Effect of conditioning regimen intensity after haploidentical transplantation

T-cell replete haploidentical donor transplants are increasingly conducted, usually with post-transplant cyclophosphamide (PTCy) plus a calcineurin inhibitor, and mycophenolate mofetil (MMF) for graft­-versus-host disease (GvHD) prophylaxis. To date, this strategy seems to be an effective alternative in cases where human leukocyte antigen (HLA)-matched sibling donors are unavailable.¹

Retrospective studies in HLA-matched and unrelated donor transplants have compared post-transplant outcomes following myeloablative conditioning (MAC) and reduced-intensity conditioning (RIC) regimens, however all but one did not show a difference in disease free survival (DFS) or overall survival (OS). Other randomized studies comparing MAC to RIC in hematological malignancies have yielded mixed results, with no randomized trials or large retrospective cohort reports currently available.

To ascertain the impact of conditioning regimen intensity on the outcomes of T-cell replete haploidentical transplant in the United States (US), Scott R. Solomon, The Blood & Marrow Transplant Group, Northside Hospital, Atlanta, GA, US, and colleagues, conducted a retrospective analysis on the efficacy and safety of MAC versus RIC following T cell-replete haploidentical transplant in patients with hematological malignancies. The results of this study were recently published in Blood Advances and summarized here.

Study design and patient characteristics

• Data obtained from the Center for International Blood and Marrow Transplant Research (CIBMTR)
  • Of the 195 US centers in the CIBMTR, 78 provided data for this analysis
  • Transplants occurred between 2008 and 2016
• Patients were aged 18–70
• Hematological malignancies included:
  • Acute myeloid leukemia (AML); n= 818
  • Acute lymphoblastic leukemia (ALL); n= 286
  • Myelodysplastic syndrome (MDS); n= 221
• T-cell replete haploidentical transplant from bone marrow (BM) or peripheral blood (PB)
  • Haploidentical relative donor, mismatched at two or more HLA loci
• GvHD prophylaxis was PTCy with a calcineurin inhibitor and MMF
• Intensity of conditioning regimen was compared:
  • MAC; n= 526
    • MAC defined as; total body irradiation (TBI) dose >800 cGy, busulfan >8mg/kg (oral) or >6mg/kg (intravenous [IV]), melphalan >150mg/m² or melphalan 140mg/m² with thiotepa
    • MAC regimens were most commonly non-TBI containing
    • Most included; busulfan and cyclophosphamide (14%), busulfan and fludarabine (13%) or busulfan, cyclophosphamide and fludarabine (22%)
    • Use of MAC regimens decreased with age
    • Patients receiving MAC were younger, more likely to be female, had a Karnofsky performance score (KPS) ≤80, had ALL, were in relapse at transplant, had a high disease risk index, and received PB
  • RIC; n= 799
    • RIC defined as; TBI dose 200 or 300 cGy, melphalan ≤140mg/m² without another alkylating agent, or busulfan ≤8mg/kg (oral) or ≤6mg/kg (IV)
    • Most common RIC regimen was low-dose TBI (200 cGy), cyclophosphamide and fludarabine (84%)
    • RIC more common in those aged 55–70 (77%) compared to MAC (23%)
• Median follow-up: 24 months in both arms
• Primary endpoint: DFS

Results

A | Patients aged 18–54

After adjusting for comorbidity score, graft type, disease and disease risk index (Tables 1 and 2):

• DFS: lower with RIC
• Relapse: higher with RIC
• Non-relapse mortality (NRM): did not differ by conditioning regimen
• Acute graft-versus-host disease (GvHD, aGvHD) grade II–IV: no difference by conditioning intensity
  • HR= 1.01, 95% CI: 0.79–1.29, p= 0.94
• Chronic GvHD (cGvHD): no difference by conditioning intensity
  • HR= 0.82, 95% CI: 0.62–1.07, p= 0.14

Table 1. Effect of conditioning regimen on DFS, NRM and OS in patients aged 18–54

Table 2. Three-year adjusted probabilities of DFS, NRM, OS, cGvHD and day 100 grade II–IV aGvHD in patients aged 18–54

