The event-free survival (EFS) of children with SR-ALL is now > 80%. Nonetheless, ALL relapse remains a major problem in pediatric cancer, and optimal therapy for these children remains to be defined. We examined post relapse outcomes for children initially treated on the Children’s Cancer Group CCG-1952 study and compared stem cell transplant (SCT) with chemotherapy (CT) as salvage treatment in second remission (CR2). Between 5-96 and 1-00, 2176 eligible patients with SR-ALL (WBC < 50,000/mcl; age ≥ 1, < 10 years) were enrolled on CCG-1952, 321 of whom experienced a relapse after initially achieving M1 remission. Among the relapses, 196 are bone marrow (BM) ± extramedullary (EM) relapses (61%) and 125 are isolated EM (iEM: CNS, testicular, or ocular) (39%) relapses. The mean time (range) from first remission to BM relapse, iCNS relapse, and itesticular relapse is 34 ± 1 (2 to 79) month (mo.), 22 ± 1 (1 to 70) mo., and 37 ± 3 (4 to 74) mo., respectively. The 3-year EFS and overall survival (OS) after relapse for all patients are 41% and 53%. Three-year EFS and OS after BM ± EM relapse are 32% and 40% and after iEM relapse are 55% and 71%. Patients with early BM relapse (CR1 < 36 months) or early iEM relapse (CR1 < 18 months) are 2.3 times (p=0.002) and 2.8 times (p=0.01) more likely to suffer a subsequent adverse event than patients with later relapse by Cox regression analysis after adjustment for age, pre-relapse treatment, day 7/14 BM status, and type of treatment in CR2. The 3-year OS for early and late relapses are 31% vs 59% (p=0.001) for BM relapse and 50% vs 87% (p=0.01) for iEM relapse. We compared 73 SCT and 215 CT patients, excluding 33 patients with adverse events prior to the median time to SCT (130 days, range 56 to 1148 days) using Kaplan-Meier life table analysis and the log rank test. The cohorts are similar with respect to age (p=0.57), duration of CR1 (p=0.96), and initial Day 14 BM status (p=0.69). Thirty-seven percent (n=62) of patients with BM relapse (33 early; 29 late) and 9% (n=11) of patients with iEM relapse (6 early, 5 late) received SCT. Thus far, outcomes are similar between cohorts for patients with any BM relapse (OS, p=0.45; EFS, p=0.70), early BM relapse (OS, p=0.95; EFS, p=0.66), and iEM relapse (OS, p=0.44; EFS, p=0.96). Among patients with later BM relapse, a trend favors CT (OS, p=0.08; EFS, p= 0.14). In summary, duration of CR1 remains the most significant predictor of outcome after either BM or iEM relapse in this SR-ALL cohort. Prognosis after early BM relapse remains poor with either SCT or CT.

Author notes

Corresponding author

Sign in via your Institution