Apraglutide/Ruxolitinib Provides Efficacy Benefit vs Ruxolitinib Alone in Refractory GI aGVHD

Acute GVHD | Image Credit:

© Mikkel – stock.adobe.com

The selective GLP-2 analog apraglutide plus ruxolitinib (Jakafi) generated improved short- and long-term efficacy outcomes vs ruxolitinib alone in patients with steroid-refractory gastrointestinal (GI) acute graft-vs-host disease (aGVHD), according to findings from a study comparing efficacy outcomes from the phase 2 STARGAZE trial (NCT05415410) vs a control cohort from the observational MAGIC trial.

The data, which were presented at the 51st Annual EBMT Meeting, showed that the all-organ overall response rates (ORRs) were higher in the STARGAZE cohort (n = 31; apraglutide vs ruxolitinib) vs the MAGIC cohort (n = 31; ruxolitinib monotherapy) at days 28 (STARGAZE, 58.1%; MAGIC, 38.8%), 56 (45.1%; 32.2%), and 91 (45.1%; 25.8%). The complete response (CR) rates at these respective time points were also higher in STARGAZE (25.8%; 29.0%; 29.0%) vs MAGIC (19.4%; 16.1%; 16.1%).

Harnessing the Lower-GI aGVHD Benefits of GLP-2 Analogs

GI aGVHD is a life-threatening complication associated with allogeneic hematopoietic stem cell transplant (allo-HCT) that has a 36-month mortality rate over 50% and targets intestinal cells such as intestinal stem cells, Paneth cells, enterocytes, and L cells. Furthermore, steroid-refractory lower-GI aGVHD is associated with high rates of non-relapse mortality (NRM) despite the availability of ruxolitinib for second-line treatment. L cells produce GLP-2 and can also be damaged by chemotherapy and total body irradiation (TBI). GLP-2 is involved in maintaining normal GI integrity, stimulating intestinal mucosal growth, reducing intestinal inflammation, and reducing enterocyte apoptosis.

A preclinical study showed that relative gene expression levels of enteroendocrine L-cell markers, GLP-2–positive cell levels, and relative gene expression levels of the GLP-2 receptor decreased in the small and large intestines of mouse models exposed to TBI, chemotherapy, and bone marrow transplant. However, this study demonstrated that treatment with a GLP-2 analog prevented GVHD-induced Paneth cell loss in the mouse models. The mean number of Paneth cells was higher in evaluable models treated with tedeglutide (n = 8) vs those that received a vehicle (n = 8). Lysozyme expression levels were also higher in evaluable models treated with tedeglutide (n = 11) vs the vehicle (n = 11). Overall, in this study, prophylactic GLP-2–based analog treatment promoted the survival and growth of intestinal stem cells and Paneth cells.

Zooming in on the STARGAZE Design

The present comparison study compared outcomes between the apraglutide/ruxolitinib arm of STARGAZE with a matched control cohort of the ruxolitinib monotherapy arm of MAGIC.

STARGAZE was a multicenter, randomized, single-blind, proof-of-concept trial that enrolled patients with stage I to IV, clinically confirmed steroid refractory lower-GI aGVHD after allo-HCT from any donor source. Patients were required to have received prior therapy with systemic glucocorticoids plus ruxolitinib. Patients were excluded if they had received prior systemic GVHD therapy other than systemic glucocorticoids and ruxolitinib; had relapsed malignant disease; or had received prior enteral glutamine, endocrine peptides (including GLP-2 or GLP-2 analogs), or known antidrug antibodies.

The MAGIC control cohort was matched to the STARGAZE cohort using optimal matching with Mahalanobis distance.

Patients in STARGAZE were randomly assigned to receive apraglutide at a weight-based low dose (n = 15) or a weight-based high-dose (n = 15). One additional patient who weighed less than 50 kg received a fixed minimal dose of apraglutide. Patients received weekly apraglutide subcutaneously in combination with systemic steroids and ruxolitinib for a maximum of 182 days. The main treatment period lasted 56 days. Patients who had a CR on day 56 stopped treatment; all other patients continued treatment at investigator discretion. The treatment extension period lasted from days 57 to 91. Patients who had a CR on day 91 stopped treatment; all other patients continued treatment at investigator discretion from days 92 to 182 (the optional treatment period) or until CR, whichever occurred first. The follow-up period lasted for up to 2 years after the first study dose.

