Post-transplant cyclophosphamide and ruxolitinib as GvHD prophylactic regimen in patients with primary or secondary myelofibrosis

In patients with primary and secondary myelofibrosis (MF), allogeneic hematopoietic stem cell transplantation (allo-HSCT) is the only potentially curative treatment, but it is associated with high rates of non-relapse mortality, early relapse, and primary graft failure (PGF).

Graft-versus-host disease (GvHD) represents one of the most common causes of death after allo-HSCT. Post-transplant cyclophosphamide (PTCy) has demonstrated to be effective for GvHD prevention in both haploidentical transplants and in matched unrelated transplants. For allo-HSCT in MF, peripheral blood stem cells represent the most appropriate stem cell source and PTCy is usually combined with a second agent, such as a calcineurin inhibitor or mycophenolate mofetil, which increase the rate of infections and reduce the graft-versus-leukemia effect.

Ruxolitinib is a Janus kinase (JAK) 1/2 inhibitor approved for both, the treatment of steroid-refractory acute GvHD (aGvHD) and MF. In addition, preclinical studies suggested that it may not impact on the graft-versus-leukemia effect.

In a study published in Acta Haematologica, Elena Vladislavovna Morozova and colleagues evaluated the combination of ruxolitinib and PTCy, as GvHD prophylactic regimen, in patients with primary or secondary MF who underwent allo-HSCT.¹

Study design

Treatment

• Reduced intensity conditioning: fludarabine (180 mg/m²) and busulfan (10 mg/kg)
• GvHD prophylaxis included
  • PTCy, 50 mg/kg on Days +3 and +4
  • Ruxolitinib, 45 mg/day from Day –7 until Day –2, followed by 15 mg/day from Day +5 to +100. The dose was reduced to 10 mg/day in cases of severe poor graft function (SPGF)

Methods

• High-performance liquid chromatography-tandem mass spectrometry (HPLC-MS/MS) was used to measure ruxolitinib plasma concentrations
• Quantitative detection of interleukin (IL) 8 (IL-8), interferon gamma, IL-17, and IL-1β was done with enzyme-linked immunosorbent assays
• STAT5B, JAK1, and JAK2 gene expression levels were analysed with qRT-PCR

The primary endpoint of the study was the incidence of aGvHD Grade 2–4 and chronicmoderate and severe GvHD. Secondary endpoints were overall survival and event-free survival, incidence of relapse, PGF, and SPGF.

Patient characteristics

This prospective study (NCT02806375) enrolled 20 patients with primary MF, post-essential thrombocythemia, and post-polycythemia vera MF. Patient characteristics are reported in Table 1.

Table 1. Patient characteristics¹

Allo-HSCT, allogeneic hematopoietic stem cell transplantation; DIPSSplus, dynamic international prognostic scoring system plus; HCT-CI, Hematopoietic Cell Transplantation Co-morbidity Index; MF, myelofibrosis; PMF, primary myelofibrosis; Post-ET-MF, post-essential thrombocythemia MF; Post-PV-MF, post-polycythemia vera MF.
Characteristics	Patients (N = 20)
Age, years (range)	51 (32–64)
Diagnosis, n (%) PMF Post-PV-MF Post-ET-MF	14 (70) 3 (15) 3 (15)
Risk profile according to DIPSSplus, n (%) Intermediate-1 Intermediate-2 High Blast crisis	2 (10) 14 (70) 4 (20) 0
Palpable spleen size at transplant, n (%) ≥ 10 cm < 10 cm Splenectomy	6 (30) 7 (35) 7 (35)
Median time between splenectomy and allo-HSCT, months (range)	2.60 (0.17–4.50)
Fibrosis grade before allo-HSCT, n (%) MF-2 MF-3	8 (40) 12 (60)
Mutational status, n (%) JAK2 V617F-positive CALR-positive MPL Triple negative	13 (65) 4 (20) 2 (10) 1 (5)
Karyotype, n Normal t(6;11) (p25;q12) del(13q21) Trisomy 8 Unknown	14 1 1 3 1
Median time of ruxolitinib therapy before allo-HSCT, months (range)	7.4 (3.0–22.0)
Response status at the moment of allo-HSCT, n (%) Clinical improvement Stable disease Progression	7 (35) 12 (60) 1 (5)
HCT-CI, n (%) 0 1 2 3	10 (50) 6 (30) 3 (15) 1 (5)
СD34+ cells/kg × 106, median (range)	6.9 (1.4–12.0)
Donor, n (%) HLA-identical sibling Haploidentical   Unrelated HLA-matched HLA-mismatched 9/10	3 (15) 4 (20)     11 (55) 2 (10)
Stem cell source, n (%) Bone marrow Peripheral blood	1 (5) 19 (95)

