Objectives: Anti-citrullinated peptide antibodies (ACPAs) are established as useful predictors of radiographic progression in rheumatoid arthritis (RA). The main objective of this study was to test the prognostic capacity of the recently developed test for anti-mutated citrullinated vimentin (anti-MCV). Methods: A cohort of 238 patients with RA was followed longitudinally for 10 years; 125 patients with complete x ray sets were included in the main analyses. Radiographs were scored according to the van der Heijde modified Sharp score (SHS). Patients were analysed for anti-MCV and anti-cyclic citrullinated peptide (CCP), and were genotyped for human leukocyte antigen (HLA)-DRB1 "shared epitope'' (SE) and protein tyrosine phosphatase, non-receptor type 22 (PTPN22) 1858T. Results: Anti-MCV and anti-CCP were strongly associated with regard to status and level. Both antibodies were associated with SE, but only anti-MCV was significantly associated with PTPN22 1858T. A positive anti-MCV test increased the odds of radiographic progression by 7.3 (95% confidence interval (CI) 3.2 to 16.5) compared to 5.7 (95% CI 2.6 to 12.5) for a positive anti-CCP. Presence of MCV antibodies gave an average increase in the total SHS of 30 U compared to an average increase of 25 U for the presence of CCP antibodies. Anti-MCVs were more strongly associated to progression in erosions than joint space narrowing. Associations remained after adjustment for other predictors of radiographic progression. The odds of progression increased with increasing anti-MCV level. Conclusions: Presence of anti-MCV predicted joint damage, and the strength of this prediction was at least as strong as for anti-CCP. Antibody status showed a stronger association to bone than to cartilage destruction. This study also indicates that higher anti-MCV levels add prognostic information compared to their mere presence or absence.