Circulating cell free tumor DNA (ctdna) in patients with immune checkpoint inhibitor associated inflammatory arthritis (ICI-IA): developing a biomarker to assess immunosuppressive drug safety

The National Institute of Arthritis and Musculoskeletal and Skin Diseases awarded the Hospital for Special Surgery Fund, Inc. (doing business as HSS Foundation) $473,752 on September 16, 2025, for research developing a circulating cell-free tumor DNA (ctDNA) biomarker to assess immunosuppressive drug safety in patients with immune checkpoint inhibitor-associated inflammatory arthritis (ICI-IA), under the Arthritis, Musculoskeletal and Skin Diseases Research program (CFDA 93.846).

The research addresses a critical clinical gap: immune checkpoint inhibitors (ICIs) improve cancer survival in advanced cancer patients but commonly cause inflammatory arthritis requiring immunosuppressive treatment that may worsen cancer outcomes. Current radiographic imaging for detecting cancer recurrence occurs months after ICI-IA treatment initiation, making it difficult to attribute recurrence to specific medications. The project employs MRD-EDGE, an ultra-sensitive approach coupling whole genome sequencing with deep learning to detect circulating tumor DNA at sensitivity as low as 10-5 and track cancer ctDNA over time, even in patients with low tumor burden. Critically, this method does not require matched tumor tissue reference samples, addressing the practical constraint that simultaneous tumor sampling is infeasible in rheumatology clinics. The study leverages plasma samples from patients enrolled in existing large prospective immune checkpoint inhibitor cohorts.

Performance occurs in New York, New York, with a period of performance ending August 31, 2028.

Abstract

Immune checkpoint inhibitors (ICI) are monoclonal antibodies that block regulators of t cell activation, activating anti-cancer immunity and greatly improving overall survival in patients with advanced cancer. However, ICI commonly cause immune related adverse events (irae), including inflammatory arthritis (ICI-IA). IT is estimated that as many as 14,000 ICI-treated patients will develop ICI-IA each year. ICI-IA can i. Be destructive to the joint, II. Persist for months to years, and III. Require treatment with glucocorticoids and immunosuppressive disease modifying antirheumatic drugs (dmards). However, immunosuppression may worsen cancer survival in this setting. Identifying the safest approach to ICI-IA management (with regard to tumor control) is a critical unmet need. However, cancer recurrence is measured on radiographic imaging which is often performed months after ICI-IA treatment is initiated and this makes IT difficult to disentangle which medication or intervention may have been responsible. In this study, we use MRD-edge, a new ultra- sensitive approach to measure circulating cell free tumor DNA (ctdna) and identify patients with cancer recurrence long before IT is demonstrated on standard of care radiographic imaging. MRD-edge couples whole genome sequencing with deep learning approaches to provide a tumor fraction detection sensitivity as low as 10-5 and allow tracking of cancer ctdna over time, even in patients with a low tumor burden. Even more critically, this approach does not require reference to matched tumor tissue. This is important because simultaneous tumor sampling is impractical in most clinical settings (e.g., rheumatology clinic). This study takes advantage of plasma samples from patients enrolled in one of two large prospective ICI-IA registries, the hospital for special surgery rheumatic irae registry (2018-2022), and the multicenter rheumatology adverse events due to immunotherapy observational study (radios) registry (first enrollment 1/2023). In aim 1, we measure ctdna levels in patients with melanoma or non-small cell lung cancer (nsclc) at their first visit for ICI-IA to quantify the association between ctdna and subsequent cancer progression. In aim 2, we measure changes in ctdna from baseline to weeks 4, 8 and 12 after institution of glucocorticoids in ICI-IA patients, to assess the sensitivity of this biomarker to the effects of immunosuppression, and the kinetics of change. Our overarching hypothesis is that ctdna can be detected in patients with ICI-IA long before cancer progression is demonstrated on imaging, and that any rise in ctdna will be detectable within 4 weeks of initiating glucocorticoids. This study will provide preliminary data and optimal plasma sampling times for a future interventional study (e.g., comparing up-front glucocorticoids to up-front IL6R inhibition for high-grade irae). If effect sizes are small, we will plan a much larger multicenter observational study powered to assess ctdna levels before and after steroid-sparing immunosuppressive agents including IL6 and TNF inhibitors.