Introduction

Diabetic retinopathy (DR) is the most common cause of vision loss among people with diabetes mellitus (DM) [1]. According to the data from a global pooled analysis, the overall prevalence was 34.6% for any DR, 6.96% for proliferative DR (PDR), and 6.81% for diabetic macular edema [2]. More seriously, 20.1% of type 1 DM cases and 25.4% of type 2 DM cases experienced vision loss within 10 years [3]. Vascular endothelial growth factor (VEGF) and placental growth factor (PlGF) are known to play important roles in the pathogenesis of PDR by inducing pre-retinal neovascularization and disrupting the blood–retinal barrier [4]. Anti-VEGF drugs, such as bevacizumab, ranibizumab, and aflibercept, were developed as new approaches to treat PDR [5]. However, they still have some unsatisfactory aspects. In November 2013, the China Food and Drug Administration (CFDA) approved conbercept (KH902; Chengdu Kanghong Biotech Co., Ltd., Sichuan, China), a new drug comprising VEGFR fusion proteins, which can specifically bind to VEGF-A, VEGF-B, and PlGF, for treatment of wet age-related macular degeneration [4]. In several previous studies, intravitreal conbercept (IVC) injection was proved to be an effective and safe treatment for patients with diabetic macular edema and retinopathy of prematurity [6]. On the other hand, the anti-VEGF drugs have been proved to be beneficial to vitrectomy procedures by reducing the chances of intraoperative bleeding owing to decreased concentrations of VEGF and PlGF [7]. Therefore, determining the concentration of VEGF and PlGF is important. Low concentrations of VEGF and PlGF would reduce the risk of intraoperative hemorrhage. In this study, we for the first time evaluated the concentrations of VEGF and PlGF in the aqueous and vitreous humor after patients received an IVC injection.

Methods

Study Subjects

Sixteen patients with PDR and nine patients with non-diabetic rhegmatogenous retinal detachment were recruited at our hospital. All patients underwent a complete ophthalmoscopic examination, including the best corrected visual acuity, slit lamp microscope, optical coherence tomography, fundus fluorescein angiography, B-scan ultrasound, intraocular pressure, and fundus photography. The patients who met the following criteria were included: age 18 years or older, diagnosis of type 2 diabetes, clinical diagnosis of PDR, and needed pars plana vitrectomy surgery. The patients were not eligible for the study if they had extremely high blood pressure, vitreoretinal disease other than DR, markedly reduced kidney and liver function, heart disease, senile macular degeneration, high myopia, retinal vasculitis, and history of retinal laser photocoagulation or intravitreal injection or vitrectomy. Informed consent was obtained from all participants. All procedures performed in this study were in accordance with the ethical standards of the Xi’an Jiaotong University committee and the Helsinki declaration of 1975, as revised in 2008.

Intravitreal Conbercept Injection and Sample Collection

Sixteen patients with PDR were randomly divided into two equal groups A (nine eyes) and B (nine eyes). The control group included nine non-diabetic rhegmatogenous retinal detachment patients (nine eyes). As shown in Fig. 1a, the patients in group A received 0.5 mg IVC injection after 100 μL aqueous humor was collected. After 7 days, the aqueous and vitreous humor were collected from groups A and B and the controls in the vitrectomy procedure. All patients completed follow-up at 1 week, 1 month, and 3 months.

Fig. 1
figure 1

Before and after the patients received the intravitreal conbercept treatment and underwent vitrectomy. a Schematic flow diagram of the process in this work. b Concentrations of VEGF and PlGF in the aqueous and vitreous humor pre- and post-intravitreal conbercept injection. c Correlation between the concentrations of VEGF and PlGF in the aqueous and vitreous humor. **p < 0.01

Measurement of VEGF and PlGF Concentrations

ELISA analysis was performed using Human VEGF ELISA kit and Human PlGF ELISA kit (Boster Biological Technology, Wuhan, China) according to the manufacturer’s instructions. Briefly, 20 μL of each sample was pipetted into a 96-well plate pre-coated with polyclonal anti-human VEGF or PlGF antibody. After incubation with a second antibody, the reaction was terminated by adding a stop solution. The absorbance of the resulting yellow product was measured at 450 nm on a microplate reader (BioTek Epoch). All ELISA experiments were performed in triplicate and were repeated three times.

Statistical Analysis

The data were expressed as mean ± standard deviation. Statistical comparisons of the three groups were performed using one-way analysis of variance (ANOVA). The Wilcoxon-signed rank test was used to compare the concentrations of PlGF and VEGF between groups. Spearman correlation analysis was used to analyze the correlation between VEGF and PlGF in aqueous and vitreous humor. All statistical analyses were performed using SPSS 18.0 for Windows (SPSS Inc., Chicago, IL, USA). For all tests, p < 0.05 was considered statistically significant.

Results

Patients’ Characteristics

In total, 27 eyes of 25 patients (11 women and 14 men) with a mean age of 51.69 ± 8.5 years (range 34–78) were recruited in the study. The mean duration of diabetes was 15.69 ± 8.73 years (range 4–31). There were no differences in sex or the mean age between these three groups (Table 1).

