Abstract Background Cell therapies are being investigated as potential disease modifying treatment options for osteoarthritis (OA). Progenza (PRG) comprises in vitro expanded mesenchymal stem cells derived from human donor adipose tissue combined with cell culture supernatant. The primary objective of this first-in-human study was to evaluate the safety and tolerability of PRG. Methods We conducted a single centre, randomized, double-blind, placebo-controlled, single ascending dose study. Twenty patients aged 40–65 years with symptomatic Kellgren–Lawrence grade 1–3 knee OA were treated in two cohorts and randomized 4:1 to PRG or placebo. Cohort 1: 3.9 million cells (PRG 3.9M, n = 8) or placebo (n = 2) and cohort 2: 6.7 million cells (PRG 6.7M, n = 8) or placebo (n = 2). Each patient received a single intra-articular injection and was followed-up for 12 months. Results The study population comprised 20 patients (placebo, n = 4; PRG 3.9M, n = 8; PRG 6.7M, n = 8). All patients reported at least one treatment-emergent adverse event (TEAE). The majority of events [143/169 (84.6%)] were mild with 34 (20.1%) being considered by the investigator to be treatment related. There were no serious AEs or withdrawals due to AEs during the study. There was a statistically significant within group improvement in VAS pain scores from baseline at all timepoints for the PRG combined group, with highly significant improvements seen at months 3, 6, 9 and 12 (p ≤ 0.005) while VAS pain scores in the placebo group showed marginal improvement. A statistically significant improvement was also seen in WOMAC pain subscale scores from baseline at all timepoints for the PRG combined group while a marginal improvement in the placebo group was not statistically significant. Between screening and month 12, there was no decrease in average lateral tibial cartilage volume in the PRG 3.9M group while the placebo group showed a statistically significant cartilage loss. This difference between the placebo and PRG 3.9M group was statistically significant (LSM difference 106.47 mm3, 95% CI 13.56 mm3, 199.37 mm3, p = 0.028). Conclusion When administered as a single intra-articular injection to patients with symptomatic knee OA, PRG was safe and well tolerated. Furthermore, measurable improvements in symptoms and knee structure outcomes warrant further studies on PRG’s potential for disease modification in OA. Trial registration ANZCTR, ACTRN12615000439549. Date registered: 7th May 2015, https://www.anzctr.org.au/Trial/Registration/TrialReview.aspx?id=368355