Real-World Patterns of Care and Outcomes for Non-Favourable Risk Acute Myeloid Leukemia Treated with Intensive Chemotherapy across Australia and New Zealand- Data Linkage between the ALLG National Blood Cancer Registry (NBCR) and the Australia and New Zealand Transplant and Cellular Therapy (ANZTCT) Registry

Allogeneic transplant (HSCT) is standard consolidation therapy for patients (pts) with newly diagnosed acute myeloid leukemia (AML) with adverse risk (AR) cytogenetics induced with intensive chemotherapy (IC), but whether this occurs commonly in routine practice is uncertain. The Australasian Leukaemia and Lymphoma Group (ALLG) National Blood Cancer Registry (NBCR) (ANZCTR 12612000337875) prospectively collects longitudinal data from diagnosis on treatment and outcomes for patients with AML from participating centres in Australia and New Zealand. The Australia and New Zealand Transplant and Cellular Therapy (ANZTCT) Registry holds outcome data on all transplants in ANZ.

We established data linkage between the ALLG NBCR and the ANZTCT Registry to describe the outcomes and patterns of treatment particularly HSCT in first response for fit pts with newly diagnosed AML bearing non-favourable cytogenetics across ANZ.

Methods: NBCR AML participants treated with IC were matched to the ANZTCT registry using the date of birth, sex, date of diagnosis (DOD), and diagnosis. Survival analysis was performed using the Kaplan-Meier method from the time of diagnosis or as a landmark from the time of first composite complete response (CRc = CR+CRi+MLFS). The impact of baseline variables, including MRC 2010 cytogenetic risk stratification, on survival was assessed using univariable and multivariable Cox proportional hazards (CoxPH) models, with transplant status treated as a time-dependent variable.

Results: Of the cohort of 600 newly diagnosed AML participants obtained from NBCR between 2013 and 2018, 102 were excluded due to non-intensive treatment, other diagnoses, favourable /no cytogenetics or incomplete data. Of the final analysis cohort of 498 AML pts treated with IC, 386 (78%) had intermediate risk (IR) and 112 (22%) had adverse risk (AR) cytogenetics. 151 (30%) and 71(14%) had NPM1 and FLT3-ITD mutations at diagnosis, respectively.

The median age was 58 years (17-78y). 48% of participants were females. A total of 334 (67%) pts received a standard dose of cytarabine and idarubicin. The remainder received intermediate/high-dose cytarabine-based induction. A total of 368 (74%) patients achieved a composite complete response (CR/CRi/MLFS), with 4% induction death. Major proportion 250 (50%) received a single consolidation chemotherapy post-response and 87 (18%) received 2 or more consolidations.

125 (25%) pts underwent HSCT in the first response (transplant cohort (TC)), whereas 373 pts were treated with IC without HSCT in the first response (No Transplant Cohort (NTC)). Pts in the TC were younger (median age, 50 vs. 60 years, P<0.0001). 82 (66%) TC and 168 (45%) NTC pts received a single cycle of consolidation chemotherapy while 19 (15%) TC and 68 (17%) NTC received two or more consolidations respectively.

80 (23%) and 40 (51%) pts in the IR and AR groups, respectively, received HSCT after the first CRc. Median time to matched sibling and unrelated donor HSCT from CRc was 2.1 (0.7-5.3) and 3.7 (0.7-11.5) months respectively. Seventy-eight (62%) and 42 (34%) pts underwent matched unrelated and sibling donor transplantation, respectively. Sixty-two (50%) received reduced-intensity conditioning. Peripheral blood stem cells were the graft source in 113 pts (90%) and double cords in 5 (4%). In vivo T-cell depletion for GVHD prophylaxis was used in 52 (42%).

At a median follow-up of 64.3 months, the median overall survival (OS) from diagnosis in IR and AR AML pts was 47.6 (33.5-78.7) and 9.4 (7.4-13.7) months, respectively. The median OS landmarked from the first CRc was superior in the TC vs NTC in both the IR (not reached vs 46.6 months, P<0.0001) and AR groups (14.5 vs 9.6 months, P<0.0001). There was no impact of time to HSCT from CRc. In multivariable Cox PH modelling, the hazard ratio (HR) for AR vs. IR was 3.36 (95% C.I. 2.55–4.42), and the HR for TC vs. NTC (as a time-dependent variable) was 0.18 (95% C.I. 0.13–0.26). Analysis of factors influencing whether HSCT was performed is ongoing.

Conclusions

This first AML and HSCT registry data linkage analysis conducted in Australia and New Zealand means we can now reliably track pts undergoing HSCT from diagnosis. Our data represent the first report of the treatment patterns in non-favourable risk AML in ANZ and confirm the superior outcomes with HSCT in both IR and AR AML. Further work is required to improve utilization of HSCT in AML patients to improve survival outcomes.