The clinical course of systemic sclerosis‐associated interstitial lung disease (SSc‐ILD) is highly variable and easily measurable biomarkers are needed to predict disease progression. Serum epithelial biomarker KL‐6 is predictive of disease progression measured by a decline in DLCO, regardless of ILD severity, and could provide increased prognostic ability to inform risk stratification in SSc‐ILD. See related Editorial ABSTRACT Background and objective The course of systemic sclerosis‐associated interstitial lung disease (SSc‐ILD) is highly variable, and accurate prognostic markers are needed. KL‐6 is a mucin‐like glycoprotein (MUC1) expressed by type II pneumocytes, while CYFRA 21‐1 is expressed by alveolar and bronchiolar epithelial cells. Both are released into the blood from cell injury. Methods Serum KL‐6 and CYFRA 21‐1 levels were measured in a retrospective (n = 189) and a prospective (n = 118) cohort of SSc patients. Genotyping of MUC1 rs4072037 was performed. Linear mixed‐effect models were used to evaluate the relationship with change in lung function parameters over time, while association with survival was evaluated with Cox proportional hazard analysis. Results In both cohorts, KL‐6 and CYFRA 21‐1 were highest in patients with lung involvement, and in patients with extensive rather than limited ILD. KL‐6 was higher in patients carrying the MUC1 rs4072037 G allele in both cohorts. In patients with SSc‐ILD, serum KL‐6, but not CYFRA 21‐1, was significantly associated with DLCO decline in both cohorts (P = 0.001 and P = 0.004, respectively), and with FVC decline in the retrospective cohort (P = 0.005), but not the prospective cohort. When combining the cohorts and subgrouping by severity (median CPI = 45.97), KL‐6 remained predictive of decline in DLCO in both milder (P = 0.007) and more severe disease (P = 0.02) on multivariable analysis correcting for age, gender, ethnicity, smoking history and MUC1 allele carriage. Conclusion Our results suggest serum KL‐6 predicts decline in lung function in SSc, suggesting its clinical utility in risk stratification for progressive SSc‐ILD.