Characteristics and Outcome Of Children and Young Adults With Early T-Precusor (ETP) ALL Treated On UKALL 2003

Early T-precursor (ETP) ALL is a recently described poor prognostic sub-group of childhood T-ALL identified by a distinctive gene expression profile and immunophenotype (CD1a^-, CD5 ^weak, CD8^- and presence of myeloid or stem cell markers). It accounts for around 12% of T-ALL and is associated with a poor outcome in retrospective US and Italian studies with a 5 year event-free survival (EFS) of less than 50% on historical treatment protocols. We sought to determine the outcome for this sub-group on a contemporary MRD stratified treatment protocol, UKALL 2003.

As full immunophenotype data was not collected centrally, individual centres were asked to provide individual patient data retrospectively. Of 387 T-ALL cases in the trial, immunophenotype data to classify ETP status was not available for 148 patients. Of the remaining 239 cases, 35 (15%) were classified as definite/probable (CD8 and 1a neg with stem cell or myeloid antigen expression) and 17 (7%) as possible (CD8 neg and stem cell or myeloid antigen expression, 1a not available) ETP. The definite, probable and possible ETP groups had similar patient characteristics and outcomes and have, therefore, been analysed together. Compared to those without an ETP phenotype, ETP patients were more likely to present with a low (<50x10⁹/l) white cell count (56% vs 26%, p<0.0001) but were less likely to be morphological rapid early responders at day 8/15 (63% vs 83% p =0.002) or MRD low risk at the end of induction (4% vs 27%, p = 0.008). Unlike previous reports, the risk of induction failure was not significantly higher in the ETP group (n= 2 vs 1 among non-ETP cases).

With a median follow-up of 4 years 10 months (range 1 year 4 months - 9 years), the ETP sub-group have a non-significantly inferior EFS (76.0% (95% CI: 64.0-88.0%) vs 84.6% (79.3-89.9%) at 5 years, p = 0.2) and OS (84.1% (73.9-94.3%) vs 90.9% (86.6-95.2%), p = 0.08) compared to the non-ETP sub-group, due primarily to a non-significantly higher relapse rate (17.7% (6.5-28.9%) vs 9.6% (5.1-14.1%), p = 0.1). Although MRD risk group was a determinant of outcome in the non-ETP group (p=0.05), its significance within the ETP group could not be reliably determined since only 2 patients were MRD low risk, of whom 1 suffered a relapse. Interestingly, of the 8 relapses in the ETP group, 7 were isolated marrow relapses and there was one isolated testicular relapse, whereas 12 of the 17 relapses in the non-ETP group involved the CNS (5 isolated and 7 combined).

Although ETP ALL is associated with an adverse outcome compared with non-ETP T-ALL, a 5 year EFS of 76% on a contemporary MRD stratified protocol is not sufficiently poor to warrant experimental treatment or first remission allogeneic transplant for the group universally.