AUTHORS
Sergio Rodriguez-Rodriguez, MD, MSc, Rebeca Bailen Sr2, Udo Holtick, MD, PhD3, Matthias Fante4, Silke Heidenreich, MD5, Nihar Desai, MD, DM6, Keven Vachon7, Brittany Salter, MD, PhD8, Christopher Lemieux, MD9, Kareem Jamani, MD, MPH10, Mohamed Elemary, MD,MSc,PhD,FRCP11, Sylvie Lachance, MD, FRCPC, DRCPC12, Kylie Lepic, MD13, Amani Ahmed, MD14, Jennifer White, MD, MSc, FRCPC15, Felipe Peña-Muñoz, MD16, Juan Martín-González17, Patricia Alcalde-Mellado, MD18, Jose Antonio Perez-Simon, MD, PhD19, Aranzazu Aranzazu Bermudez, MD20, Lucia Lopez Corral, MD, PhD21, Maria Laura Fox, MD22, Carlos De Miguel, MD23, Maria Pascual, MD24, Daniel Wolff, MD25, Mi Kwon, MD,PhD26, Dennis Dong Hwan Kim, MD/PhD27
AFFILIATIONS
1 Princess Margaret Cancer Centre, Toronto, Ontario, Canada, University of Toronto, Toronto, Ontario, Canada
2 Gregorio Marañón General University Hospital Emergency Room, Madrid, Spain
3 University of Cologne, Köln, Germany
4 University Hospital Würzburg, Würzburg, Germany
5 University Medical Center Hamburg Eppendorf Emergency Department, Hamburg, Germany
6 Princess Margaret Cancer Centre, Toronto, Ontario, Canada, University of Toronto, Toronto, Ontario, Canada
7 CHU de Québec-Université Laval, Québec, Quebec, Canada
8 Juravinski Cancer Centre, Hamilton, Ontario, Canada
9 CHU de Québec-Université Laval, Québec, Quebec, Canada
10 Arthur J.E. Child Comprehensive Cancer Centre, Calgary, Alberta, Canada
11 Saskatchewan Cancer Agency, Saskatoon, Saskatchewan, Canada
12 Maisonneuve-Rosemont Hospital, Montréal, Quebec, Canada
13 Juravinski Cancer Centre, Hamilton, Ontario, Canada
14 BC Cancer Agency, Vancouver, British Columbia, Canada
15 BC Cancer Agency, Vancouver, British Columbia, Canada
16 ICO Hospital Duran i Reinals, L'Hospitalet de Llobregat, Spain
17 Hospital Universitario Virgen del Rocío, Sevilla, Spain
18 Hospital Universitario Virgen del Rocío, Sevilla, Spain
19 Hospital Universitario Virgen del Rocío, Sevilla, Spain
20 Hospital Universitario Marqués De Valdecilla, Santander, Spain
21 Hospital Universitario de Salamanca, Salamanca, Spain
22 Hospital Vall d'Hebron, Barcelona, Spain
23 Hospital Universitario Puerta De Hierro Majadahonda, Majadahonda, Spain
24 Hospital Regional Universitario de Málaga, Málaga, Spain
25 University Hospital Regensburg, Regensburg, Germany
26 Gregorio Marañón General University Hospital Emergency Room, Madrid, Spain
27 Princess Margaret Cancer Centre, Toronto, Ontario, Canada, University of Toronto, Toronto, Ontario, Canada
ABSTRACT
Introduction: Belumosudil (BEL), a selective Rho-associated coiled-coil kinase 2 (ROCK2) inhibitor, showed clinical benefits in patients (pts) with steroid-refractory chronic graft-versus-host disease (SR-cGvHD) in a randomized phase 2 ROCKstar trial. Multiple real-world experience (RWE) studies have replicated high overall response rates (ORR), though limited by relatively small individual cohort sizes. Through international collaboration, the present study expands to 366 pts, reports ORR and failure-free survival (FFS), and compares BEL plus ruxolitinib (RUX) combination with BEL monotherapy using propensity-score matched analyses.
