Steroid-refractory GvHD: long term outcomes and risk factors

Despite the progression in transplant medicine, such as hematopoietic stem cell transplantation (HSCT), there are still high morbidity rates and reduced quality of life in surviving patients. Graft-versus-host disease (GvHD) is the leading cause of non-relapse mortality, and steroid-refractory (SR) GvHD has a particularly poor prognosis with a long-term survival of 20–30%.¹

Simona Pagliuca and colleagues conducted a retrospective study of acute (a)GvHD and chronic (c)GvHD following HSCT to determine the outcomes and risk factors of SR aGvHD and cGvHD, and to identify predictors of refractory disease. The results of this study were published in Bone Marrow Tranplantation.¹

Study design and patient characteristics

The study assessed all consecutive patients who received HSCT between 2007 and 2017 at Saint Louis Hospital, Paris, FR. They collected clinical data, such as age, gender, disease diagnosis, comorbidities, type of transplant, human leukocyte antigen (HLA) matching, conditioning regimen, GvHD prophylaxis, aGvHD and cGvHD diagnosis, and other complications based on the European Society for Blood and Marrow Transplantation registry. 

The study included 1,207 patients. Selected patient, disease, and transplant characteristics are detailed in Table 1.

• Median age at transplant was 43.14 years (range 4.13–74.77)
• Patients were predominantly male (61%)
• Median duration of follow-up was 42.2 months (range 1.6–97.7)
• Most common HSCT indication was acute leukemia (44%)
• The majority of patients were classified as intermediate risk by refined disease risk index (48%)
• Most frequently used conditioning regimen was reduced-intensity conditioning
• Donor type was most commonly matched related donor (45%)
• Most common GvHD prophylaxis was cyclosporin and mycophenolate mofetil (CSA-MMF; 60%)

 Table 1. Selected patient, disease, and transplant characteristics (adapted from Pagliuca et al.¹)

• A total of 31.3% of patients were diagnosed with Grades 2–4 aGvHD at 100 days (18.6% Grade 3–4), while SR aGVHD developed in 16.6% of patients
• cGVHD was seen in 38.1%  of patients (95% CI, 35.3–40.9.6) at 1 year (46.2% at 2 years, and 48.3% at 3 years), with SR cGvHD in 19.9%, 23.4%, and 24.7% at 1, 2, and 3 years, respectively

Results

• A significant difference was seen in survival probability between patients with steroid-responsive and SR aGvHD at 1 year (p < 0.0001; Table 2), while there was no significant difference when comparing patients with and without SR cGvHD at 3 years.
• Survival probability at 3 years of patients with steroid-responsive GvHD was considerably higher compared to patients without GvHD or patients with SR GvHD (Table 2)

Multivariate analysis

• Multivariate analysis revealed the following pre-transplant GvHD risk factors associated with Grades 2–4 SR aGvHD:
  • Very high risk by refined disease risk index (HR, 1.47; p = 0.021),
  • Transplant from a matched/mismatched unrelated donor (HR, 1.61; p < 0.001),
  • Absence of T-cell depletion (HR, 1.59; p < 0.001),
  • Reduced intensity conditioning regimen (HR, 1.79; p < 0.001)
• Pre-transplant risk factors for SR cGvHD were:
  • Transplant from unrelated donor (HR, 1.50; p < 0.001),
  • Use of PB as stem cell source (HR, 1.70; p < 0.001),
  • Absence of T-cell depletion (HR, 1.67; p < 0.001)
• In terms of the risk of SR aGvHD, the study found that in patients who had three or four SR  aGvHD risk factors, there was an 18.1% (95% CI, 12.9–24.1) probability of becoming steroid refractory within the first year after transplant
• In contrast, presence of two or three pre-transplant risk factors for SR cGvHD increased the risk of developing SR cGvHD to 80.6% (95% CI, 73.7–85.3%)

 Multistate analysis

• At the end of follow-up (121 months), there was an 18.6% probability of being alive with no event (95% CI, 14.8–23.1), a risk of being in an aGvHD state of 0.2% (95% CI, 0.1–0.5) or cGvHD state of 3.7% (95% CI, 1.5–9.1), and the probability of being alive after GvHD resolution was 36.2% (95% CI, 31.5–41.2)

• The same analysis found that the probability of being in relapse was 31.1% (95% CI, 26.8–35.8) and the probability of death was 10% (95% CI, 8.2–12.1)

Cause of death

• The principal cause of death in patients with SR GvHD was GvHD itself, followed by infection and then relapse (Table 3)

Table 2. Survival and cumulative incidence after HSCT¹

Table 3. Causes of death in patients with steroid refractory GvHD¹

Conclusion

The study found that the outcome for patients with SR aGvHD is poor, with half developing a chronic form of GvHD. The authors propose a risk model based on pre-transplant risk factors which, once validated, would allow to stratify patients who may benefit from a more intense immunosuppressive therapy. Interestingly, 3-year survival in patients with SR GvHD is not different to that of patients without any GvHD, while it is improved in patients with steroid-responsive GvHD, possibly due to enhanced graft-versus-leukemia effect. The authors feel these results should be taken with caution, though they also highlight that relapse rates are higher in patients who have never had cGvHD compared with those who have.

Although the authors understand the limitations of this retrospective, single-center study, they highlight that it is the first comprehensive study of SR GvHD incidence and risk factors and feel it can aid the management of this category of patients.