Abstract
Drug Repurposing and Artificial Intelligence—New Therapies from Old Drugs
Senior author and repeat National Psoriasis Foundation (NPF) grantee Dr. Lam Tsoi and colleagues recently published a manuscript in the Journal of Investigative Dermatology focused on drug repurposing to identify new drug candidates for the treatment of inflammatory diseases such as psoriasis [1]. The authors extracted PubMed abstracts for 2,814 drugs to create a database of word embeddings which took into account the local and global context of each word use. The database was divided into positive (drugs currently used to treat psoriasis) and negative (drugs not used to treat any immune-mediated disease) datasets to “train” a machine learning model. The machine learning model was then applied to a database of 353 drugs that are used to treat immune-mediated diseases, but not psoriasis. The author’s presented 10 drugs that were predicted to be effective treatments for psoriasis including: budesonide, hydroxychloroquine, and cyclophosphamide. Gene targets were predicted for the candidate drugs and compared to gene expression changes observed in psoriatic skin. Most of the candidate drugs exhibited predicted targets consistent with having efficacy for the treatment of psoriatic disease, that is, targeted proteins involved in the pathology of psoriatic disease. The identified drugs will require significant further evaluation; however, if effective, these drugs would have a much quicker path to approval due to their usage in other diseases and documented side effect profiles. 1
Drug Pipeline
On October 31, 2018, Sandoz, a division of Novartis, announced approval of their biosimilar adalimumab-adaz for the treatment of psoriatic arthritis and plaque psoriasis. Approval was also granted for the treatment of rheumatoid arthritis, juvenile idiopathic arthritis in patients 4 years of age and older, ankylosing spondylitis, adult Crohn’s disease, and ulcerative colitis. Adalimumab-adaz adds to the list of 6 other tumor necrosis factors targeting biosimilars approved for the treatment of psoriasis and psoriatic arthritis.
Press release: https://www.novartis.com/news/media-releases/sandoz-receives-us-fda-approval-biosimilar-hyrimoztm-adalimumab-adaz (Accessed November 16, 2018).
Cutting-Edge Research and Treatment
Psoriatic Arthritis—Identifying Potential Risk Factors
The NPF fellowship awardee Di Yan, MD, with mentor and head of NPF’s scientific advisory committee Wilson Liao, MD and colleagues published a manuscript in Dermatology and Therapy exploring risk factors associated with psoriatic arthritis in a cohort of 974 psoriasis patients. 2 Both simple and a more complex analysis found similar associations. Age, scalp psoriasis, elbow/knee psoriasis, nail involvement, skinfold involvement, plaque subtype, pustular subtype positivity, erythrodermic subtype, severe psoriasis, hypertension, type 2 diabetes, and coronary artery disease were associated with a diagnosis of psoriatic arthritis in one or both analyses. By contrast, having one or more copies of the HLA-C*06:02 allele was negatively associated with psoriatic arthritis in both analyses. The identification of specific markers of disease may also identify novel treatment targets. The ability to accurately predict and diagnose psoriatic arthritis is an important step toward increasing treatment efficacy and disease prevention. Consistent with this, NPF has identified the diagnosis of psoriatic arthritis as a critical need deserving immediate attention and has recently initiated a new funding program to develop a diagnostic test within 5 years (https://www.psoriasis.org/grants).
Does Online Care Work?
The NPF grantee and fellowship mentor April Armstrong, MD, along with several other current or former NPF grantees and colleagues published a study in JAMA Network Open that compares treatment outcomes for online versus in-person care of psoriasis. 3 The study included 296 participants and measured the mean change in Psoriasis Area and Severity Index (PASI), body surface area (BSA) affected and the patient global assessment score. Both the PASI and BSA measures were equivalent between the 2 treatment modalities. Interestingly, participants in the online cohort exhibited more improvement in the patient global assessment compared with the in-person care group. However, the authors note several important considerations when implementing a connected-health model including the quality of the communication technology, the cost and medicolegal considerations, limitations due to a workforce shortage of dermatologists and their potential reluctance to provide online care, and the need for a local telehealth “champion” to ensure an effective telehealth experience. Overall, the results indicate that online care may be as efficacious as in-person care for the treatment of psoriasis.