FDA Approval Summary: Belumosudil for Adult and Pediatric Patients 12 Years and Older with Chronic GvHD after Two or More Prior Lines of Systemic Therapy

Chronic GvHD (cGvHD) is a multisystem disorder in allogeneic hematopoietic stem cell transplant recipients that is manifested by both inflammatory and fibrosing features (1). Experimental models implicated both the innate and adaptive immune systems in the etiology. The pathogenesis is thought to evolve from an acute inflammatory response to tissue injury early posttransplant which evolves into chronic inflammation and dysregulation of both T and B cells with subsequent aberrant tissue repair and fibrotic reaction (2). Without treatment, cGvHD can cause severe multisystem tissue damage due to chronic inflammation and fibrosis which can be fatal. Standard first-line treatment is steroids with or without a calcineurin inhibitor, depending on the severity of the disease (3). Almost half the patients with cGvHD require three or more lines of therapy, which include multiple types of immunosuppressant drugs used off-label (4).

Rho-associated coiled-coil containing kinase 2 (ROCK2) is a serine/threonine kinase that regulates multiple cellular functions, including cytoskeleton organization, cell migration, and gene transcription (5). ROCK2 can be activated by the RhoA GTPase G protein in concert with Rho signaling or by cleavage of the carboxyl terminus by apoptotic proteins (6). Transcription of proinflammatory cytokines, such as IL17 and IL21, and of the profibrotic gene targets of myocardin-related transcription factor (MRTF) are controlled by the ROCK2 signaling pathway, and dysregulation of these factors has been implicated in the pathogenesis of cGvHD (7).

Belumosudil (Rezurock, KD025, SLx-2119) is a small molecule inhibitor of ROCK2. In ex vivo or in vitro studies, belumosudil inhibited the secretion of IL21 and IL17 in response to CD3-CD28 stimulation of human peripheral blood mononuclear cells, inhibited STAT3 phosphorylation and transcriptional activity, and inhibited MRTF-mediated gene expression (8, 9). In vitro, belumosudil also inhibited the mTOR pathway and, with less potency, the ROCK1 isoform (10). In vivo studies in cGvHD mouse models demonstrated that treatment with belumosudil resulted in improvement in cGvHD-related clinical and histologic abnormality scores in the lung and skin (11). Herein we summarize the FDA review (10) of the marketing application for belumosudil for treatment of cGvHD.

Belumosudil drug product is available as 200-mg tablets. The mean bioavailability was 64% in healthy subjects. The median Tmax was 1.26 hours in patients; administration of belumosudil with food increased Cmax and AUC 2.2 times and 2 times, respectively, and delayed Tmax by 0.5 hours.

The mean steady-state AUC and Cmax of belumosudil given with food were 22.7 h•mcg/mL and 2.39 mcg/mL, respectively. Belumosudil is metabolized primarily by CYP3A4 and to a lesser extent by CYP2C8, CYP2D6, and UGT1A9 in vitro. Two metabolites, KD025m1 and KD025m2, also inhibit ROCK2. The mean elimination half-life of belumosudil was 19 hours, and clearance was 9.83 L/hours in patients. Excretion was estimated as 85% in feces (30% unchanged) and less than 5% in urine.

Belumosudil pharmacokinetics (PK) was described by a two-compartment model with first order absorption and a lag time. In population pharmacokinetics (PPK) analyses, no clinically significant differences in belumosudil PK were observed with regard to age (18–77 years), sex, weight (38.6–143 kg), or mild to moderate renal or hepatic impairment; a diagnosis of GvHD was associated with decreased clearance in comparison to healthy volunteers. The effect of severe renal or hepatic impairment on belumosudil PK has not been studied.

Within the full clinical development program, belumosudil was evaluated at doses of 20 to 1,000 mg in 15 clinical studies in patients with cGvHD, psoriasis, or idiopathic pulmonary fibrosis, or in healthy volunteers. Two clinical trials were conducted in patients with cGvHD. On the basis of tolerability in earlier trials, belumosudil dosages of 200 mg daily, 200 mg twice daily, and 400 mg daily were evaluated in Study KD025–208 (NCT02841995), a single-arm dose-escalation trial of belumosudil for treatment of 54 patients with cGvHD failing one to three prior lines of therapy (12). All three dosages appeared to be active (Supplementary Table S1), so the two lower dosages were chosen for further evaluation in Study KD025–213 (NCT03640481), a randomized dose-ranging trial of belumosudil for treatment of 132 patients with cGvHD following two to five prior lines of therapy (13). Using data for all 186 patients from the two trials, pooled exposure–response analyses showed a trend for increasing risk of safety events (adverse events at Grades 4–5, leading to treatment interruption, or resulting in drug withdrawal) with increasing exposure (Supplementary Fig. S1), but no relationship for response with exposure, and it was concluded that the 200-mg daily dosage was acceptable.

