Decitabine-containing conditioning regimens for allo-HSCT may reduce rates of Grade II–IV aGvHD

Allogeneic hematopoietic stem cell transplant (allo-HSCT) is the only treatment with curative intent for patients with intermediate- or high-risk myelodysplastic syndrome (MDS) or acute myeloid leukemia (AML). However, treatment failure can occur due to relapse and graft-versus-host disease (GvHD). Decitabine is a pyrimidine analogue that inhibits cytosine methylation, resulting in activation of silent genes and thus has an anti-leukemic effect. Since recent studies have implicated epigenetic changes in the pathogenesis of MDS, the use of decitabine, in combination with standard conditioning regimens, may improve the outcome of patients who undergo allo-HSCT. To evaluate the role of decitabine in pre-allo-HSCT conditioning regimens, Qing Ya Wang, Yuan Li and colleagues, Peking University First Hospital, Beijing, CN, conducted a retrospective analysis in patients with intermediate- and high-risk MDS and transformed AML (tAML).

Study design

• Intermediate- and high-risk patients with tAML or MDS were included (N = 76)
  • Decitabine-containing conditioning (dec): n = 40
  • Conditioning without decitabine (non-dec): n = 36
• Patients had undergone allo-HSCT at the Peking University First Hospital between December 2005 and June 2018
  • Patients with high and higher- risk MDS received allo-HSCT immediately after diagnosis whilst patients with tAML received standard “7+3” induction chemotherapy
• Conditioning dosing schedule:
  • Cytarabine: 2g/m²/day for three days
  • Busulfan (Bu): 3.2mg/kg/day intravenously (IV) for three days
  • Cyclophosphamide (Cy): 1.8g/m²/day for two days OR fludarabine: 50mg for 3–6 days
  • Decitabine was administered to the dec group (15mg/m² IV) from Day -16 to -12
  • Rabbit anti-thymocyte globulin (ATG) was given for three days pre-transplant for patients with:
    • Haploidentical or unrelated donors: 7.5–10mg/kg
    • Human leukocyte antigen identical sibling donors: 0–5mg/kg
  • GvHD prophylaxis was given as below:
    • Mycophenolate mofetil: 50mg every 12 hours from the start of conditioning to up to 30 days post-transplant
    • Cyclosporin A: 5mg/kg IV from Day -6 to bowel function recovery, and was then administered orally
    • Methotrexate: 10–15mg/m² IV on Days +1, +3, +5 and +11 post-transplant
  • Stem cell mobilization: granulocyte colony stimulating factor (G-CSF) subcutaneously every 12 hours from three days prior to the collection of bone marrow stem cells. If peripheral blood was the stem cell source, harvesting was conducted on the fourth day after G-CSF was given
  • Antibiotics were used to prevent microbial infections
  • Median follow-up, months (range): 40 (1–155)

Patient characteristics

Given as dec vs non-dec

• Pre-transplant therapy and Revised International Prognostic Scoring System (IPSS-R) categories were comparable between groups
• Differences between the treatment arms were observed with:
  • Age: patients in the dec group were older (41 vs  33, p = 0.028)
  • Mononuclear cells (MNCs): patients in the dec group received a lower number of MNCs (p = 0.007) but a larger number of CD34-positive cells (p = 0.004)
  • Disease status prior to transplant: more patients in the dec group were in complete remission at the time of transplant (13 vs  7)
  • Conditioning: BuCy was used more frequently in the non-dec group (0 vs 5, p = 0.015)
• By the World Health Organization (WHO) classification, patients were diagnosed with:
  • Refractory anemia: 0 vs 1
  • Refractory cytopenia with multi-lineage dysplasia: 8 vs 13
  • Refractory anemia with excess blasts (RAEB)-1: 8 vs 3
  • RAEB-2: 7 vs 5
  • tAML: 17 vs 14
• All patients achieved neutrophil engraftment

Results

Given as dec vs non-dec

• Transplantation-related complications, such as incidence of cytomegalovirus viremia, were comparable between the two arms, aside from the incidence of respiratory infection (22.50% vs78%, p = 0.012)
• Cumulative incidence of Grade II–IV aGvHD: 12.4% (95% CI, 4.9–30.9) vs5% (95% CI, 28.1 vs 61.2), p = 0.005
  • Univariate analysis of factors associated with risk of aGvHD found patients aged > 40 years may be more susceptible (p = 0.018)
  • Multivariate analysis of these factors showed:
    • Decitabine-containing regimens protected against Grade II–IV aGvHD: hazard ratio = 0.279, 95% CI, 0.102–0.765, p = 0.013
    • Age at transplant was not significantly associated with aGvHD in multivariate analysis (p = 0.150 and p =0.180, respectively)
  • Cumulative incidence rate of chronic GvHD (cGvHD): 17.5% vs0% (p = 0.436)

Outcomes

Given as dec vs non-dec

• No significant differences were observed in relation to three-year efficacy endpoints:
  • Overall survival (OS): p = 0.980
  • Disease-free survival (DFS): p = 0.959
  • Relapse rate (RR): p = 0.573
• The median time to relapse was longer in the decitabine group, months (range): 7 (2–12) vs  3 (2–4),    p = 0.171
• There was a tendency for lower relapse rate in high-risk patients with decitabine, but this was not significant (p = 0.085)
• Subgroup analysis in higher-risk IPSS-R patients found that decitabine may:
  • Reduce the RR: 10.6% vs 21.6%, p = 0.370
  • Increase DFS: 72.2% vs 51.1%, p = 0.314
  • Enhance OS: 72.2% vs 64.3%, p = 0.628
  • NRM: 10.8% vs 0%, p < 0.001, however, no conclusions can be drawn due to the small patient numbers in this group

Conclusion

Patients in the dec group tended to be older and with higher risk disease, however a similar OS, DFS and RR was achieved. The authors suggest decitabine may, therefore, ameliorate the unfavorable characteristics of high-risk patients.

Including decitabine in the conditioning regimen prior to allo-HSCT in intermediate- and high-risk patients with MDS or AML decreased the incidence of Grade II–IV aGvHD and respiratory infections in this analysis. The authors hypothesize this is because decitabine exerts its effects via direct cytotoxicity and strengthened tumor immunity, whilst not compromising the GvHD protective effects.

A larger study would be required to confirm these results since the analysis and conclusions are limited by a small, retrospective sample of patients treated at one center, with differences in variables at baseline.