This multicentre French retrospective study of 38 patients treated with clofarabine for ALL shows results below expectations. In the group of 30 patients who received clofarabine for an ALL in relapse, the ORR obtained was only 37%. This value is lower than those of 55% reported by Hijiya et al.
 and 56% by Locatelli et al.
 for patients treated with a combination of clofarabine, cyclophosphamide and etoposide. However, these phase I and/or II studies were conducted with patients who met the eligibility criteria. We obtained better results than studies with clofarabine used alone, where remission rates are about 20%
[12, 14–16]. Two additional studies have been recently published. The study by O’Connor, similar to ours, presents the UK experience of clofarabine in the treatment of relapsed and refractory paediatric ALL, and they obtain an ORR of 52%
. The second is the phase II study of Hijiya
, and he reports an ORR of 44%. However, in our study, most of patients had advanced disease when clofarabine was administered. They were heavily pre-treated, with an average of 2.5 treatments before clofarabine, and 70% of them were refractory to their most recent prior treatment. These factors may have contributed to the low ORR we observed. Patients in the studies by O’Connor and Hijiya had received a mean of only 2 or 2.1 previous treatments, respectively. Hijiya reports that 60% of the patients were refractory to their most recent treatment
, but no information was provided about refractory patients in the O’Connor report
. Only one study published recently, reports lower results than ours for a population of refractory or relapsed ALL
. Nevertheless, this result must be interpreted with caution due to the small number of patients included (9).
Surprisingly, the ORR was similar for refractory patients (ORR 38%) and those treated with clofarabine as the first line for a relapse (33%). Also, the ORR was not better for the five patients treated with clofarabine for a first relapse and who had only received one prior treatment (ORR 40%). The results obtained for these five patients differ from those reported by O’Connor
. Effectively, in the same conditions, O’Connor reported an ORR of 86% (7 patients, 6 CR). Obviously, all these results must be interpreted with caution due to the small numbers of patients involved.
Ten of the 11 patients in remission after clofarabine therapy subsequently underwent HSCT (90%) and five of these ten were alive (50%) at the end of data analysis with a median follow-up post HSCT of 13 months (range; 4.5-22). These data are similar to those reported by other studies with transplantation rates for patients in remission after clofarabine of 50% for Locatelli et al.
, 54% for Hijiya et al.
, 63% for Hijiya et al.
 and 83% for O’Connor et al.
 The survival rates after HSCT for responding patients were 57% (4/7 patients), in the study by Locatelli with a median follow-up of 8 months after HSCT
, and 60% (6/10 patients) in the study by O’Connor, with a median follow-up of 13 months post HSCT
. Finally, the proportion of patients in remission after clofarabine was lower in our cohort than in other studies, but the post-clofarabine transplantation rate and survival rate were similar to other published values.
Eight patients in complete remission but with a high MRD level were treated by clofarabine. But clofarabine treatment only moderately improved (one log or less) the MRD levels in four of these eight patients. Most of these patients were transplanted after clofarabine treatment and the survival rate in this sub-group was high (87.5% one year after clofarabine). It is not possible to draw informative conclusions because of the small number of patients in this subset. Inaba et al.
 reported two cases treated for a high MRD, and both showed a significant improvement of their MRD.
Adverse events associated with clofarabine treatment are frequent. In our study, clofarabine was generally well tolerated. However, we report four deaths (10.5%) that could potentially be attributed to the treatment. These four patients were all in poor condition and developed multi-organ failure after clofarabine treatment. Hijiya et al.
 report a death rate of 8% (2 patients) after clofarabine treatment, but no deaths were reported by Locatelli et al.
 or O’Connor et al.
. In their phase II study, Hijiya et al.
 reported a substantial drug-related death rate of 24% (6/25 patients) due notably to severe hepatotoxicity with VOD.
For other adverse events, the incidences we observed were similar to published data. The incidence of febrile neutropenia in our study (79%) is comparable with those reported previously: 60-65%
[19, 22, 23]. We had fewer documented infections (grade ≥3) than in other studies, with only nine cases (24%) including six of sepsis. Infection rates reported previously are in a range of 30-72%
[19, 20, 22] depending on supportive therapy and infection prophylaxis. Thus, infectious complications are frequent with clofarabine treatment. To decrease their incidence, prophylactic treatments against bacterial and fungal infections with an active monitoring of patients are required.
Hepatic toxicity is frequent when clofarabine is used as single agent or in combination
[12, 14, 18]. Some severe and life-threatening cases of hepatic toxicity were described by Hijiya et al.
[19, 23] (3 VOD, 1 hyperbilirubinaemia). In our study, grade 3 hepatotoxicity occurred in nine patients (24%; 1 hyperbilirubinaemia and elevated transaminase and 8 elevated transaminases), but there were no cases of VOD. The hepatotoxicity rate in our study was not higher among patients with previous HSCT than among those without HSCT.