In Study KD025–107 (NCT03530995), a dedicated drug–drug interaction study, the strong CYP3A inducer rifampicin significantly reduced belumosudil Cmax by approximately 60% and overall exposure by approximately 72%, and the proton pump inhibitor (PPI) rabeprazole reduced belumosudil Cmax by up to 87% and overall exposure by up to 80% (14). The PPK models confirmed the reduced exposures with these drugs. The recommended dosage of belumosudil is increased to 200 mg twice daily when used concomitantly with a strong CYP3A inducer or a PPI.

Study KD025–213 (NCT03640481) was a randomized, open-label, multicenter dose-ranging study of belumosudil for treatment of patients with cGvHD who had received two to five prior lines of systemic therapy and required additional treatment. Concomitant treatment with GvHD prophylaxis and standard care systemic cGvHD therapies were permitted when the subject was on a stable dose for at least 2 weeks prior to study.

Enrolled patients were randomized 1:1 to one of two oral dosages of belumosudil (200 mg daily and 200 mg twice daily), and randomization was stratified according to prior cGvHD treatment with ibrutinib and presence of severe cGvHD. The primary objective was to evaluate efficacy as determined by achievement of overall response rate (ORR) that included complete response (CR) or partial response (PR) at any time on study as determined by the investigator according to the 2014 NIH consensus criteria (15). The results were to be assessed in each arm individually; the study was not powered to assess for differences in outcomes between arms. A sample size of 63 patients per arm was calculated to have 90% power with a two-sided alpha 0.045 to exclude an ORR of 30% with a true ORR of 55%. The analysis set included all treated patients. The primary analysis was to occur when 126 patients completed 6 months on study, and a follow-up analysis was to occur when 126 patients completed 12 months on study.

There were 132 patients randomized, 66 to each arm. Both arms met the objective of excluding a 30% ORR as reported by the investigator at the primary analyses, and the study was concluded to be positive. For FDA's assessment of efficacy, the analysis was limited to the 65 patients with active cGvHD at study baseline in the cohort receiving belumosudil 200 mg daily (Table 1). FDA adjudicated all responses and determined the duration of response (DOR) using all available raw data for the components of the response criteria (Supplementary Table S2) using the dataset with 12 months of follow-up.

The efficacy outcomes in the pivotal cohort are summarized in Table 2. The ORR through Cycle 7 Day 1 was 75% (95% CI, 63–85). The responses were consistent across subgroups, including in patients who previously received treatment with ibrutinib or ruxolitinib (Supplementary Table S3). The median time to first response was 1.8 months (95% CI, 1.0–1.9). The median DOR was 1.9 months (95% CI, 1.2–2.9), and the median time from first response to death or new systemic therapy, was not reached. In patients who achieved response, no death or new systemic therapy initiation occurred in 62% (95% CI, 46–74) of patients for at least 12 months since response.

An exploratory analysis of patient-reported symptom bother showed at least a 7-point decrease in the Lee Symptom Scale (LSS) score by Cycle 7 Day 1 in 52% (95% CI, 40–65) of patients and was consistent with the ORR results in Study KD025–213 efficacy population.

Because belumosudil's mechanism of action involves fibrosis pathways and Th17 cells, impaired wound healing and infections were anticipated. Belumosudil also inhibited UGT1A1 at concentrations that could be achieved clinically, so transient isolated hyperbilirubinemia was also anticipated. Photoabsorbance was observed in vitro, and the drug distributed to photosensitive areas such as skin and uveal tract, suggesting a potential for phototoxicity. Also in vitro, belumosudil blocked the human ether-à-go-go-related gene (hERG) current, indicating a potential for QT prolongation.

In animal studies of belumosudil, toxicities were observed in the gastrointestinal tract, liver, kidney, hemolymphoid system, and reproductive system. In safety pharmacology studies in dogs, belumosudil was associated with lowering of the blood pressure. In reproductive toxicity studies, belumosudil was associated with embryo-fetal toxicity and impaired male fertility.

The adverse reactions of belumosudil across all doses in the 15 studies in the clinical development program were reviewed elsewhere (10). The safety profile of belumosudil in the intended population was based on the results for 83 adult patients with cGvHD treated with belumosudil 200 mg daily in Studies KD025–208 and KD025–213. The median duration of treatment for these patients was 9.2 months (range, 0.5–44.7 months).

One patient on belumosudil 200 mg daily had a fatal adverse reaction, including severe nausea, vomiting, diarrhea, and multi-organ failure. Permanent discontinuation of belumosudil due to adverse reactions occurred in 18% of patients, predominantly nausea (4%). Adverse reactions leading to dose interruption occurred in 29% of patients, predominantly infections (11%) and diarrhea (4%). The common clinical adverse reactions are shown in Table 3, and the laboratory adverse reactions are shown in Table 4.

Additional adverse events of special interest included infections, relapse, QT prolongation, hypotension, and impaired wound healing. Infections occurred in 65% of the patients treated with belumosudil 200 mg daily (Supplementary Table S5); the infection event was Grades 3 to 4 in 18%. The most common infections were upper respiratory tract infection (31%) and pneumonia (11%). Relapse was reported in 6%, QT prolongation in 2%, hypotension in 5%, and impaired wound healing in 1%.

