Contributions
Abstract: S872
Type: Oral Presentation
Presentation during EHA23: On Saturday, June 16, 2018 from 16:15 - 16:30
Location: Room A6
Background
The efficacy of allogeneic hematopoietic cell transplantation (HCT) in patients (pts) with acute myeloid leukemia (AML) depends both on chemo/radiotherapy given in the conditioning regimen and on immune-mediated graft-versus-leukemia (GVL) effects. Previous studies have observed an association between occurrence of acute (a) and chronic (c) graft-versus-host disease (GVHD) and a lower risk of AML relapse in pts with de novo AML (GvL effects).
Aims
It is well established that, in comparison to de novo AML, secondary AML (sAML) is less sensitive to chemotherapy. However, its susceptibility to GvL effects has not been studied in large cohorts of pts. This is the focus of the current study.
Methods
The study population included adult pts with sAML in first or second CR, bone marrow (BM) or peripheral blood stem cells (PBSC) as stem cell source, HLA-identical sibling or 10/ 10 (MUD) or 9/10 (MMUD) HLA-matched unrelated donor, no ex vivo T-cell depletion, and transplantation between 2005 and 2016. Previous studies might have overestimated the beneficial impact of GVHD on relapse prevention given the tight association between GVHD and non-relapse mortality (NRM) (since relapse and NRM are competing events). Thus, here we assessed the evolution of relapse rate over time according to GVHD condition by calculating the relapse rate per pt-year within sequential 90-day intervals, as previously reported (Baron et al., J Intern Med 2018, 283(2):178-18). The smoothed rates were plotted as curves for each GVHD condition. The impact of GVHD on relapse, NRM and overall survival (OS) was also assessed with classical multivariate Cox models modeling GVHD as time-dependent covariates.
Results
3303 pts met the study inclusion criteria. Status at transplantation was CR1 in 2919 pts (88%) and CR2 in the remaining 384 ps. 1517 pts received grafts from MSD, 1427 from MUD and 359 from MMUD. 41% of the pts received a myeloablative conditioning (MAC), and 49% anti-thymocyte globulin (ATG). Stem cell source was PBSC in 90% of the patients. The proportion of patients with grade II and III-IV aGVHD was 15% and 10%, respectively. At 2 years, the cumulative incidence of cGVHD was 43% (18% extensive), the cumulative incidence of relapse 30% and OS 53%. Relapse rates declined gradually over time and were significantly lower in pts with cGVHD than in those without (P=0.009). In multivariate Cox models, grade III-IV aGVHD and cGVHD were each associated with a lower risk of relapse but not grade II aGVHD. Other factors associated with a lower risk of relapse included absence of poor-risk cytogenetic (P=0.008), MAC versus reduced-intensity (RIC) conditioning (P<0.001), and MUD (P=0.01). Interestingly, while there was a trend for better OS in pts with limited cGVHD (P=0.07), all other forms of GVHD were associated with lower OS. Other factors associated with increased mortality included high pt age (P<0.001), intermediate (P=0.01) and poor (P<0.001) risk cytogenetics, and MMUD (P=0.01).
Conclusion
Although this study showed that limited cGVHD was associated with lower risk of sAML relapse and a trend to a better OS, all other form of GVHD were associated with lower OS. In addition, while MUD recipients had a lower risk of relapse, they had higher NRM offsetting any potential OS benefit. Further studies should focus at optimizing the conditioning regimens and administering post-transplant maintenance therapies in an attempt to further reduce sAML relapse post HCT.
Session topic: 23. Stem cell transplantation - Clinical
Keyword(s): Acute Myeloid Leukemia, Graft-versus-host disease (GVHD), Graft-versus-tumor effect, Secondary
Abstract: S872
Type: Oral Presentation
Presentation during EHA23: On Saturday, June 16, 2018 from 16:15 - 16:30
Location: Room A6
Background
The efficacy of allogeneic hematopoietic cell transplantation (HCT) in patients (pts) with acute myeloid leukemia (AML) depends both on chemo/radiotherapy given in the conditioning regimen and on immune-mediated graft-versus-leukemia (GVL) effects. Previous studies have observed an association between occurrence of acute (a) and chronic (c) graft-versus-host disease (GVHD) and a lower risk of AML relapse in pts with de novo AML (GvL effects).
Aims
It is well established that, in comparison to de novo AML, secondary AML (sAML) is less sensitive to chemotherapy. However, its susceptibility to GvL effects has not been studied in large cohorts of pts. This is the focus of the current study.
Methods
The study population included adult pts with sAML in first or second CR, bone marrow (BM) or peripheral blood stem cells (PBSC) as stem cell source, HLA-identical sibling or 10/ 10 (MUD) or 9/10 (MMUD) HLA-matched unrelated donor, no ex vivo T-cell depletion, and transplantation between 2005 and 2016. Previous studies might have overestimated the beneficial impact of GVHD on relapse prevention given the tight association between GVHD and non-relapse mortality (NRM) (since relapse and NRM are competing events). Thus, here we assessed the evolution of relapse rate over time according to GVHD condition by calculating the relapse rate per pt-year within sequential 90-day intervals, as previously reported (Baron et al., J Intern Med 2018, 283(2):178-18). The smoothed rates were plotted as curves for each GVHD condition. The impact of GVHD on relapse, NRM and overall survival (OS) was also assessed with classical multivariate Cox models modeling GVHD as time-dependent covariates.
Results
3303 pts met the study inclusion criteria. Status at transplantation was CR1 in 2919 pts (88%) and CR2 in the remaining 384 ps. 1517 pts received grafts from MSD, 1427 from MUD and 359 from MMUD. 41% of the pts received a myeloablative conditioning (MAC), and 49% anti-thymocyte globulin (ATG). Stem cell source was PBSC in 90% of the patients. The proportion of patients with grade II and III-IV aGVHD was 15% and 10%, respectively. At 2 years, the cumulative incidence of cGVHD was 43% (18% extensive), the cumulative incidence of relapse 30% and OS 53%. Relapse rates declined gradually over time and were significantly lower in pts with cGVHD than in those without (P=0.009). In multivariate Cox models, grade III-IV aGVHD and cGVHD were each associated with a lower risk of relapse but not grade II aGVHD. Other factors associated with a lower risk of relapse included absence of poor-risk cytogenetic (P=0.008), MAC versus reduced-intensity (RIC) conditioning (P<0.001), and MUD (P=0.01). Interestingly, while there was a trend for better OS in pts with limited cGVHD (P=0.07), all other forms of GVHD were associated with lower OS. Other factors associated with increased mortality included high pt age (P<0.001), intermediate (P=0.01) and poor (P<0.001) risk cytogenetics, and MMUD (P=0.01).
Conclusion
Although this study showed that limited cGVHD was associated with lower risk of sAML relapse and a trend to a better OS, all other form of GVHD were associated with lower OS. In addition, while MUD recipients had a lower risk of relapse, they had higher NRM offsetting any potential OS benefit. Further studies should focus at optimizing the conditioning regimens and administering post-transplant maintenance therapies in an attempt to further reduce sAML relapse post HCT.
Session topic: 23. Stem cell transplantation - Clinical
Keyword(s): Acute Myeloid Leukemia, Graft-versus-host disease (GVHD), Graft-versus-tumor effect, Secondary