Outcome of Patients with Hematologic or Molecular Relapsed/Refractory (R/R) Acute T- Lymphoblastic Leukemia - Data from the GMALL Study Group

Background

Outcomes of newly diagnosed T-ALL has improved considerably (Gökbuget et al, ASH 2023). However, options in patients (pts) with hematological (hem) refractory/relapsed (R) disease are limited and outcomes are poor, whereas the prognosis of molecular (mol) R/R disease is largely unknown. One goal for treatment optimization is to identify poor responders and upcoming relapses by regular measurement of minimal residual disease (MRD). Furthermore, integration of potential new compounds is urgently required. Since randomized trials in R/R T-ALL are hampered by the rarity of the disease, outcomes of current standard therapies must be evaluated and compared retrospectively.

Aims

We analyzed data from 171 adult pts with either hem or mol R disease during/after pediatric based first-line therapies. Data were prospectively collected between 2011 and 2023 within the nationwide GMALL registry from 73 hospitals, being representative for the current standard of care in Germany. The major aim was to analyze outcomes for different subtypes of T-ALL, frequencies of hem/mol relapses, identification of prognostic factors for complete remission (CR) and overall survival (OS), and to evaluate the impact of stem cell transplantation (SCT). In hemR the CR rate was assessed according to standard criteria whereas molecular failure (molFail) and molecular relapse (molRel) was defined as MRD >=10-4 after induction/consolidation or after prior achievement of molCR (MRD neg with sensitivity of 10-4). MRD was measured by quantitative PCR of clonal IG/TR in the central GMALL reference laboratory. We focused on the results of first salvage of hemR and molR under chemotherapy, which was given according to physicians’ choice.

Results

101 pts with first hemR ALL were evaluable. Median age was 30 (18–54) yrs. 74 pts had early (< 18 mo of 1st remission duration) and 27 late relapse. The phenotypes were distributed as follows (N=93): 47 early T (51%), 39 thymic T (42%), 7 mature T (8%). The CR rate after first salvage was 48% and lower in early vs late relapse (42% vs 64%; p>.05). CR rates were different for subtypes (43%, 47% and 60% for early, thymic and mature; p>.05). The most frequent regimens in early relapse were induction (N=11) with 36% or Nelarabin+/- Cyclo (N=31) with 42% CR rate resp. In late relapses most frequently, induction was repeated (N=12) yielding a CR rate of 75%.

130 patients with molR ALL (molFail 61% or molRel 39%) were evaluable, with a median age of 32 (18-55) years. 58%, 34% or 8% had early, thymic or mature T-ALL resp. The molCR rate after first salvage was 29% and with no difference between molFail/molRel (29% vs 30%). The most frequently used regimen was Nelarabine (N=49) with a molCR rate of 20%. Response rates for subtypes were 27% for early, 20% for thymic and 0% for mature.

Prognostic factors for OS were analyzed in 101 pts with hemR. OS at 5 yrs was superior in late vs early relapse (29% vs 10%; p=0.02), similar in pts aged 15–25 vs 26–45 vs 46–55 yrs (19% vs 13% vs 17%; ns) and superior in thymic vs early or mature T-ALL (23% vs 8% vs 0%; p>.05). 30% of evaluable pts received SCT in any stage after relapse. Their survival rate at 5 yrs was 28% vs 3% without SCT (p<0.0001). Survival was significantly better if SCT was performed in CR after 1st salvage vs later CR vs no CR (42% vs 31% vs 9%; p=0.02). Comparing different periods (2011-2016 and 2017-2023) there was a slight improvement in 5y-OS (11% vs 19%, >.05).

For molFail/molRel the 5y-OS was 39%. OS was significantly superior in pts with SCT (N=76; 56%) compared to those without SCT (N=20; 8%) (p<0.0001). OS at 5 yrs was significantly superior after molR (33%) compared to hemR (15%) (p=0.0003). Analyses of later salvages, different regimens, relapse localizations and older pts will be presented.

Summary

Survival of pts with T-ALL and hemR disease during/after chemotherapy was 15% in the prospectively analyzed GMALL cohort. Over the yrs survival improved only slightly. Outcome was dependent on phenotype, time to relapse and SCT. In addition, OS was significantly superior for molR vs hemR underlining the need to identify upcoming relapse by stringent MRD testing. Nelarabine however appeared not be effective in molFail/molRel T-ALL. Our analysis provides reference for future trials with new compounds in hemR/R or molR/R T-ALL.