COMMENT NATURE REVIEWS | RHEUMATOLOGY © 2018 Macmillan Publishers Limited, part of Springer Nature. All rights reserved. Gene therapy has great potential in the treatment of multiple diseases, including osteoarthritis (OA) and rheumatoid arthritis (RA). In the past few years, nota- ble strides have been made towards moving gene ther- apeutics into the clinic and a number of promising approaches are currently being tested in clinical trials. In 2017, the FDA approved three new gene therapeu- tics 1 and the South Korean Ministry of Food and Drug Safety approved the first arthritis gene therapy: Invossa (TissueGene) (TABLE 1). These new developments make it an exciting time in the field of genetic medicine. Intra-articular gene therapy was initially developed as a means of overcoming the pharmacokinetic barriers to delivering biologics to joints. Such agents have diffi- culty entering joints from the systemic circulation and, once in the joint, they are rapidly removed 2 . Indeed, most substances introduced into joints have dwell times of only a few hours, which is inappropriate for diseases such as arthritis that last for decades. Both of these delivery prob- lems can be overcome by introducing cDNAs that encode therapeutic products, such as proteins or non-coding RNA molecules, into the joint. Upon successful gene transfer and expression of the gene in the joint by resid- ing cells, the therapeutic gene products are synthesized endogenously, and continue to be synthesized for poten- tially extended periods of time. Furthermore, unlike their recombinant counterparts, the encoded proteins likely undergo authentic post-translational processing. Genetic modification of joints has been achieved in preclinical models by ex vivo and in vivo strategies using a variety of vectors 2 . The first human trial used an autologous, ex vivo protocol in conjunction with a retro- viral vector to deliver IL-1 receptor antagonist (IL-1Ra) cDNA to metacarpophalangeal joints of patients with RA 2 . This approach had the disadvantage of requiring two interventions, one to harvest the cells and another to return them to the patient. The need to expand the cells under conditions of good manufacturing practice added further complexity and cost to this strategy. Moreover, instances of insertional mutagenesis when using retrovi- ruses in other settings created major barriers to the use of retroviruses in non-lethal conditions such as arthritis. These complications have been overcome in two ways: via the use of either allograft cells (cells derived from another individual) transduced ex vivo or vectors that are delivered directly into the joint (in vivo delivery). The use of allograft cells obviates the need for autol- ogous cell harvest from each patient. Invossa, an allo- geneic cell-mediated gene therapy, is an example of this approach 2 . The cell line for Invossa derives from chondro- cytes harvested from cartilage recovered during surgery to remove a digit from an infant with polydactyly 2 . A popula- tion of these cells have been transduced with a retrovirus encoding transforming growth factor-β (TGFβ). To avoid the possibility of malignancies arising as a result of inser- tional mutagenesis, the transduced cells are irradiated before being injected into patients. The radiation dose is sufficient to prevent cell division, and hence tumour for- mation, without reducing TGFβ production. To amplify and perpetuate the response, the irradiated, transduced cells are mixed with unirradiated, untransduced cells from the same donor before injection into the affected joint. In vivo gene delivery to the joint by direct intra- articular injection is an alternative way to expedite treatment. Although a number of different vectors successfully transduce cells in intra- articular tissues following injection into the joint, many are unsuitable for clinical translation because the vectors are inflamma- tory, immunogenic, unsafe or provide only short-term transgene expression. Adeno-associated virus (AAV) has emerged as a popular vector for in vivo delivery, because the virus is safe, effective and less immunogenic than other vectors 1 . Moreover, AAV provides extended periods of intra-articular transgene expression 2 . When injected into joints, recombinant AAV transduces syno- vial lining cells as well as chondrocytes throughout the thickness of the articular cartilage 3 . The latter capability promises to confer a considerable advantage when treat- ing OA, a disease in which chondrocyte dysfunction has a key role. As the turnover of chondrocytes is low, even in OA, transduction of chondrocytes in situ provides additional potential for long-term transgene expression. Arthritis gene therapy is becoming a reality Christopher H. Evans 1 *, Steven C. Ghivizzani 2 and Paul D. Robbins 3 After more than 25 years of development, arthritis gene therapy is finally entering clinical practice. In South Korea, a gene therapeutic has been approved for the treatment of osteoarthritis, and other gene therapeutics are in the pipeline elsewhere. Genetic medicines for arthritis should enter the rheumatological armamentarium in the foreseeable future. 1 Rehabilitation Medicine Research Center, Mayo Clinic, Rochester, MN, USA. 2 Department of Orthopaedics and Rehabilitation, University of Florida College of Medicine, Gainesville, FL, USA. 3 Department of Molecular Medicine, The Scripps Research Institute, Jupiter, FL, USA. *e-mail: Evans.Christopher@ mayo.edu https://doi.org/10.1038/ s41584-018-0009-5