A Bortezomib-Based Regimen Offers Excellent Outcomes In Myeloablative HLA-Mismatched and Unrelated Donor Transplantation: Phase II Results

In a phase I/II trial of a novel bortezomib-based regimen for reduced-intensity conditioning (RIC) hematopoietic stem cell transplantation (HSCT) recipients of HLA-mismatched peripheral blood stem cell (PBSC) grafts, we documented low rates of graft-versus-host disease (GVHD) and non-relapse mortality (NRM), with promising survival. In registry analyses, myeloablative conditioning (MAC) HSCT recipients of both HLA-matched (MUD) and 1-locus mismatched donor (MMUD) grafts also have impaired outcomes, with day +100 grade III-IV acute GVHD rates of 28% and 37% respectively, 1-year NRM of 36% and 45% respectively, 1-year progression-free (PFS) of 47% and 38% respectively, and 1-year overall survival (OS) of 52% and 43% respectively. We therefore evaluated a similar bortezomib-based regimen in MAC HSCT recipients lacking 8/8 HLA-matched (-A, -B, -C, -DRB1) related donors.

In a prospective single-arm phase II trial, we enrolled patients with hematologic malignancies, aged 18-60 years, receiving MUD, MMUD, or mismatched related donor (MMRD) grafts. Myeloablative conditioning was IV busulfan (130 mg/m², without PK dose adjustment) and fludarabine (40 mg/m²) once daily for 4 doses (days -7 to -4). T-replete PBSC grafts with ≥ 2x10⁶ CD34+ cells/kg were infused on day 0. GVHD prophylaxis comprised bortezomib (1.3 mg/m² IV on days +1, +4, +7), methotrexate (15 mg/m² IV on day +1, 10 mg/m² on days +3, +6, +11) and tacrolimus from day -3, with a planned taper starting day +100 and complete by day +180. The primary endpoint was day +100 acute GVHD incidence. Secondary endpoints included NRM, relapse, PFS, OS and chronic GVHD at 1 year.

The 34 patients (19 male, 15 female), accrued between March 2011 and November 2012, had a median age of 49 years (range, 21-60) and variable diagnoses (17 AML, 6 MDS, 4 NHL, 3 MPD, 2 ALL, 1 CML, 1 MM) and disease risk indices (Low 1, Intermediate 24, High 9). They received 8/8 MUD (n=14), 7/8 MMUD (n=18) or 7/8 MMRD (n=2) PBSC grafts. Mismatches (16 antigen-, 4 allele-level) involved HLA-A (9), -B (1), -C (6) and -DRB1 (4). The median follow up in survivors is 20 months (range, 7.2-25.5).

The regimen was feasible and well tolerated. Mucositis was noted in the MAC recipients, but no doses were missed due to toxicity. Excluding 1 patient who died of sepsis prior to engraftment, neutrophil and platelet engraftment was prompt, at a median of 14 (range, 3-33) and 17 (range, 7-54) days respectively. Median day +30 donor chimerism was 99% (range, 90-100). Grade II-IV acute GVHD incidence by day +100 and +180 was 32% and 36% respectively, but only 18% and 21% respectively if upper GI GVHD (which had excellent long term outcomes) was excluded. Grade III-IV acute GVHD incidence by day +100 and +180 was 12%. 2-year cumulative incidence of NRM and relapse was 8.8% and 5.9% respectively. 2-year PFS and OS was 85% and 84% respectively (Figure). 2-year cumulative incidence of extensive chronic GVHD was 57%. For 8/8 MUD vs. 7/8 MMRD/MMUD, grade II-IV acute GVHD by day +180 was 30% vs. 40% (p=0.47) (excluding upper GI GVHD, 16% vs. 25%, p=0.42), and grade III-IV acute GVHD was 7% vs. 15% (p=0.48) respectively.

Figure

View largeDownload slide

Study Survival

Figure

View largeDownload slide

Study Survival

Close modal

In conclusion, bortezomib-based prophylaxis for MAC HSCT recipients of HLA-mismatched and unrelated donor grafts was safe and well-tolerated, with low rates of severe acute GVHD, NRM and relapse, and excellent long-term survival. On preliminary landmark analysis, upper GI GVHD did not appear to impair transplantation outcomes. Bortezomib-based prophylaxis is suitable for prospective randomized evaluation in myeloablative transplantation recipients lacking HLA-matched related donors.