In this retrospective study employing OCT and OCTA, we investigated the impact of DR on the longitudinal morphological and functional alterations in type 1 nAMD-associated MNV following the LP and 12-month follow-up of anti-VEGF treatment. Our findings indicate that eyes with mild NPDR exhibited distinct responses to treatment for nAMD. Specifically, the Diabetic group demonstrated a limited reduction in the MNV lesion area after 1 year of intravitreal therapy. This observation underscores the importance of considering such differential responses, particularly because OCTA-based metrics are currently utilized in the management of both nAMD and DR.
It has been suggested that alterations in the structure and function of the RPE and the choroidal circulation, which are associated with diabetes, may elevate the risk of developing neovascular nAMD.
18,23 Indeed, previous histological reports have demonstrated basement membrane thickening of the CC and other small choroidal blood vessels associated with luminal narrowing and capillary dropout in diabetic eyes with prolonged disease duration.
23 Following OCTA analysis, it has been hypothesized that eyes with NPDR may exhibit macular hypoperfusion and photoreceptor damage as a consequence of CC impairment.
18
However, there is limited evidence regarding the influence of DR on the management of patients with nAMD. Specifically, this subset of patients might exhibit divergent responses to anti-VEGF treatment over the long term, particularly concerning retinal anatomical changes. In our previous study examining the impact of DR on the response to the anti-VEGF loading phase treatment in patients with nAMD, we noted that the Diabetic group showed a smaller reduction in the size of the MNV compared to patients solely affected by nAMD.
10 Therefore, in this study, we chose to evaluate these preliminary results 1 year after the initiation of treatment by analyzing a similar group of patients. The results were encouraging, confirming the minor reduction in MNV after the LP in the Diabetic group. However, the most interesting finding pertained to the behavior of the neovascular lesion after 1 year of treatment initiation. Specifically, the Diabetic group did not exhibit a reduction in MNV after 12 months; instead, there was a lack of significant reduction of the area of the MNV. In contrast, the Not Diabetic group showed a continuous reduction in the size of the MNV. This last result appears to be highly significant, especially considering that the size of MNV evaluated with OCTA serves as a valuable biomarker in assessing the response or lack of response to treatment.
25,26
The lack of significant reduction of the neovascular membrane area in the Diabetic group after 1 year of treatment suggests a distinct treatment response compared to the continuous reduction observed in the Not Diabetic group. This finding underscores the importance of considering DR as a potential modifier of treatment outcomes in neovascular AMD management.
Multiple mechanisms could be contributing to this result, and they may even coexist simultaneously; for example, it is plausible that the presence of DR alters the angiogenic pathways involved in MNV. DR is associated with dysregulated VEGF signaling and increased levels of inflammatory cytokines, which could potentially lead to a different response to anti-VEGF therapy. These altered pathways may result in a reduced responsiveness to anti-VEGF therapy, leading to a less pronounced reduction of neovascularization over time. Another possibility is that the microvascular changes associated with DR, such as capillary dropout and basement membrane thickening, create a less conducive environment for the resolution of neovascular complexes. The compromised vascular structure and function in diabetic eyes may impede the ability of anti-VEGF therapy to induce significant regression of the MNV, resulting in a lesser degree of reduction compared to non-diabetic eyes.
This latter observation is further supported by another of our previous studies, which found that the Diabetic group exhibited greater ischemia of the CC surrounding the MNV even before treatment initiation compared to the Not Diabetic group. Additionally, following the anti-VEGF loading phase, the Diabetic group did not experience significant changes, unlike the Not Diabetic group, which demonstrated a gradual reperfusion of the CC around the neovascular membrane.
19
Additionally, we assessed changes in the MNV flow in both groups, revealing a decrease after the LP procedure and also after 1 year of anti-VEGF treatment. Our findings corroborate previous research, such as the study by Mastropasqua et al.
27 Importantly, we did not observe any significant differences in MNV flow between the two groups, although the Not Diabetic group exhibited a decreasing trend 12 months post-treatment. These results support the theory that anti-VEGF treatment leads to a reduction in the number and perfusion of smaller pathological vessels, but larger trunks remain adequately perfused, even in cases of nAMD complicated by initial stages of DR.
25
The differential response to anti-VEGF therapy observed between the Diabetic and Non-Diabetic groups might be related to variations in vitreous VEGF concentrations. Although we did not directly measure VEGF levels in our study, previous research has shown that diabetic patients often have higher baseline VEGF concentrations. For example, Dell'Omo et al.
