Osteoarthritis Phenotyping Makes Progress

Clinical medicine ― and rheumatology in particular ― is frequently about pattern recognition. As we age in our profession and as our patient panels grow, we develop a gestalt, so that the most seasoned of us can recognize polymyalgia rheumatica, rheumatoid arthritis, psoriatic arthritis, or gout within a few minutes of meeting a patient.

Dr Karmela Chan

But osteoarthritis (OA) confounds us. Why do some people have more pain than others? Is knee OA biologically the same as spine degenerative disc disease? How is it that some folks have horribly misshapen hands but feel no pain at all? And speaking of hands, what is erosive osteoarthritis even?

OA has become a leading contributor to years lived with disability, as tracked by the Global Burden of Disease Collaborative Network. Yet our treatment options are embarrassingly limited.

To improve outcomes in OA, we need to do a better job of defining it and understanding mechanisms behind it. Though we do not typically think of "phenotypes" when we think of OA, we also instinctively know that the "wear-and-tear" conception of OA is not the whole picture either, as illustrated by those pesky Heberden nodes.

Historically, many attempts have been made to phenotype OA based on a variety of dimensions, such as joint involvement, pain, psychological profiles (eg, depression, coping ability, catastrophizing behavior), comorbidities (metabolic syndrome), imaging features, or disease trajectory. A systematic literature review on this topic by Deveza et al is worth a read. Another possible way to define disease is via the endotype, which is the molecular mechanism that gives rise to a disease. Possible endotypes in OA, for example, might be whether it starts in the bone, chondrocyte, or synovium.

In the past decade, more advanced imaging and lab methods, statistical tools, and machine learning have contributed to more sophisticated phenotyping attempts. This work has become so relevant and critical that in 2020, a panel of 25 osteoarthritis researchers published a framework for doing this work. Defining phenotypes and endotypes will lead to a better understanding of the natural history of the various types of OA and allow for better treatment paradigms rather than the limited and rather blunt instruments of pain management and joint replacement surgery that we currently use.

At the 2021 ACR Convergence meeting, there were two notable abstracts presenting what I thought were elegant attempts at phenotyping and endotyping. Both involved looking at synovial tissue histology from patients undergoing joint replacement therapy.

Abstract 0212, from my colleagues at Hospital for Special Surgery, examined ultrasound and histologic features from 157 subjects undergoing knee arthroplasty. Using unsupervised clustering of 23 ultrasound and 35 histologic features, they were able to group patients into three subtypes with distinct features, including one subtype with features of synovial inflammation. The authors concluded that "these integrative approaches have the potential to generate prognostic platforms that can in turn guide targeted treatments."

Abstract 0943 by Jessica Murillo Saich et al looked at cytokine and metabolomic profiles of 37 synovial tissue samples (28 knees and nine hips). After grading the samples on inflammation and fibrosis scores, the researchers found that inflammatory synovial tissue had a distinct metabolic profile and that interleukin (IL)-6, but not tumor necrosis factor or IL-1 beta, correlated positively with inflammation.

The hope, of course, is that the better we understand the pathobiology of the disease, the better we can target treatment. There are no approved disease-modifying drugs yet, but there are certainly more candidates now, with more basis in science, than there have been before. Some notable examples include the recombinant human fibroblast growth factor sprifermin, which showed improvement in cartilage thickness over a 2-year period but not any change in pain scores. Tanezumab, an anti-nerve growth factor antibody, has demonstrated efficacy at reducing pain scores, although it has thus far failed to get Food and Drug Administration approval because of the development of rapidly progressive OA in a subset of patients. At the 2021 ACR Convergence meeting, lorecivivint, a WNT pathway modulator, demonstrated benefit for pain up to 12 months following one intra-articular injection (abstract 0729).

These attempts have not been uniformly successful, of course. For example, despite preclinical and clinical evidence that IL-1 is elevated in the synovium of OA patients, a phase 2 trial of the subcutaneous IL-1 alpha/beta blocker lutikizumab failed to show improvement in synovitis. Similarly, in data presented at the 2021 ACR Convergence meeting (abstract 0728), intra-articular canakinumab, an IL-1 beta blocker, was no better than placebo at reducing pain in patients with knee OA.

These mixed and negative results underscore the importance of defining OA better in order to direct the appropriate treatment toward the appropriate population. There is so much work being done in this field at the moment, and it makes me hopeful that in our lifetimes we will have something to offer our patients with OA.

Karmela Kim Chan, MD, is an assistant professor at Weill Cornell Medical College and an attending physician at Hospital for Special Surgery and Memorial Sloan Kettering Cancer Center in New York City. Before moving to New York City, she spent 7 years in private practice in Rhode Island and was a columnist for a monthly rheumatology publication, writing about the challenges of starting life as a full-fledged rheumatologist in a private practice. Follow her on Twitter.

Follow Medscape on Facebook, Twitter, Instagram, and YouTube.

Comments

Commenting is limited to medical professionals. To comment please Log-in.

Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.