Abstract： Objective:To compare and observe the efficacy and safety of different administration methods of conbercept in the treatment of diabetic macular edema (DME).Methods:A retrospective clinical study. From November 2016 to November 2020, 135 eyes of 92 patients with foveal DME who were diagnosed in the Department of Ophthalmology of General Hospital of Central Theater Command received conbercept treatment were included in the study. All patients underwent best corrected visual acuity (BCVA) and optical coherence tomography (OCT) examinations. All the affected eyes received intravitreal injection of conbercept once a month for 3 months or 5 months, and then pro re nata (PRN) treatment based on the patient’s visual acuity and OCT examination results, namely 3+PRN or 5+PRN treatment plan, and divided into 3+PRN group (84 eyes) and 5+PRN group (51 eyes), respectively. There was no statistically significant difference in baseline information between the two groups of patients ( P>0.05). The changes of BCVA, centre retinal thickness (CRT), the number of intravitreal injections and the occurrence of complications in the two groups were compared and observed at the end of 12 months after treatment. The independent sample Mann-Whitney test was used to compare continuous variables between groups; the χ2 test was used to compare categorical variables. Results:At the end of 12 months after treatment, compared with baseline, the BCVA of 3+PRN group and 5+PRN group increased by 8.8±4.4, 9.2±6.1 letters, and CRT decreased by 145.1±50.5, 148.5±82.5 μm; there was no statistically significant difference between the two groups of eyes with BCVA increasing letter number and CRT decreasing value ( P=0.295, 0.548). In the 3+PRN group and 5+PRN group, the BCVA increased by more than 10 and 15 letters were 40 (47.6%, 40/84), 21 (25.0%, 21/84) eyes and 27 (52.9%, 27/51), 16 (31.4%, 16/51) eyes; there was no statistically significant difference in the proportion of the two groups with BCVA improvement> 10, 15 letters ( χ2=0.360, 0.648; P=0.549, 0.421). During PRN, in the eyes of 3+PRN group and 5+PRN group, the prognosis of eyesight was unstable in 22 (26.2%, 22/84) and 6 (11.8%, 6/51) eyes; the prognosis of eyes in the two groups was unstable. Compared with the number of eyes, the difference was statistically significant ( χ2=4.017, P=0.045). The number of injections into the vitreous cavity of the two groups of eyes were 4.1±2.9 and 2.4±1.8, the difference was statistically significant ( P<0.001); the times of remedial photocoagulation were 1.9±1.0, 1.5±0.8 times, the difference was statistically significant ( P=0.034). During the follow-up period, the overall incidence of ocular adverse events and serious adverse events in the 3+PRN group and 5+PRN group were similar, being 22.6% (19/84), 25.5% (13/51), and 8.3% (7/84), 7.8% (4/51), respectively. Conclusion:Both the 3+PRN and 5+PRN regimens of conbercept can treat DME safely and effectively; the 5+PRN regimen only requires fewer PRNs to maintain a more stable therapeutic effect.