Novel Gene Signature Found in Cutaneous Chronic GVHD Responders to Topical Ruxolitinib

Alina Markova, MD

A differentially expressed gene profile was identified in patients with cutaneous nonsclerotic and superficially sclerotic chronic graft-vs-host-disease (GVHD) who responded to treatment with topical ruxolitinib (Jakafi), according to findings from an exploratory analysis of a phase 2 study (NCT03954236) presented at the 2022 Transplantation & Cellular Therapy Meetings.

In comparing the gene expression profiles of responders (n = 8) and nonresponders (n = 3) to topical ruxolitinib, investigators identified 383 differentially expressed genes, with a fold-change of greater than 2 between populations (P < .01). The top 10 differentially expressed genes were VPS35, MCM5, MAP3K15, CCNB3, AL008638.1, DQX1, AC011525.1, MT1F, NKX1-1, and ALPK2.

Notably, the profiles were easily distinguishable in a heat map, showing hierarchical clustering of responder and nonresponder samples using differentially expressed genes, reflecting a novel gene signature for responders.

“This is the first study to characterize the effect of topical JAK1/2 blockade with ruxolitinib cream on cutaneous GVHD. [These] data support the need for deeper interrogation to understand the clinical significance for patients with cutaneous nonsclerotic and superficially sclerotic chronic GVHD,” lead study author Alina Markova, MD, director of inpatient consultative dermatology and codirector of the Multidisciplinary GVHD Clinic with the BMT Service at Memorial Sloan Kettering Cancer Center, said in the presentation.

There are currently no FDA-approved topical therapies for cutaneous GVHD. Oral ruxolitinib is approved for the treatment of steroid-refractory acute and chronic GVHD following failure of 1 or 2 prior lines of therapy in patients at least 12 years of age. Topical ruxolitinib 1.5% cream is approved for the treatment of mild to moderate atopic dermatitis in nonimmunocompromised patients at least 12 years of age whose disease is not adequately managed with topicals or is contraindicated for other topicals.

In the prospective, double-blind, phase 2 randomized controlled trial, investigators evaluated the efficacy and safety of topical ruxolitinib 1.5% vs vehicle cream for cutaneous chronic GVHD.

To be eligible for enrollment, patients had to have history of allogeneic hematopoietic stem cell transplant, be at least 12 years of age, and have at least 2% body surface area of clinically or histologically confirmed nonsclerotic or superficially sclerotic cutaneous chronic GVHD. Additionally, systemic therapy for GVHD had to be stable for 4 weeks prior to enrollment, and concurrent topical therapies had to be discontinued on study day 0. 

Patients were randomly assigned 1:1 to vehicle cream on the left side of the body and topical ruxolitinib on the right side of the body, or topical ruxolitinib on the left side of the body and vehicle cream on the right side of the body. At day 28, patients were allowed to continue with open-label treatment with topical ruxolitinib. All agents were administered twice daily in both arms.

The primary outcome was the efficacy of ruxolitinib vs vehicle cream at day 28 according to differences in body surface area. Secondary outcomes included the composite assessment of index lesion severity (CAILS) and physician’s global assessment of clinical condition (PGA) of the ruxolitinib-treated site vs the vehicle cream–treated site.

Thirteen patients were randomized from June 2019 to May 2021. The median age was 52.6 years, and most patients were women (n = 7; 53%). Most patients had a primary diagnosis of acute leukemia (n = 8; 62%), followed by non-Hodgkin lymphoma (n = 3; 23%), myeloproliferative neoplasm (n = 1; 8%), and aplastic anemia (n = 1; 8%).

The median time from transplant to enrollment was 665 days (interquartile range [IQR], 433-1355), and the median time from chronic GVHD onset to enrollment was 283 days (IQR, 115-867).

Regarding GVHD characteristics, most patients had GVHD classified as severe (n = 8; 62%) and classic (n = 11; 85%). The number of organs involved ranged from 1 (n = 4; 31%) to at least 4 (n = 6; 46%), and more than half of patients had received 2 prior lines of therapy (n = 7; 54%).

Within the nonsclerotic subtype (n = 10; 77%), most patients had Lichen planus–like disease (n = 8; 62%); all patients within the sclerotic subtype had Lichen sclerosus-–ike disease (n = 3; 23%).

The most common prior line of therapy was topical corticosteroids (n = 10; 77%).

Prior interim results of the primary outcome showed a nonsignificant reduction in body surface area of the ruxolitinib-treated site vs the vehicle cream–treated site at day 14 (P = .06) and day 28 (P = .15). Results of a secondary outcome assessment showed a statistically significant improvement in CAILS score with topical ruxolitinib at day 14 (P = .02) and an insignificant trend toward improvement at day 28 (P = .09). Futher, statistically significant improvements in PGA score with ruxolitinib were observed at day 14 (P = .024) and day 28 (P = .026).

An exploratory analysis included the evaluation of molecular expression profiles of lesional GVHD skin with the DermTech Smart Sticker following treatment with topical ruxolitinib or vehicle.

“The technology is a form of skin stripping that allows us to collect 1.5 mg of lesional and non-lesional stratum corneum tissue and capture genomic material from epidermal keratinocytes, T cells, dendritic cells, and melanocytes,” Markova said.

To determine differential expression between the treatment arms, the noninvasive skin samples were collected from all patients: 11 in the ruxolitinib arm and 11 in the vehicle arm. A total of 310 genes were differentially expressed, with a fold-change of greater than 2 between arms (P < .01). The top 10 differentially expressed genes were CAPN12, KRT84, FP236383.7, CTNNB1, TCF7, HNRNPU, ZNF664, GARS1, GIGYF2, and SLC44A1.

Principle component analysis, which was done to provide a visual representation of differentially expressed genes in the ruxolitinib and vehicle samples, demonstrated distinct clustering. The top ranked pathways included keratinization, transcriptional regulation by RUNX3, and NOTCH3 activation and transmission of signal to the nucleus. Others included binding of TCF/LEF:CTNNB1 to target gene promoters, RUNX3regulation of WNT signaling, deactivation of the β-catenin transactivating complex, and oncogene induced senescence.

Reference

1. Markova A, Whitaker JW, Ai R, et al. Noninvasive genomic characterization of patients with nonsclerotic and superficially sclerotic chronic cutaneous graft-versus-host disease identified a novel gene signature in responders to ruxolitinib cream. Presented at: 2022 Transplantation & Cellular Therapy Meetings; April 23-26, 2022. Abstract 33.