A E Nelson1, F Fang2, L Arbeeva3, R J Cleveland4, T A Schwartz5, L F Callahan6, J S Marron7, R F Loeser8. 1. Thurston Arthritis Research Center, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA; Department of Medicine, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA. Electronic address: aenelson@med.unc.edu. 2. Department of Statistics and Operations Research, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA. Electronic address: fangf@live.unc.edu. 3. Thurston Arthritis Research Center, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA. Electronic address: liubov_arbeeva@med.unc.edu. 4. Thurston Arthritis Research Center, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA; Department of Medicine, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA. Electronic address: Rebecca_cleveland@med.unc.edu. 5. Thurston Arthritis Research Center, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA; Department of Biostatistics, Gillings School of Global Public Health, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA. Electronic address: tschwart@email.unc.edu. 6. Thurston Arthritis Research Center, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA; Department of Medicine, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA. Electronic address: leigh_callahan@med.unc.edu. 7. Department of Statistics and Operations Research, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA. Electronic address: marron@unc.edu. 8. Thurston Arthritis Research Center, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA; Department of Medicine, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA. Electronic address: Richard_loeser@med.unc.edu.
Abstract
OBJECTIVE: Knee osteoarthritis (KOA) is a heterogeneous condition representing a variety of potentially distinct phenotypes. The purpose of this study was to apply innovative machine learning approaches to KOA phenotyping in order to define progression phenotypes that are potentially more responsive to interventions. DESIGN: We used publicly available data from the Foundation for the National Institutes of Health (FNIH) osteoarthritis (OA) Biomarkers Consortium, where radiographic (medial joint space narrowing of ≥0.7 mm), and pain progression (increase of ≥9 Western Ontario and McMaster Universities Osteoarthritis Index [WOMAC] points) were defined at 48 months, as four mutually exclusive outcome groups (none, both, pain only, radiographic only), along with an extensive set of covariates. We applied distance weighted discrimination (DWD), direction-projection-permutation (DiProPerm) testing, and clustering methods to focus on the contrast (z-scores) between those progressing by both criteria ("progressors") and those progressing by neither ("non-progressors"). RESULTS: Using all observations (597 individuals, 59% women, mean age 62 years and BMI 31 kg/m2) and all 73 baseline variables available in the dataset, there was a clear separation among progressors and non-progressors (z = 10.1). Higher z-scores were seen for the magnetic resonance imaging (MRI)-based variables than for demographic/clinical variables or biochemical markers. Baseline variables with the greatest contribution to non-progression at 48 months included WOMAC pain, lateral meniscal extrusion, and serum N-terminal pro-peptide of collagen IIA (PIIANP), while those contributing to progression included bone marrow lesions, osteophytes, medial meniscal extrusion, and urine C-terminal crosslinked telopeptide type II collagen (CTX-II). CONCLUSIONS: Using methods that provide a way to assess numerous variables of different types and scalings simultaneously in relation to an outcome of interest enabled a data-driven approach that identified key variables associated with a progression phenotype.
OBJECTIVE:Knee osteoarthritis (KOA) is a heterogeneous condition representing a variety of potentially distinct phenotypes. The purpose of this study was to apply innovative machine learning approaches to KOA phenotyping in order to define progression phenotypes that are potentially more responsive to interventions. DESIGN: We used publicly available data from the Foundation for the National Institutes of Health (FNIH) osteoarthritis (OA) Biomarkers Consortium, where radiographic (medial joint space narrowing of ≥0.7 mm), and pain progression (increase of ≥9 Western Ontario and McMaster Universities Osteoarthritis Index [WOMAC] points) were defined at 48 months, as four mutually exclusive outcome groups (none, both, pain only, radiographic only), along with an extensive set of covariates. We applied distance weighted discrimination (DWD), direction-projection-permutation (DiProPerm) testing, and clustering methods to focus on the contrast (z-scores) between those progressing by both criteria ("progressors") and those progressing by neither ("non-progressors"). RESULTS: Using all observations (597 individuals, 59% women, mean age 62 years and BMI 31 kg/m2) and all 73 baseline variables available in the dataset, there was a clear separation among progressors and non-progressors (z = 10.1). Higher z-scores were seen for the magnetic resonance imaging (MRI)-based variables than for demographic/clinical variables or biochemical markers. Baseline variables with the greatest contribution to non-progression at 48 months included WOMAC pain, lateral meniscal extrusion, and serum N-terminal pro-peptide of collagen IIA (PIIANP), while those contributing to progression included bone marrow lesions, osteophytes, medial meniscal extrusion, and urine C-terminal crosslinked telopeptide type II collagen (CTX-II). CONCLUSIONS: Using methods that provide a way to assess numerous variables of different types and scalings simultaneously in relation to an outcome of interest enabled a data-driven approach that identified key variables associated with a progression phenotype.