The short-term efficacy outcomes used in this comparison study were all-organ ORR and CR rate at days 28 and 56; durable all-organ ORR and CR rate at day 56; lower-GI ORR and CR rate at days 28 and 56; and durable lower-GI ORR and CR rate at day 56. The long-term efficacy outcomes were all-organ ORR and CR rate at day 91; lower-GI ORR and CR rate at day 91; NRM rates at 6 and 12 months; and overall survival (OS) rates at 6 and 12 months.

Expanding on Efficacy Findings From the Comparison Study

Among patients in the STARGAZE cohort (n = 31), the median age was 62.0 years (range, 40.0-77.0), 87.1% of patients had stage III or IV aGVHD in the second-line setting, and 64.5% of patients had stage III or IV lower-GI aGVHD in the second-line setting. In the MAGIC cohort (n = 31), the median age was 60.0 years (range, 29.0-75.0), 90.3% of patients had stage III or IV aGVHD in the second-line setting, and 64.6% of patients had stage III or IV lower-GI aGVHD in the second-line setting.

Patients in the STARGAZE cohort also had higher lower-GI GVHD response rates vs those in the MAGIC cohort at all time points. At day 28, the lower-GI ORR and CR rate were 54.8% (95% CI, 36.0%-72.7%) and 29.0% (95% CI, 14.2%-48.0%), respectively, in the STARGAZE cohort vs 35.5% (95% CI, 19.2%-54.6%) and 22.6% (95% CI, 9.6%-41.1%), respectively, in the MAGIC cohort. At day 56, in STARGAZE, the lower-GI ORR, durable lower-GI ORR, lower-GI CR rate, and durable lower-GI CR rate were 51.6% (95% CI, 33.1%-69.8%), 45.1% (95% CI, 27.3%-64.0%), 29.0% (95% CI, 14.2%-48.0%), and 29.0% (95% CI, 14.2%-48.0%), respectively. These respective day 56 rates in the MAGIC cohort were 35.5% (95% CI, 19.2%-54.6%), 29.0% (95% CI, 14.2%-48.0%), 16.1% (95% CI, 5.5%-33.7%), and 16.1% (95% CI, 5.5%-33.6%). At day 91, the lower-GI ORR and CR rate were 48.4% (95% CI, 30.2%-66.9%) and 32.3% (95% CI, 16.7%-51.4%), respectively, in the STARGAZE cohort vs 25.8% (95% CI, 11.9%-44.6%) and 16.1% (95% CI, 5.5%-33.7%), respectively, in the MAGIC cohort.

The cumulative NRM incidence at 6 and 12 months was lower in the STARGAZE cohort (6-month NRM, 33.9% [95% CI, 17.7%-50.8%]; 12-month NRM, 38.8% [95% CI, 20.9%-56.4%]) vs the MAGIC cohort (41.9% [95% CI, 24.7%-58.3%]; 51.6% [95% CI, 33.0%-67.4%]).

Furthermore, the 6- and 12-month OS rates were higher in the STARGAZE cohort (6-month OS rate, 58.1% [95% CI, 39.0%-73.1%]; 12-month NRM, 58.1% [95% CI, 39.0%-73.1%]) vs the MAGIC cohort (54.8% [95% CI, 36.0%-70.3%]; 45.2% [95% CI, 27.4%-61.4%]).

The study authors noted that limitations of this research include the use of observational data for the control arm, as well as the study’s small sample size.

Reference

Zeiser R, Ferrara JLM, Louloudis I, et al. Comparisons of glucagon-like peptide-2 analog apraglutide in combination with ruxolitinib for steroid-refractory gastrointestinal acute graft-versus-host disease with MAGIC control cohort: the phase 2 STARGAZE trial. Presented at: 51st Annual EBMT Meeting; March 30 to April 2, 2025; Florence, Italy. Abstract GS2-04.