Selection of Preparative Regimens for Haploidentical Bone Marrow Transplantation With Post-transplant Cyclophosphamide: Age Is Not Just a Number

Human leukocyte antigen (HLA) –haploidentical hematopoietic stem cell transplantation (HSCT) has become one of the most commonly employed alternative donor transplant techniques since the optimization of post-transplant cyclophosphamide (PTCy) to control the potent immune responses that accompany an HLA-mismatched allogeneic donor.¹  The safety of this type of transplant when used with nonmyeloablative conditioning regimens has been demonstrated in children and adult patients aged up to 75 years.^2-4  Furthermore, several large, registry-based retrospective studies have confirmed the efficacy of haplo–bone marrow transplantation (BMT) with PTCy, achieving results comparable to those of HLA-matched HCT.^5,6  The majority of the data for this transplantation platform is associated with a reduced-intensity conditioning approach that either consists of fludarabine/cyclophosphamide (Cy) and low-dose total body irradiation (TBI). Investigators have explored higher intensity conditioning regimens,^7-11  but no consensus has been reached regarding the optimal myeloablative conditioning regimen (MAC) for haplo-HSCT with PTCy for patients with high-risk hematologic malignancies.

In the current article, Dr. Heather Symons and colleagues report the results of 96 children and adult patients (median age, 42 years; range, 1-65 years) with high-risk hematologic malignancies in remission who were treated at Johns Hopkins University School of Medicine on a prospective phase 2 study of haplo-BMT following a MAC regimen of busulfan (Bu) and Cy or Cy and TBI. The trial was conducted from 2008 until 2014 and enrolled patients with acute myeloid leukemia (AML; 43%) and acute lymphoblastic leukemia (21%) in first- and second-morphologic complete remission (CR), myelodysplastic syndrome (11.5%), mixed-lineage leukemia (5%), lymphoma (11.5%), chronic myeloid leukemia (4%), and other (4%). Patients with AML or lymphoma received a Bu/Cy conditioning regimen, and patients with acute lymphoblastic leukemia or lymphoblastic lymphoma received a Cy/TBI preparative regimen. Allograft source was unmanipulated bone marrow from an HLA-haploidentical (matched for at least one allele each of HLA-A, HLA-B, HLA-C, HLA-DRB1, and HLA-DQB1) first-degree relative (full or half sibling, parent, or child) with a target cell dose of 4 × 10⁸ marrow nucleated cells per kilogram of recipient ideal or actual body weight, whichever was lower. The median donor age was 37 years (range, 11-77 years), and the median number of HLA allele mismatches was 4/10 (range, 2-5). See Figure for treatment schema.

The primary objective of the trial was to characterize donor engraftment at day 60 after MAC haplo-BMT with PTCy; the researchers hypothesized that graft failure would not exceed 10 percent. Primary graft failure was defined as less than 5 percent donor chimerism measured on day 60. Secondary endpoints included non-relapse mortality (NRM), acute graft-versus-host disease (aGVHD), chronic graft-versus-host disease (cGVHD), event-free survival (EFS), and overall survival (OS).

The study met its primary endpoint. Engraftment of donor cells with donor chimerism higher than 95 percent at day 60 was achieved in 80 (91%) of 88 evaluable patients (95% CI, 83, 96), and thus a graft failure of 9 percent. Of the patients who did not achieve higher than 95 percent donor chimerism, four patients had AML and one patient had myelodysplastic syndrome. Interestingly, all five patients were older than 55 years and had received a Bu/Cy preparative regimen. The median time to neutrophil and platelet recovery was 24 and 29 days, respectively. The cumulative incidence of neutrophil and platelet recovery at day 28 was 76 percent (95% CI, 0.7, 0.9) and 46 percent (95% CI, 0.9, 1), respectively. The overall incidence of neutrophil and platelet recovery was 91 percent (95% CI, 0.9, 1) and 87 percent (95% CI, 0.8, 1), respectively.

Regarding the main secondary endpoints, the cumulative incidence of aGVHD grades II to IV and grades III to IV at day 100 for the full cohort was 11 percent (95% CI, 0.05, 0.18) and 4 percent (95% CI, 0, 0.08), respectively. The cumulative incidence of cGVHD at six and 12 months was 4 percent (95% CI, 0, 0.08) and 15 percent (95% CI, 0.08, 0.22). Six percent of patients had moderate to severe cGVHD.

A subset analysis of the adolescent and young adult population, defined in this study as one to 24 years, revealed the cumulative incidence of aGVHD grades II to IV and grades III to IV at eight months was 17 percent and 4 percent, respectively. The cumulative incidence of cGVHD at 12 months was 28 percent (95% CI, 0.01, 0.3), and 14 percent (95% CI, 0.01, 0.3) had moderate to severe cGVHD.

The Table contains selected NRM and survival data. Multivariate analysis confirmed that age was associated with inferior EFS and OS and identified donor or recipient cytomegalovirus seropositivity as associated with inferior EFS and OS. The cumulative incidence of relapse at one year was 35 percent (95% CI, 0.26, 0.45) and 43 percent (95% CI, 0.33, 0.53) at three years. Again, multivariate analysis revealed that relapse was associated with increased age (hazard ratio [HR], 5.4 [95% CI, 1.4, 21.4; p=0.02]) for patients 20 to 55 years and HR of 5.5 (95% CI, 1.3, 23.3; p=0.02) for patients older than 55 years, and detection of minimal residual disease before transplantation (HR, 1.9; 95% CI, 1, 3.4; p=0.02).

Finally, the incidence of bacterial, fungal, and viral infections that occurred in the post-transplant period (within 180 days) appears similar to other published data and is described in detail in the manuscript, for the interested reader. Deaths due to infection in the posttransplant period were low; one child, aged 22 months, died from Gram-negative sepsis and disseminated adenovirus; one adult patient, aged 53 years, died from rhinocerebral fungal infection; and one adult patient, aged 65 years, died from polymicrobial infection with multiorgan system failure.