Poster Biomarkers

P039 - Association Between Osteoarthritis-related Serum Biomarkers over 11 Years and Knee MRI-biomarkers in Middle-aged Adults

Presentation Topic
Biomarkers
Date
13.04.2022
Lecture Time
09:30 - 09:30
Room
Exhibition Foyer
Session Name
7.3 - Poster Viewing / Coffee Break / Exhibition
Session Type
Poster Session
Speaker
  • A. Singh (Hobart, AU)
Authors
  • A. Singh (Hobart, AU)
  • L. Blizzard (Hobart, AU)
  • A. Venn (Hobart, AU)
  • G. Jones (Hobart, AU)
  • J. Burgess (Hobart, AU)
  • V. Parameswaran (Hobart, AU)
  • F. Cicuttini (Melbourne, AU)
  • L. March (Sydney, AU)
  • C. Ding (Hobart, AU)
  • B. Antony (Hobart, AU)
Disclosure
No Significant Commercial Relationship

Abstract

Purpose

There is a lack of data on osteoarthritis (OA)-related serum biomarkers' (cartilage oligomeric matrix protein (COMP), matrix metalloproteinase (MMP)-3, and hyaluronan (HA)) association with MRI-markers in the middle-aged general population. Therefore, our aim was to investigate the associations between OA-related biomarkers and MRI-markers in middle-aged adults followed up over 10-13 years.

Methods and Materials

Blood samples were collected during the Childhood Determinants of Adult Health (CDAH)-1 study at baseline (year: 2004-06, age: 26–36 years) and 10-13 year follow-up (CDAH-3; year: 2014–2019, age: 36–49 years). Serum samples from baseline (n=156) and follow-up (n= 167) were analyzed for three OA-related biomarkers – COMP, MMP-3, and HA– using non-isotopic ELISA. Knee MRI scans were obtained during the CDAH-knee study (year: 2008-10, age: 30-40 years, n=313), and were assessed for cartilage volume, cartilage thickness, subchondral bone area, cartilage defects, and bone marrow lesions (BML). Univariable and multivariable (adjusted for age, sex, and BMI) linear regression and logistic regression were used to describe the association of biomarker at CDAH-1 and MRI-biomarkers at CDAH-knee; Tobit regression was used to describe the association of MRI-biomarkers at CDAH-knee and biomarkers at CDAH-3.

Results

In the multivariable model a significant negative association of COMP with medial femorotibial compartment cartilage thickness (-0.010 (-0.019, -0.000);p=0.045), and MMP-3 with patellar cartilage thickness (-9.075 (-16.344, -1.807);p=0.015) and total bone area (-0.047 (-0.086, -0.007);p=0.020) was found.

In the multivariable model a significant negative association of total cartilage volume (-0.0005 (-0.0008, -0.0002);p=0.001) and total cartilage thickness (-0.628 (-1.143, -0.114);p=0.017) with MMP-3, and total bone area with COMP (-0.270 (-0.474, -0.006);p=0.010) was observed.

Conclusion

COMP and MMP-3 levels were negatively associated with knee cartilage thickness assessed 4-years later. Knee cartilage thickness and volume were negatively associated with COMP and MMP-3 levels assessed 6-9 years later. These results suggest that OA-related biomarkers may predict future MRI-markers in middle-aged adults.

Collapse