28 reported significantly elevated vitreous VEGF levels in diabetic patients compared to non-diabetic controls, with levels further increased in those with proliferative DR. In the context of AMD, Tong et al.
29 found that VEGF levels were significantly higher in the vitreous of patients with active choroidal neovascularization compared to those without. These findings suggest that our Diabetic group might have had higher baseline VEGF levels, potentially requiring more aggressive or prolonged anti-VEGF therapy to achieve reductions in MNV size similar to those observed in the Not Diabetic group.
It is crucial to note that, although all previously significant parameters assessed with OCTA showed notable changes, there were no significant differences between the two groups in terms of functional (BCVA) and anatomical parameters measured with structural OCT, specifically CMT. Indeed, both groups exhibited a notable improvement in BCVA and a significant reduction in CMT compared to baseline following the LP and 1 year after treatment initiation. However, no significant differences were observed in terms of CMT and BCVA between the LP assessment and measurements taken 1 year into treatment. This suggests that, although there might be variances in the underlying anatomical changes detected with OCTA, these differences do not substantially affect visual function. This conclusion is supported by the similar improvements in visual acuity observed in both groups following treatment.
Our study has limitations that should be considered when interpreting our findings. First, our sample size was relatively small, potentially limiting the generalizability of our results. Additionally, the absence of a control group means that observed differences during the follow-up period could have occurred even without anti-angiogenic therapy, thus affecting the interpretation of treatment effects. Furthermore, the use of spectral-domain OCTA, which utilizes shorter wavelength light compared to swept-source OCTA, may result in reduced signal penetration through the RPE, potentially affecting the accuracy of our imaging data. Among the limitations of our study, we acknowledge that our analysis focused primarily on flow and density parameters of the MNV, in line with our previous research. Although this approach ensured consistency and comparability with our earlier findings, it may have limited our ability to provide a more comprehensive analysis of the vascular components affected by anti-VEGF therapy. A more detailed examination of additional OCTA parameters, such as the differentiation between trunk vessels and newly formed fine neovascularization or measurements of retinal vessel density in the macular area at baseline, could potentially offer deeper insights into the mechanisms underlying the differential responses observed between eyes with and without DR. Moreover, our study did not directly assess choroidal perfusion, which could potentially influence MNV changes, especially in diabetic eyes. Future studies incorporating choroidal perfusion measurements could provide valuable insights into the relationship between choroidal ischemia and MNV response to anti-VEGF therapy in eyes with DR. Finally, our analysis did not account for potential confounding factors related to systemic and ocular parameters, despite our efforts to ensure homogeneity in terms of age and axial length between the two patient samples. However, it is important to consider the strengths of our study, as well. Specifically, we exclusively enrolled patients with type 1 MNV who were treatment naïve, allowing for a focused investigation that leveraged the capabilities of OCTA, particularly in quantitative assessment. Additionally, to the best of our knowledge, our study is the first to analyze eyes with nAMD even in the presence of mild NPDR and to evaluate long-term morphovascular and morphofunctional changes following anti-VEGF treatment. Through OCTA imaging, we observed a distinct response to intravitreal therapy in terms of MNV lesion area in the Diabetic group, highlighting DM as a significant risk factor to consider in patients with nAMD who are undergoing anti-angiogenic therapy.
In conclusion, our study represents the first comprehensive investigation of the impact of DR on longitudinal morphological and functional changes in type 1 MNV associated with AMD in patients undergoing anti-VEGF therapy for 1 year. Utilizing OCTA, we unveiled a divergent response to intravitreal therapy in terms of MNV lesion area in the DR group. Specifically, the Diabetic group did not demonstrate a reduction in MNV area after 12 months; instead, the area of MNV remained largely unchanged. In contrast, the Not Diabetic group exhibited a continuous reduction in MNV size. The lack of significant reduction of the neovascular membrane area in the Diabetic group after 1 year of treatment indicates a distinct treatment response compared to the continuous reduction observed in the Not Diabetic group. This highlights the significance of considering DR as a potential modifier of treatment outcomes in nAMD management, with DM serving as a risk factor to consider during anti-angiogenic treatment. Future larger studies utilizing swept-source OCTA and longer follow-up durations are necessary to validate our preliminary findings.