their initial CNS status and risk group. In MLL-Baby pts, the initial ATRA pulse in daily dose 25 mg/m² was given after debulking of the main tumor from day 36 to day 43, followed by 12 in IRG and 16 in HRG 2-weeks of ATRA courses alternating with chemotherapy. Between September 2003 and November 2006, 40 pts. aged 1–12 months were enrolled onto MLL-Baby (n=19) and ALL-MB 2002 (n=21) protocols. Both groups were balanced for median age 8 (1–12) and 6 (2–12) months, p=0.47; initial WBC 115,5 (1,2–450) and 105,5 (2.6–350) per microliter, p=0.76; initial CNS disease 3 vs. 5, p=0.5; 11q23/MLL translocations 12 vs. 10 pts., including t(4;11) - 6 vs. 7 pts, p=0.74 and BI phenotype 9 and 4 cases correspondingly, p=0.37, respectively. No significant differences were observed between MLL-Baby and ALL-MB 2002 with respect to rates of induction deaths: 1 vs. 3 cases, CR rates 94.7% vs. 85.7%, deaths in CR 1 vs. 0; whereas the early relapse rate was higher with ALL-MB 2002 compared to MLL-Baby: RFS was 0.24±0.13 vs. 0.94±0.05 (p=0.02) with a median of follow-up 20 months (1–55). EFS differed significantly: 0.21±0.11 vs. 0.84±0.08 (p=0.03). Most relapses on ALL-MB 2002 (n=11) and MLL-Baby (n=1) developed very early: 7 of 11 pts. and 1 of 1 pt. respectively and were localized only to bone marrow in most cases: 6 of 11 pts. and 1 of 1 pt. correspondingly. Our data indicate that ATRA administration contributes positively towards early relapse-free outcome in infants with ALL. A prospective randomized clinical trial with longer follow-up is needed to confirm these results.