Simultaneous inhibition of fibroblast growth factor-2 and vascular endothelial growth factor-a with RC28-E in diabetic macular edema: a phase 2 randomised trial

Objective To compare different doses and dosing regimens of RC28-E, a novel bispecific antibody that simultaneously binds vascular endothelial growth factor-A (VEGF-A) and fibroblast growth factor-2 (FGF-2), with conbercept in patients with diabetic macular edema (DME). Design Prospective, randomised, active comparator-controlled, open-label, multicentre, phase 2 clinical trial.cente Participants The trial enrolled patients aged 18 years or older with centre-involving DME, best-corrected visual acuity (BCVA) of 73 to 24 Early Treatment Diabetic Retinopathy Study (ETDRS) letters, and central subfield thickness (CST) of 300 mu m or more. Methods Patients were assigned randomly to one of five treatment regimens: 1.0 mg RC28-E for three initial monthly doses and then every 8 weeks (1.0mgQ8); 1.0 mg RC28-E for five initial monthly doses and then on a pro re nata (PRN) basis (1.0mgPRN); 2.0 mg RC28-E for three initial monthly doses and then every 8 weeks (2.0mgQ8); 2.0 mg RC28-E for five initial monthly doses and then on a PRN basis (2.0mgPRN); or 0.5 mg conbercept for three initial monthly doses and then on a PRN basis. Assessments were made at baseline and every 4 weeks thereafter. Main outcome measures The primary endpoint was the change in BCVA compared with baseline at 24 and 52 weeks. Secondary endpoints included the change in CST from baseline at 52 weeks; the proportion of patients who gained/lost >= 15 letters, >= 10 letters and >0 letter in BCVA; and the number of injections and safety outcomes. Results The trial enrolled 156 patients. Mean improvements in BCVA in the RC28-E groups at week 24 were 7.1, 11.0, 7.4 and 10.5 letters for 1.0mgQ8, 1.0mgPRN, 2.0mgQ8 and 2.0mgPRN regimens, respectively, versus 9.7 letters for the conbercept group (p=0.146). By week 52, the RC28-E groups exhibited respective mean BCVA enhancements of 5.5, 9.5, 9.2 and 9.7 letters, compared with 8.4 letters of the conbercept group (p=0.469). Mean reductions in CST in the RC28-E groups at week 52 were -163.2 mu m, -136.9 mu m, -142.5 mu m and -153.6 mu m, versus -160.7 mu m for the conbercept group (p=0.948). The Per Protocol Set analysis indicated that at 24 weeks, the BCVA improvement in the 2.0mgPRN group was significantly greater than that in the conbercept group (14.0 vs 9.8, p=0.019). In patients with poor baseline glycaemic control (HbA1c >= 7.5%), the 2.0mgPRN group showed greater BCVA improvement than the conbercept group (14.4 vs 4.2, p=0.039) at week 52. During the maintenance phase, the 2.0mgPRN group had fewer injections (2.8, 95% CI 1.8 to 3.7) compared with the conbercept group (4.4, 95% CI 3.5 to 5.2). RC28-E was generally well tolerated. The incidence of ocular adverse events in study eyes was comparable between RC28-E groups (22.6% in 1.0mgQ8 group, 26.7% in 1.0mgPRN group, 34.4% in 2.0mgQ8 group, 25.0% in 2.0 mg PRN group) and conbercept group (32.3%). The number of ocular serious adverse events was 1 (1.0mgQ8), 0 (1.0mgPRN), 1 (2.0mgQ8), 2 (2.0mgPRN) and 0 (conbercept). Conclusions Intravitreous RC28-E improved BCVA and CST in eyes with centre-involved DME. Compared with conbercept, the 2.0mgPRN regimen of RC28-E was recommended due to its superior efficacy in improving vision particularly for patients with poor glycaemic control, fewer treatment injections during the maintenance phase and comparable safety profile. Trial registration number NCT04782115.