Aim: To assess the role of autologous transplantation (AT) compared with chemotherapy (CT) for low risk relapsed childhood acute lymphoblastic leukaemia (ALL).

Patients and Methods: Since 1997 at a single Institution, 30 pediatric consecutive patients (pts), lacking a compatible related donor, underwent immunologically purified peripheral stem cell AT, for B cell precursor ALL in second remission (CR2) after late (>30 ms after diagnosis) or extra-medullary (BM) relapse, belonging to S1–S2 BFM risk group. Since 2000 the positivity of minimal residual disease (MRD) at transplant was considered an exclusion criterium. For each AT patient all possible controls were selected among 236 S1 or S2 pts in CR2 treated with CT in all BFM Centers, matched for: site of relapse, CR1 duration, relapse period and waiting time to transplant. Outcome data are expressed according to the KM estimator for the AT group; a weighted version of the KM estimator is used for the CT group in order to account for the variable proportion of matching; a p-value for the comparison at 4 years (ys) is provided by a permutation test. The role of gender and age at relapse are assessed in a multivariate analysis by a Cox model. The impact of MRD is evaluated.

Results: 103 CT controls were selected with a median matching ratio of 4 controls (1–8) for each AT patient. Eight pts in the AT group presented with subsequent relapse at a median of 17 ms (11–48) and 50 in the CT group at a median of 22 ms (5–59) after first relapse; 6 of 8 and 18 of 50 relapsed pts in the two groups underwent allogeneic transplant in CR3 and 4 and 15 of them, respectively, are alive. All events were relapses and no pts died of treatment related complications in CR2. The probability of DFS at 4 ys was 71.3% (SE 8.8) for the AT pts and 44.3% (SE 6.2) for the CT group (p-value: 0.0165) and the probability of survival was 85.8% (SE 6.6) and 65.7% (ES 6.5) (p-value: 0.0385) with a median follow-up of 4.9 ys. The advantage of AT was consistent within the subgroups of late BM and extra-BM relapsed pts. Age and gender did not significantly affect DFS (HR male vs female: 1.82, p-value: 0.08; age >10 ys vs <10 ys: 1.24, p-value: 0,53). The time period of relapse seemed to play a role among pts receiving CT, which reported a better survival when relapse occurred after 2000, which could be influenced by MRD driven patient selection. The impact of MRD was striking in the AT pts, among whom all the 4 MRD positive pts transplanted before 2000, relapsed. MRD data after induction were available for 17 of 30 AT and 36 of 103 CT pts; 5 of 9 MRD positive AT pts and 9 of 12 MRD positive CT pts relapsed; 2 of 8 and 5 of 22 MRD negative pts relapsed among the AT and CT pts, respectively.

Conclusions: Children treated with autotransplantation reported a better outcome compared to matched patients receiving chemotherapy. These results in low risk relapsed ALL should be confirmed in a prospective controlled study.

Disclosure: No relevant conflicts of interest to declare.

Author notes


Corresponding author