Authors: D T Felson; J Niu; T Neogi; J Goggins; M C Nevitt; F Roemer; J Torner; C E Lewis; A Guermazi Journal: Osteoarthritis Cartilage Date: 2015-09-30 Impact factor: 6.576
Authors: Frank W Roemer; C Kent Kwoh; Michael J Hannon; David J Hunter; Felix Eckstein; Tomoko Fujii; Robert M Boudreau; Ali Guermazi Journal: Arthritis Rheumatol Date: 2015-05 Impact factor: 10.995
Authors: H An; J S Marron; T A Schwartz; J B Renner; F Liu; J A Lynch; N E Lane; J M Jordan; A E Nelson Journal: Osteoarthritis Cartilage Date: 2015-11-24 Impact factor: 6.576
Authors: A Guermazi; F Eckstein; D Hayashi; F W Roemer; W Wirth; T Yang; J Niu; L Sharma; M C Nevitt; C E Lewis; J Torner; D T Felson Journal: Osteoarthritis Cartilage Date: 2015-07-08 Impact factor: 6.576
Authors: F W Roemer; A Guermazi; M K Javaid; J A Lynch; J Niu; Y Zhang; D T Felson; C E Lewis; J Torner; M C Nevitt Journal: Ann Rheum Dis Date: 2008-10-01 Impact factor: 19.103
Authors: Martin Englund; Ali Guermazi; Frank W Roemer; Piran Aliabadi; Mei Yang; Cora E Lewis; James Torner; Michael C Nevitt; Burton Sack; David T Felson Journal: Arthritis Rheum Date: 2009-03
Authors: Frank W Roemer; Ali Guermazi; Jamie E Collins; Elena Losina; Michael C Nevitt; John A Lynch; Jeffrey N Katz; C Kent Kwoh; Virginia B Kraus; David J Hunter Journal: BMC Musculoskelet Disord Date: 2016-11-10 Impact factor: 2.362
Authors: Hikmat N Daghestani; Joanne M Jordan; Jordan B Renner; Michael Doherty; A Gerry Wilson; Virginia B Kraus Journal: PLoS One Date: 2017-12-29 Impact factor: 3.240
Authors: David J Hunter; Leticia A Deveza; Jamie E Collins; Elena Losina; Jeffrey N Katz; Michael C Nevitt; John A Lynch; Frank W Roemer; Ali Guermazi; Michael A Bowes; Erik B Dam; Felix Eckstein; C Kent Kwoh; Steve Hoffmann; Virginia B Kraus Journal: Arthritis Care Res (Hoboken) Date: 2022-05-03 Impact factor: 5.178
Authors: Jonas Bianchi; Antônio Carlos de Oliveira Ruellas; João Roberto Gonçalves; Beatriz Paniagua; Juan Carlos Prieto; Martin Styner; Tengfei Li; Hongtu Zhu; James Sugai; William Giannobile; Erika Benavides; Fabiana Soki; Marilia Yatabe; Lawrence Ashman; David Walker; Reza Soroushmehr; Kayvan Najarian; Lucia Helena Soares Cevidanes Journal: Sci Rep Date: 2020-05-15 Impact factor: 4.379