SUMMIT Prospective, Randomized, Controlled Trial: Response Rates To Matrix-induced Autologous Chondrocyte Implant (MACI) Versus Microfracture (MFX) By Lesion Characteristics

Abstract

Objectives:

To compare the response rates of patients treated with the MACI implant with those treated with MFX for the repair of symptomatic articular cartilage defects of the knee, based on lesion characteristics, in the phase 3 SUMMIT (Superiority of Matrix-induced autologous chondrocyte implant versus Microfracture for Treatment of symptomatic articular cartilage defects) trial (sponsored by Sanofi/Genzyme Biosurgery).

Methods::

Patients with ≥1 symptomatic focal articular cartilage defect of the femoral condyles (MFC/LFC) and/or trochlea and baseline Knee Injury and Osteoarthritis Outcome Score (KOOS) pain score ≤55 participated in this prospective, randomized, controlled trial (NCT00719576). The co-primary endpoint was improvement in KOOS pain and function (sports/recreational activities) at 2-year follow up. A patient was considered a responder if they had at least a 10-point improvement over baseline in KOOS pain and function scores. The Cochran-Mantel-Haenszel χ² (α=0.05) was used to analyze differences in response rates between groups by lesion size (>4 cm², >5 cm²), lesion location (MFC/LFC/trochlea), and osteochondritis dissecans (OCD) etiology (yes/no).

Results::

Patients (N=144) had a mean age of 33.8 years and 51% were male. The mean lesion size was 4.8 cm², and most lesions were on the MFC (74%). The KOOS pain and function (sports/recreational activities) co-primary endpoint improvement was clinically and statistically significantly better in patients treated with the MACI implant versus MFX (P=0.001) at 2-year follow up. Significantly better improvements were also observed for KOOS activities of daily living (P<0.001), quality of life (P=0.029) and symptoms (P<0.001) with the MACI implant compared with MFX. Response rates for the overall and subgroup analyses are shown in the Table. The incidence of adverse events was similar between the treatment groups and no unexpected safety findings were reported. Based on the small patient numbers in the subgroups, further research on response predictors is warranted.

Conclusion::

Clinically and statistically significantly better results in KOOS pain and function co-primary scores were observed with the MACI implant versus MFX for treating cartilage defects of the knee 2 years following surgery. More patients responded with the MACI implant compared with MFX when their lesions were >4 cm², on the MFC, and not of OCD origin.