Results

After a median follow-up of 27 months (1–51):

• Primary engraftment was observed in 17 patients; two patients died before engraftment (one due to severe Pseudomonas aeruginosa sepsis and one because of gastrointestinal bleeding), and one died after engraftment (due to thrombotic microangiopathy and infectious complications)
• Median time to neutrophil engraftment, 27 days (18–44)
• Median time to platelet engraftment, 38 days (15–219)
• Median time to achieve red blood cell transfusion independence, 59 days (20–540)

Toxicity

Observed toxicities are reported in Table 2.

Table 2. Toxicity¹

Toxicity	Patients, n (%)
Veno-occlusive disease None Mild	19 (95) 1 (5)
Sepsis	7 (35)
Invasive mycosis	1 (5)
Viral reactivation or infection	9 (45)
Viral hemorrhagic cystitis	3 (15)

 

Primary endpoint: Incidence of GvHD

The incidence of GvHD was the following:

• Grade 2–4 aGvHD, 25% (n = 5)
• Grade 3–4 aGvHD, 15% (n = 3)
• Overall incidence of chronic GvHD, 40% (n = 8) was
  • moderate, 20% (n = 4)
  • mild, 20% (n = 4)

Calcineurin inhibitors were used as a first-line therapy in six patients while two patients required systemic steroid therapy. GvHD-related mortality was reported in one case.

Secondary endpoints

• One patient that experienced PGF is in remission 1 year after a second allo-HSCT
• Eleven (55%) patients experienced SPGF, which was
  • resolved in two patients without ruxolitinib dose modifications
  • resolved in the other eight patients when the ruxolitinib dose was reduced from 15 to 10 mg/day:
    • In one, the SPGF resolved at Day +77
    • In another one, the SPGF resolved at Day +100 after ruxolitinib discontinuation
    • Three required СD34+ boost administration
    • Three required donor lymphocyte infusion

• Hematological and molecular remission and splenomegaly regression were achieved in all engrafted patients
• One patient relapsed at Day +665
• The 2-year non-relapse mortality was 15% (95% confidence interval [CI], 4–34%); 2-year overall survival was 85% (95% CI, 60–93%); and 2-year event-free survival was 72% (95% CI, 45–87%)
• Peaking levels of inflammatory cytokines, observed after graft transfusion, decreased after PTCy and ruxolitinib administration. In contrast, elevated levels of IL-8 were observed after engraftment and were linked with higher ruxolitinib concentrations.
• A decrease in STAT5B, JAK1, and JAK2 expression was observed after HSCT

Conclusion

This study showed that the combination of PTCy and ruxolitinib was well tolerated and provided a promising control of GvHD rates associated with stable ruxolitinib plasma concentration. However, the high incidence of SPGF and the relatively long median time to the leukocyte and platelet engraftment may require ruxolitinib dose adjustments. Relapse rates in this study were low, with no early relapses reported. In addition, there were no episodes of severe nephrotoxicity and a low rate of veno-occlusive disease was observed, which compares favorably with calcineurin inhibitors.

Despite the small sample size, these promising results prompted the initiation of a multicenter randomised phase II trial, which will compare PTCy + ruxolitinib with conventional PTCy + tacrolimus + mycophenolate mofetil prophylaxis in patients with acute leukemia undergoing unrelated and haploidentical transplantation.