Thiotepa–busulfan–fludarabine conditioning regimen in myelofibrosis

For patients with myelofibrosis (MF), allogeneic hematopoietic cell transplantation (allo-HCT) is still the only potential curative approach, but no consensus exists on the optimal conditioning regimen or its intensity.¹

Thiotepa, an alkylating agent, in combination with busulfan and fludarabine (TBF) seems a valid conditioning regimen for patients with a range of hematologic malignancies undergoing allo-HCT, but data on its use for MF are limited to small, single-center series.¹

Recently, results from a retrospective, multi-center, European Society for Blood and Marrow Transplantation (EBMT) registry-based study have been published in Bone Marrow Transplantation.¹ This study evaluated the outcomes of adult patients with MF who underwent TBF-conditioned allo-HCT.

Study design

Patients included in the study were aged ≥18 years and underwent TBF-conditioned first allo-HCT for MF between 2010 and 2017. All donor types were included. The authors reported the outcomes of 187 patients.

A performance status (PS) <90% was defined as poor, according to the Karnofsky Performance scale. Neutrophil engraftment was positive when the absolute neutrophil count was ≥0.5 × 10⁹/L for 3 days, consecutively. Similarly, platelet engraftment required an absolute platelet count ≥20 × 10⁹/L for 3 days, consecutively.

Patient characteristics

Key patient-, disease-, and transplant characteristics are reported in Table 1. The majority of donors were haploidentical (41%), and the stem cell source was peripheral blood in 60% of cases. Reduced-intensity conditioning was used in 52% of cases, and in vivo T-cell depletion using antithymocyte globulin occurred in 41% of cases.

Table 1. Patient characteristics*

Results

The transplant outcomes are shown in Table 2. The median follow-up for survival outcomes was 29 months (interquartile range, 23–47).

Table 2. Main transplant outcomes*

Engraftment

The median time to neutrophil engraftment was 21 days (95% confidence interval [CI], 20–22) and, according to donor type, it took longest in patients transplanted from haploidentical donors (23 days) compared with those transplanted from human leukocyte antigen (HLA)-identical donors (20 days) or unrelated donors (19 days). The median time to platelet engraftment was 34 (95% CI, 31–49) days.

Results from the univariate analysis showed that lower rates of 100-day neutrophil engraftment were seen

• after haploidentical HCT versus HLA-identical sibling or unrelated donor HCT (84% vs 92% vs 92%, respectively; p < 0.01);
• with the use of bone marrow (BM) versus peripheral blood (PB) as stem cell source (89% vs 91%; p = 0.01); and
• when using posttransplant cyclophosphamide (PTCy) as GvHD prophylaxis versus antithymocyte globulin or other prophylaxis strategies (84% vs 92% vs 93%; p < 0.01).

GvHD

Results from the univariate analysis showed the following:

• The 100-day incidence of Grade II–IV acute graft-versus-host disease (aGvHD) was higher after allo-HCT from unrelated donors versus HLA-identical sibling or haploidentical donors (38% vs 13% vs 17%; p < 0.01)
• The 100-day incidence of Grade III–IV aGvHD was higher
  • after allo-HCT from unrelated donors versus HLA-identical sibling or haploidentical donors (21% vs 11% vs 4%; p = 0.02); and
  • with the use of PB versus BM as stem cell source (16% vs 6%; p = 0.05).
• The incidence of 3-year chronic GvHD (cGvHD) was higher
  • in patients transplanted after an interval from diagnosis to allo-HCT ≤44 months versus those transplanted in ˃44 months (49% vs 28%; p = 0.02);
  • in patients receiving a dose of thiotepa of 10 mg/kg versus those receiving 5 mg/kg (42% vs 31%; p = 0.04);
  • with the use of BM versus PB as stem cell source (44% vs 35%; p = 0.05); and
  • in patients with primary MF versus secondary MF (45% vs 22%; p < 0.01).

The multivariate analysis confirmed that receiving a graft from an unrelated donor compared with an HLA-identical sibling was associated with a higher risk of Grade II–IV aGvHD (hazard ratio [HR], 3.22; 95% CI, 1.11–9.31; p = 0.03). No significant differences in GvHD incidence were observed with respect to recipient age, conditioning intensity, Karnofsky PS, stem cell graft source, or GvHD prophylaxis.

Survival outcomes

The univariate analysis for the main survival outcomes showed that

• patients aged ≤58 years versus >58 years (63% [95% CI, 53─73] vs 46% [95% CI, 34─57]; p = 0.03) and a good versus poor Karnofsky PS (62% [95% CI, 52─72] vs 44% [95% CI, 31─57]; p = 0.02) were associated with higher probability of overall survival; and
• a good versus poor Karnofsky PS (48% [95% CI, 36–59] versus 38% [95% CI, 24–51]; p = 0.04) was associated with a better graft-versus-host disease relapse-free survival (GRFS).

Results from the multivariate analysis confirmed that a good Karnofsky PS was associated with superior GRFS (HR 0.62, [95% CI, 0.39–1.01]; p = 0.05). No significant differences in survival outcomes were observed according to donor type, recipient age, conditioning intensity, stem cell graft source, or GvHD prophylaxis.

At last follow-up, 77 patients had died, with the main causes being

• infections, 36% (n = 28);
• GvHD, 21% (n = 16); and
• multiple organ failure, 14% (n = 11).

Conclusion

Results from this study showed a delayed neutrophil engraftment in transplanted patients with MF, with a cumulative incidence of neutrophil engraftment of 71% at 28 days, increasing to 91% at 100 days.

A higher risk of Grade II–IV aGvHD was observed in patients receiving a graft from an unrelated donor compared with an HLA-identical sibling in both uni- and multivariate analyses.

A poor Karnofsky PS at time of allo-HCT was associated with a worse GRFS in both uni- and multivariate analyses, highlighting the importance of early referral for allo-HCT for high-risk patients with MF.

No statistically significant differences were observed in the main transplant outcomes according to conditioning intensity, suggesting that TBF with reduced-intensity conditioning is a potential alternative to consider, especially for more fragile patients.

The retrospective nature of the study and the small sample size represent potential limitations.