Outcome of Treatment after First Relapse in Adults Patients with Acute Lymphoblastic Leukemia (ALL) Initially Treated by the LALA-94 Trial.

Introduction: More than half of adults with ALL suffer a relapse. It is generally accepted that the only curative approach at this stage is stem cell transplantation (SCT). In the LALA-94 trial, we analyzed the outcome in patients (pts) undergoing a first relapse, and evaluated the possibility to undergo SCT.

Methods: Of 771 ALL pts (15–55 years (y)) who entered the trial and achieved complete remission (CR), 421 experienced a first relapse between 1994 and 2004. Except for 48 pts who died early before any chemotherapy, all other pts were given re-induction therapy with various salvage chemotherapy regimens (355 pts), autologous SCT (3 pts) with cells harvested in first CR, donor lymphocyte infusions (DLI) (1 pt) or allogeneic (allo) SCT (13 pts) from a geno-identical (7 pts) or a pheno-identical (6 pts) donor. After CR achievement with chemotherapy, pts with an HLA-identical sibling or unrelated donor were systematically assigned to allo SCT (61 pts: 33 from a related-identical donor, 27 from MUD, and one from cord blood). When possible, allo SCT was also proposed to pts refractory to re-induction chemotherapy (24 pts: 14 from a related-identical donor, 7 from MUD, and 3 from cord blood).

Results: Overall, 187/421 pts (44%) achieved CR. CR proportion according to risk-groups defined in first line therapy were: 52% for standard-risk ALL, 37% for high-risk ALL, 40% for Ph+ ALL, and 36% for CNS+ ALL. There were no significant differences in CR proportion according to the first CR duration. With a median follow-up of 4.3 y, the median overall survival (OS) was 6.3 months (m). Median disease free survival (DFS) was 5.2 m with 5-y DFS at 12%. The estimated 5-y DFS rates were 11%, and 14% for T- and (Ph neg) B-lineage ALL, respectively. SCT (p<0.0001), first CR duration (p=0.03), and platelets at relapse (p=0.02) were predictors of survival in multivariate analysis. 111/421 pts (26%) had a sibling donor identified during first line therapy and available at relapse. 9 more pts had a sibling donor identified at the time of relapse. Overall allo SCT from a sibling donor could be performed in 54 pts (45% of pts with a sibling donor), and 3 pts received DLI. 40 SCT could be performed from a match unrelated donor (MUD) and 4 from cord blood. Our results showed a significantly higher survival rate at 3 y after SCT in second CR as compared to that of SCT at time of relapse (p = 0.02) or to that of SCT with active disease after failure of reinduction chemotherapy (p = 0.005). When comparing allo SCT from MUD and allo SCT from related-identical donor, survival tended to be better with MUD (3-y OS: 31% vs 21%), in relationship with a lower relapse incidence (RI) (3-y RI: 41% vs 64%) while treatment related mortality (TRM) was quiet similar (TRM at 100 days: 19% vs 35%, and 1-y TRM: 37% vs 46%).

Conclusions: Our results showed a second CR in less than 50% of pts and difficulties to organize allo SCT in second CR with however a definitive advantage for pts undergoing SCT when compared to chemotherapy alone. This questions about the opportunity to perform allo SCT, when possible, systematically earlier in the evolution of the disease.