Extracorporeal Photopheresis (ECP) for Acute Graft-Versus Host Disease (GvHD) after Allogeneic Hematopoietic Stem Cell Transplantation (allo-HSCT)

Background

Graft-versus-host disease (GvHD) is a major limitation after allo-HSCT and remains a frequent cause of death. The 5-years survival is 25% and 5% for grade III and IV, respectively. Acute GvHD occurs in up to 45% of HLA-matched and up to 75% in case of unrelated donors. The standard-treatment consists of methylprednisolone (usually 2 mg/kg/day) and a calcineurin-inhibitor. No standardized second-line treatment for acute GvHD exists. Here, we report a pilot single-centre experience with extracorporeal photopheresis (ECP) for acute GvHD: the objective was to investigate the efficacy of ECP for patients with steroid-refractory/-dependent acute GvHD as well as an early intervention in patients with low-grade acute GvHD to avoid/taper steroids. Furthermore, we evaluated the reduction of immunosuppressive therapy.

Patients' characteristics

Between 2013 and 2014, 17 patients with acute GvHD (of whom two patients developed GvHD after donor lymphocyte infusion) were treated. Eight patients had a maximum grade of GvHD I/II (2/6 patients) and nine patients were graded as III/IV (6/3 patients). Organ involvement was as follows: skin only in 10, skin and liver in one, skin and gastrointestinal tract in two and all three organs were involved in four patients.

Treatment before ECP consisted of topical steroids in one and 0.5 mg/kg methylprednisolone (due to side-effects of calcineurin-inhibitor) in the other patients with grade I. Six patients received 1 mg/kg and eight patients received 2 mg/kg methylprednisolone. One patient was treated with 2 mg/kg methylprednisolone and weekly methotrexate. Before start of ECP, one patient was steroid-free, six patients were steroid-refractory and nine patients were steroid-dependent. Thus, we treated patients with acute GvHD not only for steroid-refractory disease but also steroid-dependent disease and grade I GvHD to avoid a treatment with steroids.

Results

The median number of ECP sessions per patient was 12 (range, 5 - 36), seven patients received ECP twice a week. Best response to ECP was complete remission in 71%, partial response in 12% and no response in 17%, after a median number of 6 treatments (range, 2 - 9). Response was better for grade I/II: 87.5% received complete remission compared to 56% with grade III/IV, partial response was observed in 12.5% in patients with grade I/II versus 11% with grade III/IV. No responders comprised 33% with grade III/IV and 0% with grade I/II.

Immunosuppressive therapy could be tapered successfully: mean reduction of steroids was 95% (range, 60 - 100) and mean reduction of calcineurin-inhibitor was 83% (range, 40 - 100).

Six patients developed a rebound of GvHD during tapering (two patients) or after discontinuation (four patients) of ECP. Eleven patients (78%) developed chronic GvHD (two patients with severe grade), whereas it appeared in four patients during tapering of ECP and in seven patients after discontinuation of ECP after a median time of 116 days (range, 30 - 287).

We could observe seven bacterial, 14 viral and one fungal infection in 14 patients, which are expected rates after allo-HSCT in patients with acute GvHD. After a median follow up of one year, two patients relapsed from their underlying disease and five patients died (one due to relapse and four due to infections).

Conclusion

In this single-centre pilot experience, we could show that ECP is an efficient and safe treatment in patients with steroid-refractory or steroid-dependent acute GvHD as well as an upfront-treatment in patients with low-grade GvHD. We were able to taper immunosuppressive therapy with a mean reduction of steroids of 95% and mean reduction of calcineurin-inhibitor of 83%.

Best responses were seen in patients with I/II grade GvHD which concludes that ECP should be started as early as possible.

Further studies are warranted to investigate a schedule to reduce the risk of rebound of acute GvHD (42% in our cohort) and development of chronic GvHD (78%).