Efficacy and Safety of Baricitinib in Refractory Chronic Graft-Versus-Host Disease (cGVHD): Preliminary Analysis Results of a Phase 1/2 Study

Background: Chronic Graft-versus-Host Disease (cGVHD) is a multi-organ disorder characterized by immune dysregulation, impaired organ function, and decreased survival. Refractory cGVHD remains an unmet need with limited effective treatment options. Janus kinase (JAK) inhibition targeting JAK-STAT is promising in cGVHD due to this multifaceted pathway's key role in immune cell development, function, and inflammation. Baricitinib is a JAK1/2 inhibitor FDA-approved at 2 mg daily for treatment of rheumatoid arthritis, and which has shown therapeutic promise in preclinical models of GVHD. Herein, we report the preliminary analysis results of a phase 1/2 study exploring the efficacy and safety of the JAK 1/2 inhibitor baricitinib in therapy-refractory cGVHD.

Methods: Patients (pts) with therapy-refractory moderate-severe cGVHD per NIH criteria were enrolled on an NCI sponsored treatment protocol (NCT02759731). This is a phase 1/2 single center, single arm study of baricitinib starting at an oral dose of 2 mg daily, given for 28-day cycles, with intra-patient dose escalation to 4 mg daily after 3 months (mos) for a total of 12 cycles. Adverse events (AEs) were monitored and graded per CTCAEv4.3. Clinical responses were measured every 3 mos per the 2014 NIH cGVHD Response Criteria. The primary efficacy endpoint was ORR (CR+PR) at 6 mos; if PR or SD, pts were treated for up to 12 mos and monitored for 12 mos post-completion. Subjects self-report questionnaires included the Lee symptom bother scale (LSS). Peripheral blood samples were collected for pharmacokinetic and correlative studies at several time points throughout study.

Results: Twenty pts with refractory cGVHD were enrolled. Median patient age was 54 years (24-68), 55% female, 90% Caucasian. Median time from cGVHD diagnosis to study enrollment was 36.8 mos, with median baseline KPS of 80%. cGVHD NIH global score was severe in all pts (due to sclerotic skin in all except one oral), involving a median of 4 (2-6) organs, and treated with a median of 4 (2-7) lines of prior systemic therapy including ibrutinib in 3. At study enrollment, all but 3 (85%) were on concurrent systemic immunosuppression, including steroids for 12 (60%) at a median dose of 10 mg (2.5-30) prednisone daily.

ORR at 6-months was 63% (95% CI 47-87%), with durable responses seen in 7 of 8 evaluable pts at 12-mos. ORR at any time was 90% (85-100%), with median time to ORR of 1.4 mos (1.4-6.3). Organ-specific responses were seen in all involved organs except for lungs (11 organ-specific CRs at 6-mos). Clinically significant improvements in LSS (≥ 6-point reduction) were seen in 7 of 14 (50%) evaluable pts at 6-mos. Median steroid dose-reduction of 4 mg prednisone daily was achieved in 6 (50%) of those on steroids at enrollment. With a median follow up of 24.2 mos, median failure-free survival (FFS) was 20.6 mos (19-NR). 1- and 2-year FFS was 74% (57-97%) and 37% (18-78%), respectively. Of the 8 evaluable pts who completed study, 3 progressed at a median of 7.3 mos (6.9-11.9) after drug discontinuation.

No DLT was observed with the 2 mg dose of baricitinib; 16/20 (80%) pts reached dose-escalation to 4 mg at 3 mos. Three pts required dose-reductions: two due to neutropenia, and one due to refractory myalgias. Common AEs, possibly treatment-related, included upper respiratory infection (URI) (13), neutropenia (6), hypophosphatemia (12), and hypertriglyceridemia (5). Asymptomatic, transient reactivation of CMV occurred in 6, EBV in 7, BK viruria in 5 - none of which required treatment. HPV-related eyelid papilloma developed in 2, with no visual changes. One patient with EBV viremia/lymphadenopathy at enrollment was diagnosed with biopsy-proven EBV-PTLD <1 cycle into study. 11 SAEs were reported, of which 5 were possibly drug-related, including hospitalizations for joint infection (n=1), cellulitis (n=2), and URI (n=2). Eleven pts (55%) required dose-interruptions due to AEs, and 9 pts (45%) discontinued treatment at a median 5.6 mos (0.8-10.7) post-initiation due to PD (n=3), EBV-PTLD (n=1), decreased DLCO (n=1), viral URI (n=1), pneumonia (n=1), and other (n=2). There were no deaths on study.

Conclusion: The JAK 1/2 inhibitor baricitinib appears well-tolerated and effective for pts with severe cGVHD who progressed through multiple lines of prior therapy. Responses were achieved rapidly and seen across all involved organs, except lungs. Full study results for a total of 31 pts are expected in early 2022.