Table 1 Summary of patients (mean and standard deviation with statistical signification of the postvitrectomy vs. previtrectomy, *p < 0.05)

Concentrations of VEGF and PlGF in Aqueous Humor

As shown in Fig. 1b, the concentrations of VEGF (269.3 ± 118.1 pg/mL) and PlGF (355.5 ± 204.9 pg/mL) in group A (pre-IVC) were higher than those (VEGF, 128.7 ± 60.7 pg/mL; PlGF, 219.6 ± 151.5 pg/mL) in group A (post-IVC) (both Z = − 2.666, p = 0.008) and the control group (VEGF, 11.8 ± 5.5 pg/mL; PlGF, 12.4 ± 7.3 pg/mL) (both Z = − 3.578, p = 0.001). The concentrations of VEGF (301.0 ± 173.6 pg/mL) and PlGF (478.7 ± 273.9 pg/mL) in group B was also higher than those in the control group (both Z = − 3.578, p = 0.001) (Fig. 1b). However, there was no difference between group A (pre-IVC) and group B in the concentrations of VEGF (Z = − 0.221, p = 0.863) and PlGF (Z = − 0.927, p = 0.354).

Concentrations of VEGF and PlGF in Vitreous Humor

As shown in Fig. 1b, the concentrations of VEGF (911.8 ± 330.9 pg/mL) and PlGF (763.8 ± 373.5 pg/mL) in group B were higher than those (VEGF, 2.0 ± 10.1 pg/mL; PlGF, 40.7 ± 20.2 pg/mL) in the control group (VEGF, Z = − 3.578, p = 0.001; PlGF, Z = − 3.578, p = 0.001). Interestingly, the concentrations of VEGF (267.6 ± 76.1 pg/mL) in group A (post-IVC) were lower than those in group B (Z = − 3.576, p = 0.001), whereas no difference in the concentration of PlGF (PlGF in group A post-IVC, 721.0 ± 369.0 pg/mL) (Z = − 0.309, p = 0.796).

Correlation of Concentrations Between VEGF and PlGF

In aqueous humor, the concentrations of VEGF and PlGF have a positive correlation in group A (pre-IVC), group A (post-IVC), and group B (Fig. 1c), but not in the control group (r = 0.075, p = 0.847). In vitreous humor, there was a positive correlation between VEGF and PlGF concentrations in group B (Fig. 1c), but not in group A (post-IVC) (r = 0.567, p = 0.112) and the control group (r = 0.375, p = 0.345).

Visual Outcome

After vitrectomy, all patients had visual improvement at 1 week, 1 month, and 3 months follow-up. The mean logMAR BCVA was improved significantly at 3 months follow-up in all three groups (Table 1).

Discussion

In recent years, anti-VEGF agents considerably changed the treatment algorithms and improved prognosis of center-involving PDR [8]. As a new member of the anti-VEGF family of drugs, conbercept has been shown to be a safe and effective adjunct to vitrectomy in accelerating postoperative vitreous clear-up, and acquiring stable visual acuity restoration [9]. In comparison to ranibizumab, the first US FDA-approved medication used to treat wet age-related macular degeneration, Xu et al. reported that conbercept showed a longer treatment interval and fewer intravitreal drug injections were needed [6]. Su et al. reported that preoperative IVC injection could reduce the chances of intraoperative bleeding, which is beneficial to the management of PDR and the vitrectomy procedure [7]. Recently, Zhang et al. reported that the IVC combined with laser therapy could improve the vision of patients with Coats’ disease [10].

Conbercept consists of the VEGF-binding domains of human VEGFR-1 and VEGFR-2 combined with the Fc portion of human immunoglobulin G-1 [11]. In the rat retinal edema model, the concentrations of VEGF decreased after IVC injection [12]. However, there was no study, at least to our knowledge, to prove that the concentrations of VEGF and PlGF decreased after IVC injection in human aqueous and vitreous humor samples as a result of the difficulty in sampling the vitreous humor. In our study, we demonstrated that conbercept could significantly block the VEGF in the aqueous and vitreous humor of patients with PDR. Consistent with previous reports, the concentration of VEGF in vitreous was always higher than that in aqueous humor in diabetic patients [13]. Interestingly, we also found that the concentrations of PlGF decreased in aqueous humor 7 days after IVC injection, whereas there was no change in vitreous humor. The mechanism of this difference is unclear, but it may relate to some components in vitreous humor affecting the binding between PlGF and conbercept. Previous studies showed that conbercept could specifically bind to PlGF in vitreous humor and was critical to the development of PDR [14]. In PlGF knockout mouse strain, its genetic deletion protects the retina from diabetic damaged by inhibiting Akt activation and the HIF1α-VEGF pathway [15]. This all suggested that PlGF could be another therapeutic target of conbercept.

Conclusions

All our results, combined with those of previous reports, consistently showed that conbercept was effective in the treatment of PDR by blocking the VEGF and PlGF, whilst showing no signs of retinal toxicity.