Anti-citrullinated protein antibodies (ACPA) are a key biomarker in rheumatoid arthritis (RA), but a recent study has suggested that ACPA studies may not fully identify the biological distinctions between patients with RA who are ACPA-negative or ACPA-positive, and that testing for additional biomarkers could be warranted to help target early treatments. 

“A diagnostic test that is based on high-dimensional profiling of blood proteins and metabolites could be useful for early diagnosis of rheumatoid arthritis when ACPA testing is negative,” senior study author John Davis III, MD, chair of rheumatology at Mayo Clinic in Rochester, Minnesota, told Medscape Medical News. “However, further development and validation are necessary before this can be useful in the clinic.”

When asked to comment on the study, Ted Mikuls, MD, MSPH, professor of rheumatology at the University of Nebraska Medical Center, Omaha, Nebraska, told Medscape Medical News that “these data really provide molecular proof of a real-world observation that many of us have shared since ACPA testing first became available; that is, ACPA-negative RA is really a distinct beast from seropositive disease.”

Although it may be too early to take anything from the study into the clinic, Mikuls said, “It gives me hope that we will soon have more powerful tools that will empower rheumatologists and other health care providers to eliminate diagnostic delays that are still so common, especially in seronegative RA, and allow us to provide the most effective treatments to our patients as early in the disease process as possible.”

The retrospective, observational cohort study, published in the journal Frontiers in Immunology, evaluated the effectiveness of high-throughput proteomic and metabolomic profiling on plasma from a cohort of 120 patients, divided evenly among those with ACPA-negative RA, ACPA-positive RA, and healthy control patients. The goal was to improve the biological understanding of heterogeneity in RA and gather data that would enable the development of more precise approaches for diagnosing and stratifying RA. 

“Seronegative RA is harder to confirm early because standard serology (ACPA/rheumatoid factor) is negative or low, so clinicians must lean on evolving symptoms and imaging, often delaying diagnosis and treatment,” said lead author Jaeyun Sung, PhD, computational biologist and rheumatology researcher at Mayo Clinic in Rochester.

Potential of Multi-omics Profiling

The diagnostic sensitivity of antibody testing for ACPA has been reported to range from about 29% to 58%, depending on the antibody isotype. The Mayo study demonstrated the potential of using multi-omics profiling to improve diagnosis and management of RA earlier in the course of the disease, he said.

“Multi‑omics blood profiling — proteins and metabolites in our study — can distinguish RA and, importantly, ACPA-negative RA from healthy controls with high accuracy, while also revealing immune and metabolic pathways that differ by ACPA status,” Sung said. “This argues support for digital biomarker panels that not only are more insightful than current practices based on ACPA-based serology but also aid disease diagnosis and subgrouping.”

The study used proteomic profiling to measure 7273 human proteins, including cytokines, growth factors, proteases, and hormones. It also used ultra-high performance liquid chromatography-tandem mass spectrometry for untargeted profiling of 1061 metabolites.

The study found that levels of five complement proteins (A2M, CFB, CFD, CFHR5, and F9) and two cytokines (IL1RN and TNFRSF17) were higher in ACPA-negative patients than in others, whereas only one complement protein (C9) was high in ACPA-positive patients in a subcohort that hadn’t had any treatment, he said.

The metabolomic analysis observed significant enrichment of pathways related to lipid metabolism (carnitine, sphingosine, and medium-chain fatty acid metabolic pathways) in ACPA-negative RA and those related to pyrimidine and endocannabinoid metabolism in ACPA-positive disease. Conversely, metabolites that were reduced in abundance in both ACPA-negative and ACPA-positive RA compared with control patients were significantly enriched in arginine and proline metabolic pathways, suggesting shared alterations in urea cycle-linked metabolism.

The study also identified a number of correlates with inflammatory burden, Sung said. MMP19 correlated most strongly with erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) in ACPA-negative RA but not in ACPA-positive RA, “nominating it as a candidate marker of inflammatory activity specific to the ACPA-negative subgroup,” he said. Bilirubin‑derived metabolites correlated negatively with ESR and CRP in ACPA-negative RA but not in ACPA-positive disease, which is “consistent with prior reports that higher bilirubin associates with lower RA activity,” Sung said.

Why Confirming Seronegativity Early Matters

Confirming seronegative RA early in the course of the disease “is essential to optimizing the prognosis, in terms of enhancing the probability of achieving the target of low disease activity or remission on DMARD [disease-modifying antirheumatic drug] therapy, preventing structural joint damage, and minimizing the risk of respiratory and cardiovascular complications,” Davis said.

The study’s findings could give rise to new hypotheses about mechanisms driving autoimmunity and chronic inflammation in ACPA-negative patients with RA, Davis added. “This could potentially lead to identification of new therapeutic targets for ACPA-negative patients that could improve treatment outcomes.”

The study has a number of strengths and limitations, Sung said. Among its strengths were the number of proteins and metabolites measured; adjustments for potentially confounding factors such as age, sex, smoking status, and RA medication use; the subanalysis of treatment-naive patients; and the use of network-based machine learning and deidentified datasets.

Limitations Sung noted included the need for external validation; predominantly White participant cohorts; unknown variations among participants’ therapies; the lack of a separate cohort; and limited data on metabolic comorbidities and lifestyle factors. “These could affect metabolomic profiles,” he said.

Sung reported having a licensing agreement with Rheumasense. Davis reported receiving a research grant from Pfizer and having licensing agreements with Rheumasense and Remission Medical. Both Sung and Davis had a nonprovisional US patent application for a biomarker for seronegative RA. Mikuls was a consultant for UCB, Olatec Therapeutics, and Merck and reported receiving past research support from Horizon Therapeutics/Amgen. 

Richard Mark Kirkner is a medical journalist based in Philadelphia.