Improved Outcome in Adult T-ALL by Risk Adapted Treatment Strategies.

T-cell acute lymphoblastic leukemia (T-ALL) accounts for 25% of adult ALL and is characterised by specific clinical and biologic features such as phenotype with early (cyCD3+,CD7+) (E-T), thymic (CD1a+) ) (Thy-T) and mature T-ALL (M-T) (sCD3+). The formerly poor prognosis of T-ALL was more recently improved in some studies albeit not in all, even not in childhood ALL.

Patients: To improve outcome of T-ALL the German Multicenter Study Group for Adult ALL initiated two consecutive studies with subtype adapted therapies. 503 T-ALL pts were recruited between 4/93 and 10/03. The median age was 30 (15–55)yrs, 75% were male, 66% had mediastinal tumor (MedTu), 24% WBC > 100.000 and 7% CNS involvement with subtypes as follows: 53% Thy-T, 26% E-T and 21% M-T.

Study Design: In Study 05/93 all T-ALL pts were treated uniformly with 8drug induction incl. prophylactic CNS (24 Gy) (CNSRAD) and proph. mediastinal (24 Gy) irradiation (MEDRAD) followed by 7x consolidation (HDAC/MITOX, HDMTX/ASP, reinduction, 2xVM26/AC, 2xCYCLO/AC) and maintenance (6MP/MTX). In Study 06/99 treatment of T-ALL was risk adapted with a shortened, intensified 8drug induction (CNSRAD in all but MEDRAD only in pts with residual MedTu after induction) followed by consolidation I (HDAC/HDMTX/VP16). Thy-T was then treated as standard risk with 6x consolidation (3xHDMTX/ASP,reinduction,VM26/AC,CYCLO/AC). E-T and M-T were considered as high risk and scheduled for stem cell transplantation (SCT) in CR1.

Results: In Study 05/93 the CR rate in 291 pts was 89% (94%, 73% and 90% for Thy-T, E-T and M-T; p=.0006) and even 97% for Thy-T in adolescents (15–25 yrs) . Overall 4% failed to achieve CR and 7% died in induction. The probability of continuous CR (CCR) at 5 yrs was overall 53% and 64% for Thy-T, but only 30% (p<.0001) for E-T and M-T. The survival of CR pts (S) was overall 44% and 61%, 19% and 28% (p<.0001) for Thy-T, E-T and M-T. 10% died in CR mainly due to infections - particularly pts with prolonged cytopenias after prior MEDRAD. Immunophenotype was the only prognostic factor confirmed in multivariate analysis. This led to a risk adapted strategy in the subsequent study 06/99. The CR rate in 212 pts was again high (85% overall and 90%, 80% and 76% for subtypes; p=.04). 10% failed to achieve CR and 5% died in induction. 68% of pts with early/mature T-ALL received SCT in CR1 (13 allo sibling, 25 MUD, 11 auto). In these high-risk pts survival after SCT was surprisingly good with 64% at 3 yrs. Survival of all CR pts at 3 yrs was 57% (69% for Thy-T and 50%/34% for E-T/M-T; p=.007). Death in CR did not increase despite SCT (9% overall; 7% after SCT).

Conclusion: In a very large cohort of prospectively treated pts excellent results for adult T-ALL could be achieved with subtype adapted strategies. For early/mature T-ALL a doubling of survival of CR pts from 23% to 43% in studis 05/93 versus 06/99 was achieved by SCT. Thus this is the first study showing a substantial benefit of SCT in these subtypes of T-ALL. To improve outcome further, new drugs for T-ALL (e.g. Campath, Cladribine, Nelarabine) are evaluated in relapse pts and in the ongoing GMALL study additional pts for SCT are identified by prospective analysis of minimal residual disease (supported by Deutsche Krebshilfe and partly BMBF, Grant No 01GI 9971)