The 75% ORR with durability in the 200-mg daily dosage arm of Study KD025–213, supported by the LSS results, the ORR in the 200-mg twice daily dosage arm, and the ORRs in Study KD025–208 were considered substantial evidence of effectiveness. On the basis of the biology of treatment-refractory cGvHD and the mechanism of action of belumosudil, the efficacy of belumosudil for treatment-refractory cGvHD was extrapolated from the adult population to pediatric patients. The pharmacometrics analysis showed no clinically significant effect of weight on exposure in adults, but no PK or safety data were available for smaller patients, so extrapolation was limited to children 12 years and older to meet the weight range with observed data.

A major strength of the belumosudil clinical development program for cGvHD was the attention to dose optimization. For targeted therapies, use of the maximum tolerated dose is frequently unnecessary (16); instead, to minimize toxicity, the lowest effective dose would be advantageous. Randomized dose-finding trials would be optimal to generate unbiased safety, response, and pharmacodynamic marker data, but such trials need not be powered to demonstrate significant differences in efficacy by dose. To this end, the PK, safety, tolerability, and efficacy data from Studies KD025–208 and KD025–213 were adequate to enable the pharmacometrics analyses that informed the recommended dosage.

At the present time, FDA considers a CR or PR within the first 6 months of treatment to be representative of a clinically meaningful response if durable. In Study KD025–213, the DOR was only 1.9 months (8 weeks). Although the median DOR is quite short, the definition of DOR (Supplementary Table S2) does not take into account that cGvHD may flare and resolve without additional treatment. An additional measure of the durability of response would be the time to death or new systemic therapy. For the 49 responders, the median time to either death or new systemic therapy for cGvHD was not estimable due to the low number of events, but no death or new systemic therapy initiation occurred in 62% (95% CI, 46–74) of patients for at least 12 months since response. This is not to suggest that 12-month DOR is a clinical benefit, but rather it is to provide context to the durability when the median is not yet reached due to the limited follow-up.

The mainstays of treatment of cGvHD have largely been immunosuppressive drugs that interfere with the number or function of T cells and/or B cells, resulting in a substantial risk of infection. Belumosudil is somewhat unique in the cGvHD armamentarium in that its proposed mechanism of action is to restore the balance of regulatory and effector T cells, as would be needed to induce tolerance, in addition to addressing the fibrotic reaction to the inflammation-related tissue damage. In view of these potential salutary effects, the low CR rate in Study KD025–213 was somewhat disappointing. However, as is the case for most trials of treatments for cGvHD, the study participants were quite heterogeneous with regard to the concomitant immunosuppressive drug used, and it is not clear whether some of these drugs might negate the effects of belumosudil. For example, a drug that eliminates all T cells or that dampens all cytokine production might interfere with the ability of belumosudil to restore the balance of regulatory and effector T cells. To design more rationale combinations for drugs for treatment of cGvHD and restoration of tolerance, additional studies are needed to characterize the immune system of patients with cGvHD and to determine how this is affected by various drugs (17).

There are also transporter- or metabolism-based interactions that may affect safety when belumosudil is used with other concomitant medications. Belumosudil is known to inhibit transporters such as p-glycoprotein, BCRP, and OATP1B1 and metabolism enzymes such as CYP1A2, 2C19, 2D6, and UGT1A1, but it has little effect on CYP3A (10). Calcineurin and mTOR inhibitors used commonly in patients with cGvHD include substrates of CYP3A, p-glycoprotein, and OATP1, but formal studies of belumosudil interactions with calcineurin inhibitors or sirolimus were not performed. Standard therapeutic drug monitoring of calcineurin inhibitors and sirolimus is currently considered sufficient when these drugs are used in combination with belumosudil. The potential for other drug interactions will be studied as a postmarketing requirement.

In the absence of a randomized control for comparison, it was challenging to clearly confirm a causal association between belumosudil and adverse events observed in Study KD025–213, but the risks of elevated transaminases, anemia, lymphocytopenia, headache, nausea, and diarrhea were confirmed in early-phase trials of belumosudil in other disease settings and in healthy volunteer studies (see Section 8.3 in ref. 10). These risks appear to have been minimized by optimal dose selection for the cGvHD population and instructions for dose reduction if needed for elevated transaminases or other serious adverse reactions. The risk of infection is less clear, given the concurrent use of other immunosuppressive drugs as well as standard prophylactic antibiotics. The incidence of viral infections appeared to increase numerically with dose (Supplementary Table S5), although the confidence intervals are wide due to the small number of patients treated with the highest dose. Data with longer follow-up would be needed to confirm the safety of belumosudil with long-term use in patients with cGvHD.

Given the observed response rate and associated durability, and with the mitigation strategies in place in labeling, the clinical benefit of belumosudil appears to outweigh the risks of treatment of adult and pediatric patients 12 years and older with cGvHD after failure of at least two prior lines of systemic therapy.

No disclosures were reported.

The Editor handling the peer review and decision-making process for this article has no relevant employment associations to disclose.

We thank Rosa Lee-Alonzo, PharmD, RAC, for expert review project management.