Patterns of joint involvement in juvenile idiopathic arthritis and prediction of disease course: A prospective study with multilayer non-negative matrix factorization

• Published in: PLoS medicine 2019-02, Vol.16 (2), p.e1002750-e1002750
• Main Authors: Eng, Simon W M, Aeschlimann, Florence A, van Veenendaal, Mira, Berard, Roberta A, Rosenberg, Alan M, Morris, Quaid, Yeung, Rae S M
• Format: Article
• Language: English
• Subjects: Adolescent; Anti-inflammatory agents; Arthritis; Arthritis, Juvenile - diagnosis; Arthritis, Juvenile - epidemiology; Biology and Life Sciences; Care and treatment; Child; Child, Preschool; Children; Classification; Clustering; Cohort Studies; Consortia; Disease; Disease Progression; Epidemiology; Factorization; Female; Health aspects; Homunculus; Hospitals; Humans; Hypothesis testing; Immunology; Infant; Inflammation; International conferences; Joint diseases; Joints (Anatomy); Joints - pathology; Juvenile rheumatoid arthritis; Knee; Learning algorithms; Localization; Machine learning; Male; Medical prognosis; Medical research; Medicine and Health Sciences; Nonsteroidal anti-inflammatory drugs; Patients; Pediatric diseases; Pediatric research; Pediatrics; Pelvis; Permutations; Phenotypes; Polyarthritis; Predictive Value of Tests; Principal components analysis; Prospective Studies; Psoriasis; Psoriatic arthritis; Rank tests; Rheumatoid arthritis; Rheumatoid factor; Rheumatology; Statistical models; Trajectory analysis
• ISSN: 1549-1676; 1549-1277; 1549-1676
• DOI: 10.1371/journal.pmed.1002750

Joint inflammation is the common feature underlying juvenile idiopathic arthritis (JIA). Clinicians recognize patterns of joint involvement currently not part of the International League of Associations for Rheumatology (ILAR) classification. Using unsupervised machine learning, we sought to uncover data-driven joint patterns that predict clinical phenotype and disease trajectories. We analyzed prospectively collected clinical data, including joint involvement using a standard 71-joint homunculus, for 640 discovery patients with newly diagnosed JIA enrolled in a Canada-wide study who were followed serially for five years, treatment-naïve except for nonsteroidal anti-inflammatory drugs (NSAIDs) and diagnosed within one year of symptom onset. Twenty-one patients had systemic arthritis, 300 oligoarthritis, 125 rheumatoid factor (RF)-negative polyarthritis, 16 RF-positive polyarthritis, 37 psoriatic arthritis, 78 enthesitis-related arthritis (ERA), and 63 undifferentiated arthritis. At diagnosis, we observed global hierarchical groups of co-involved joints. To characterize these patterns, we developed sparse multilayer non-negative matrix factorization (NMF). Model selection by internal bi-cross-validation identified seven joint patterns at presentation, to which all 640 discovery patients were assigned: pelvic girdle (57 patients), fingers (25), wrists (114), toes (48), ankles (106), knees (283), and indistinct (7). Patterns were distinct from clinical subtypes (P < 0.001 by χ2 test) and reproducible through external data set validation on a 119-patient, prospectively collected independent validation cohort (reconstruction accuracy Q2 = 0.55 for patterns; 0.35 for groups). Some patients matched multiple patterns. To determine whether their disease outcomes differed, we further subdivided the 640 discovery patients into three subgroups by degree of localization-the percentage of their active joints aligning with their assigned pattern: localized (≥90%; 359 patients), partially localized (60%-90%; 124), or extended (