Long-Term Results of a Sequential Hematopoietic Stem Cell Transplantation Protocol Based on FLAG-Ida and High-Dose Melphalan for Non-Remission High-Risk Myeloid Malignancies

Introduction

Despite significant recent progress in acute myeloid leukemia (AML) therapy, a subgroup of patients failing to achieve a response before allogeneic stem cell transplantation (SCT) or those with early relapse remain a challenge. SCT with sequential conditioning regimen (e.g., FLAMSA-RIC, Schmid C. et al. 2005) was proposed in the early 2000s for this subgroup of patients, as a strategy to induce a rapid reduction of tumor burden to allow engraftment and the potential benefit of graft versus leukemia effect.

The CETLAM group designed a specific sequential protocol that combined FLAG-Ida regimen with high-dose Melphalan (FLAG-Ida/Mel), based on its capability to establish a rapid complete donor chimerism in the setting of high-risk myeloid malignancies. We report the long-term results and predictors of outcome with this strategy.

Methods

We performed a retrospective study, including all patients of the CETLAM group who received the sequential FLAG-Ida/Mel SCT protocol (n=138) between 2005 and 2019. The protocol consisted of a cytoreductive phase based on FLAG-Ida (fludarabine 30 mg/m², cytarabine 2 g/m² and granulocyte-colonies stimulating factor 300 µg from day -11 to -7; idarubicin 12 mg/m², days -11, -9 and -7) shortly followed by a reduced-intensity conditioning phase with Melphalan (70 mg/m², days -3 and -2). Graft versus host disease (GvHD) prophylaxis was based on cyclosporine and mofetil mycophenolate. Antithymocyte globulin (ATG) was generally administered in cases with one mismatch in HLA. Some more recent cases included post-transplantation cyclophosphamide for non-related and haploidentical donors. All patients gave informed consent for this study. Statistical analysis was performed with R (3.6.3).

Results

Median follow-up for survivors was 59 months. Patient baseline characteristics are summarized in Table 1. Notably, high-risk features such as secondary AML, high-risk cytogenetics and relapse after allogeneic SCT were present in 40%, 44%, and 17% of patients.

Median time to neutrophil engraftment was 15 days (range 7-44), being reached by 97% of surviving patients at day +30. Platelet recovery was reached in a median time of 14 days (range 8-100). Seven patients died before response was evaluated (early mortality rate of 5%). Only 8 patients progressed despite therapy, which resulted in a 94% remission rate at day +30 after transplantation. In congruence with this, full donor chimerism was reached by 86% of patients at day +30. Cumulative incidence of acute GvHD grade III to IV by day +100 after alloSCT was 31% (CI 95%, 23% to 38%), considering death as a competing event. Severe chronic GvHD at 2-year occurred in 7% (CI 95% 3% to 11%).

Relapse cumulative incidence at 1 and 5 years were 23% (95% CI, 16% to 30%) and 29% (95% CI, 21% to 38%). Non-relapse mortality cumulative incidence at 1 and 5 years were 42% (95% CI, 33% to 50%) and 45% (95% CI, 37% to 54%) respectively (Figure 1). Overall survival (OS) at 1 and 5 years was 39% (95% CI, 31% to 47%) and 26% (95% CI, 24% to 40%), respectively. OS, Leukemia-free survival (LFS) and graft versus host disease-free, relapse-free survival (GRFS) at 1, 2 and 5-year are summarized in Table 2 and Figure 2.

In the univariate analysis 4 variables showed a negative impact on overall survival: age >50 years, circulating blast count >10%, primary refractoriness and adverse-risk cytogenetics. In multivariate Cox model, only pre-transplant peripheral blood blast count retained significance, whereas age showed a border-line impact.