Patients and methods: Clinical characteristics were captured retrospectively in 366 pts treated with BEL in Canada, Spain, Germany, and Austria. ORR, corticosteroid (CS) dose reduction, and FFS were analyzed; FFS was defined as time from BEL initiation to addition/switch of cGvHD therapy, non-relapse mortality (NRM), or relapse. BEL+RUX combination (excluding other combination regimens) was compared with BEL monotherapy (+/- prednisone) using unadjusted and adjusted Cox models (country, cGvHD subtype/severity, organ count, lung involvement, prior RUX/ECP exposure, prior treatment lines, time from cGvHD onset, refractoriness to last therapy). Overlap-weighting (ATO) and propensity-score matching (PSM 1:1 nearest-neighbor, 0.2 standard deviation caliper) balanced prior RUX exposure, refractoriness to last therapy, and time from cGvHD onset.
Results: Overall, 77% (n=274/355) had severe cGvHD, with a median of 3 organs (IQR 2–4) involved before BEL. Of 366 pts, 314 were evaluable for ORR. With a median follow-up of 8.7 (IQR 4.7–14.4) months, ORR was 49.7% (3 mo), 52.8% (6 mo), and 44.2% (12 mo); best ORR by 6 and 12 mo was 51.7% and 47.4%, respectively. Median CS dose was reduced from 0.14 at baseline to 0.06 mg/kg/d by 12 months. FFS rate was 85.1% and 72.8% at 6 and 12 months, with OS 96.2% and 90.9%, respectively. The most common failure cause was addition/switch of systemic therapy (17.2% by 12 months), followed by intolerance (5.6%), NRM (4.3%), and relapse (0.8%).
Treatment regimens included BEL+RUX combination (n=123), BEL monotherapy (n=164), and other combinations (e.g., BEL+RUX+ECP, BEL+ECP; analyzed separately). In comparison to BEL monotherapy, BEL+RUX combination treatment was preferentially used in the pts with more aggressive cGvHD features: BEL+RUX combination group had higher prior RUX exposure (97% vs. 70%, p<0.001), older age (median 59 vs. 54 years, p=0.004), slightly higher rate of refractoriness to last prior therapy (59.8% vs. 46.7%, p=0.082), and shorter interval from cGvHD onset to BEL (median 31.4 vs 49.6 months, p=0.012). Baseline cGvHD characteristics were similar between the 2 groups for severe cGvHD (both 74%, p=1.0), median previous treatment lines (4 vs. 5, p=0.050), number of organs involved (both median 3, p=0.087), and lung involvement (39% vs. 41%, p=0.710).
FFS at 6 and 12 months was 84.2% and 63.4% with BEL+RUX, and 89.6% and 82.5% with monotherapy (unadjusted HR 1.92 for BEL+RUX [1.16-3.17], p=0.011); NRM was numerically higher with BEL+RUX (5.7% vs. 0.6%, p=0.056). After adjustment (HR 1.13 [0.71–1.81], p=0.610), with ATO (HR 1.27 [0.86–1.89], p=0.226), and in 1:1 PSM (n=146, 73 per group) (HR 1.56 [0.87–2.80], p=0.134), the 12-month FFS difference attenuated (54.4% BEL+RUX vs. 68.6% monotherapy).
Median CS dose at baseline and 12 months was 0.12 and 0.04 mg/kg/d with BEL+RUX vs. 0.13 and 0.07 mg/kg/d with BEL monotherapy (p=0.790 by paired t-test). ORR was numerically higher with combination at every time point (best ORR by 6 months 55.0% vs. 46.7%; by 12 months 46.8% vs. 45.7%, p=0.90), but not statistically significant on univariate (OR 1.08 [0.65-1.80], p=0.760) or multivariate analysis (OR 1.27 [0.71–2.3], p=0.420). In the 1:1 PSM analysis, ORR did not differ significantly at 6 months (51.6% vs. 42.2%, OR 1.45 [0.69–3.11], p=0.376) or 12 months (42.2% vs. 40.0%, OR 1.09 [0.51–2.34], p=0.859).
Conclusion: This expanded RWE cohort (n=366 pts) confirms durable BEL efficacy in heavily pre-treated SR-cGvHD pts. BEL+RUX combination treatment was preferentially used in pts with more aggressive cGvHD: more prior RUX exposure, greater refractoriness to prior line therapy, and rapid transition to BEL, features associated with attenuation of the FFS difference across multiple analyses. These findings warrant prospective evaluation of BEL-RUX sequencing.
DISCLOSURES
Sergio Rodriguez-Rodriguez, MD, MSc: No relevant conflicts of interest to declare.