Purpose: Age-related changes in articular cartilage are likely to play a role in the aetiology of osteoarthritis (OA). One of the major age-related changes in cartilage is the accumulation of advanced-glycation-endproducts (AGEs). The present study evaluates whether pentosidine can predict radiographic progression and/or burden over 5 years follow-up in a cohort of early knee and/or hip OA. Methods: The Cohort Hip and Cohort Knee (CHECK) is a cohort of participants with very early OA. In this study data of 5 years follow-up were used. Radiographic progression and burden were assessed by X-rays of both knees and hips (K&L and Altman scores). Radiographic OA burden was expressed as the summed K&L grade or Altman scores of both hips and knees. Radiographic OA progression was expressed as the area under the curve (AUC) of the summed K&L grade over the 5 years minus the baseline value over 5 years. The burden of radiographic joint damage was expressed as the AUC over 5 years. For the summed Altman scores for JSN and osteophyte formation the same approaches were used to determine progression and burden scores per patient. Baseline skin pentosidine levels (and urinary CTXII as a comparator) were measured by HPLC (and ELISA). Univariate and multivariate associations including baseline radiographic damage, age, gender, BMI and kidney function were performed. Results: Of 183 participants all data were available. At 5-year follow-up radiographic progression of OA was seen (sum K&L score 1.69 ± 1.13 at T0 versus 3.30 ± 1.81 at T5). Both pentosidine and uCTXII correlated with this radiographic progression and burden (ie progression AUC sum K&L score R=0.167 p=0.024 and R=0.323 and p=0.000 for skin pentosidine and urine CTXII, respectively). In multivariate analysis, in general pentosidine did not have an added predictive value to uCTXII for progression nor burden of the disease. The best prediction was obtained for burden of radiographic damage (R2=0.60-0.88), but this was predominantly determined by baseline radiographic damage (without this parameter R2=0.07-0.17). Interestingly, pentosidine significantly added to prediction of osteophyte formation, whereas uCTXII significantly added to prediction of JSN in multivariate analysis. Conclusion: Pentosidine adds to prediction of radiographic progression and burden of osteophyte formation and uCTXII to radiographic progression and burden of JSN, but overall skin pentosidine did not perform better than uCTXII in predicting radiographic progression or burden. Burden of damage over 5 year is mainly determined by radiographic joint damage at baseline. This study was supported by the Dutch Arthritis Association.