B | Patients aged 55–70

After adjusting for comorbidity score, graft type and disease risk index (Tables 3 and 4):

• DFS: did not differ by conditioning regimen
• Relapse: did not differ by conditioning until several years post-transplant
• NRM: lower with RIC
• aGvHD grade II–IV: no difference by conditioning intensity
  • HR= 0.88, 95% CI: 0.64–1.07, p= 0.14
• cGvHD: no difference by conditioning intensity
  • HR= 0.86, 95% CI: 0.59–1.26, p= 0.43

Table 3. Effect of conditioning regimen on DFS, NRM and OS in patients aged 55–70

Table 4. Three-year adjusted probabilities of DFS, NRM, OS, cGvHD and day 100 grade II–IV aGvHD in patients aged 55–70

C | TBI-containing vs non-TBI containing

• DFS: did not differ by conditioning regimen
  • MAC: HR= 1.13, 95% CI: 0.88–1.45, p= 0.35
  • RIC: HR= 1.23, 95% CI: 0.91–1.65, p= 0.17
• Relapse: did not differ by conditioning regimen
  • MAC: HR= 0.98, 95% CI: 0.72–1.35, p= 0.92
  • RIC: HR= 0.90, 95% CI, 0.6–1.33, p= 0.59
• NRM:
  • MAC: did not differ between TBI and non-TBI containing regimens:
    • HR= 1.39, 95% CI: 0.91–2.11, p= 0.13
  • RIC: higher with non-TBI-containing than with TBI-containing regimens
    • HR= 2.14, 95% CI: 1.36–3.37, p= 0.03

D | Overall

• Deaths (MAC vs RIC): 236/526 (45%) vs 405/799 (51%)
  • Cause of death did not differ by conditioning regimen (p= 0.35)
  • Main cause of death: recurrent disease (MAC: 54%; RIC: 61%)
  • Other frequent causes of death: infection, GvHD and organ failure
• Hematopoietic recovery:
  • Median time to neutrophil engraftment and platelet recovery: did not differ by conditioning regimen
    • Median time to neutrophil engraftment: 17 days for both
    • Median time to platelet recovery: 27 days for both
  • One-year incidence of graft failure did not differ by conditioning regimen (p= 0.13)
  • No interaction between conditioning intensity and graft type on graft failure (p= 0.25)
• GvHD rates:
  • Transplantation with PB: higher rates of grade II–IV aGvHD and cGvHD independent of age or regimen intensity:
    • aGvHD: HR= 1.49, 95% CI: 1.23–1.79, p< 0.0001
    • cGvHD: HR= 2.19, 95% CI: 1.76–2.73, p< 0.0001

E | Myeloid malignancies only

• A subgroup analysis was conducted in patients with AML and MDS
• In patients aged 18–54, results were consistent with the total cohort analysis of higher DFS and lower relapse rates with MAC. However, there was a survival advantage with MAC that was not observed in the total cohort analysis
  • OS (MAC vs RIC): HR= 1.35, 95% CI: 1.04–1.77, p= 0.027
• In patients aged 55–70, results were consistent with the main analysis

Conclusion

This retrospective analysis has demonstrated lower relapse and higher DFS rates in patients aged 18–54 who receive MAC regimens compared to RIC, and in patients with myeloid malignancies, MAC regimens conferred a survival advantage. For patients aged 55-70, no difference in DFS and relapse rates were observed with the different conditioning regimens, while NRM mortality was lower in the RIC group. Both aGvHD and cGvHD rates did not differ by conditioning regimen in either age group, with risks being mitigated by graft type.

Some limitations of this study include its retrospective nature, the short follow-up period and a lack of knowledge of all the risk factors that influenced the choice of conditioning regimen and subsequent outcome. Additionally, MAC and RIC were defined broadly, rather than by specific regimen.

However, the study also has many advantages such as the large cohort size, the ability to evaluate outcome in both younger and older patients and the ability to identify differences between TBI- and non-TBI-containing regimens.

This study supports the use of MAC for adults with acceptable comorbidity scores and the ability to tolerate intensive chemotherapy, irrespective of age, to lower relapse rates and improve DFS in the haploidentical transplant setting.