Combination treatment of extracorporeal photopheresis and belumosudil for chronic graft-versus-host disease

Allogeneic hematopoietic cell transplant recipients often develop chronic graft-versus-host disease (cGVHD), including sclerodermatous changes [1]. First-line therapy includes corticosteroids and calcineurin inhibitors [2]. Unfortunately, > 70% of patients require treatment beyond first-line due to toxicity and/or insufficient response [3]. There is a paucity of data on the efficacy of agent combinations. A study of 26 cGVHD patients reported a 77% response to combination therapy of with belumosudil and one to three of the following treatments: extracorporeal photopheresis (ECP), ruxolitinib, prednisone, mycophenolate mofetil, sirolimus and/or calcineurin inhibitors [4].

Herein, we report on thirteen cGVHD patients (8 by retrospective chart review and 5 by prospective observation) with age range from 34 to 77 (Table 1) treated with combination ECP and belumosudil. ECP is a well-established second-line therapy that induces T-regulatory cells (T-reg) and monocyte-to-dendritic cell differentiation, while the recently approved ROCK2 inhibitor belumosudil has been shown to induce T-regs and inhibit fibrotic pathways. All had involvement of the skin/joint/fascia, while ten (77%) had eye, two (15%) had mouth, and four (29%) had lung involvement. At the start of treatment, the NIH cGVHD Global Severity scores included one (8%) with mild disease, six (46%) with moderate, and six (46%) with severe. By the NIH Clinical Response Scale at four months, eight (62%) demonstrated a partial response and five (38%) experienced stable disease. Of eight patients followed for 11 to 18 months, six (75%) demonstrated continuous improvement, one experienced stable disease (13%), and one experienced worsening (13%). No serious adverse events were observed. Two (15%) reported mild fatigue, two noted mild pruritus (15%), and one experienced nausea (8%). Baseline and follow up images were taken with a range of 3 to 18 months (Fig. 1).

Table 1.

Patient characteristics including duration of disease, prior treatment, baseline NIH score, which treatment was started first and for how long before starting the combination of the two agents, ECP and belumosudil dose during combination treatment, and 4-month NIH clinical response score

					Combination Treatment
Subject	Duration of Chronic GvHD	Prior Therapies	Baseline NIH Score	Treatment Started First and Duration Before Starting Combination	ECP Dose	Bel Dose	4 Month NIH Clinical Response
1	5 year 10 mo	Pred, rux, siro, tacro	Mild	ECP, 2 year 10 mo	Every 2 weeks	200 mg QD	Partial Response
2	4 yr	Rux, siro	Moderate	Bel, 6 mo	Weekly	200 mg QD	Stable
3	4 year 5 mo	Rux, tacro	Moderate	Bel, 2 mo	Weekly	200 mg QD	Stable
4	5 year 2 mo	Pred, rux, tacro	Moderate	ECP, 2 mo	Twice weekly	200 mg QD	Partial Response
5	5 year 8 mo	Rux, tacro	Moderate	ECP, 4 year 2 mo	Every 2 weeks	200 mg QD	Partial Response
6	5 year 11 mo	Pred, rux, siro	Moderate	ECP, 5 mo	Weekly	200 mg QD	Partial Response
7	10 year 4 mo	MMF, pred, rux, siro	Moderate	ECP, 3 year 10 mo	Weekly	200 mg QD	Partial Response
8	3 year 4 mo	Pred, rux tacro	Severe	ECP, 1 mo	Twice weekly	200 mg QD	Stable
9	3 year 9 mo	Rux, siro,	Severe	ECP, 11mo	Every 2 weeks	200 mg QD	Stable
10	4 year 11 mo	Pred, tacro	Severe	Bel, 4 mo	Twice weekly	200 mg QD	Stable
11	5 year 8 mo	Pred, rux	Severe	ECP, 4mo	Twice weekly	200 mg BID	Partial Response
12	5 year 11 mo	Pred, rux, tacro	Severe	ECP, 1 year 2 mo	Every 2 weeks	200 mg QD	Partial Response
13	10 year 6 mo	MMF, tacro	Severe	Bel, 1 year 11 mo	Twice weekly	200 mg QD	Partial Response

Belumosudil: bel, BID: twice a day, MMF: mycophenolate mofetil, Pred: prednisone, Tacro: tacrolimus, QD: daily, Rux: ruxolitinb, Siro: sirolimus

Fig. 1.

Representative images of cutaneous, subcutaneous, and joint range of motion improvement following initiation of combination treatment with extracorporeal photopheresis and belumosudil, with baseline photographs on the left of each set and after treatment photographs on the right

Four had matched physical therapy measurements and a significant decrease in percent deviation from normal range of motion was observed: shoulder (p = 0.0001), hip (p = 0.0046), and ankle (p = 0.0091). Five had matched PFT results: two started and remained above 80% predicted FVC and FEV1 (normal findings), one started just below 80% and crossed to above, and two remained below.

This retrospective/observational study supports that combination ECP and belumosudil therapy is well tolerated and may improve cutaneous and systemic fibrosis in patients with cGVHD, with 62% (eight subjects) demonstrating improvement and the other 38% (five subjects) experiencing stable disease at four months of combination treatment. While belumosudil therapy alone has demonstrated a 59–62% symptom reduction in patients in another study [3] and ECP alone has been reported with overall response rates as high as 80% [5], we suggest that the clinical responses in patients receiving combination therapy (Fig. 1) is particularly noteworthy for improvement in subcutaneous/fascial involvement. Though limited to a subset of subjects with matched data, other objective measures of physical therapy range of motion and pulmonary function testing also support the benefits of this combination therapy. Limitations include the retrospective and observational aspects which limited follow-up times, lack of standardization of initiation of combination therapy, and lack of physical therapy and pulmonary data across all patients. Our findings further suggest that prospective, randomized, single versus combination therapy studies are warranted.

Author contributions

MAS, JC, and MG wrote the main manuscript text. MAS and MG prepared Fig. 1. All authors reviewed the manuscript.

Data availability

Data supporting the findings of this work are available from reasonable request from the corresponding author, Michael Girardi, MD.

Declarations

Competing interests

This work was supported in part by an external collaborative research grant from Mallinckrodt Pharmaceuticals.