Biomarkers of optical coherence tomography for choroidal neovascular membrane: correspondence

Dear Editor, We would like to discuss the publication in your esteemed Delta Journal of Ophthalmology: “Biomarkers of optical coherence tomography in evaluating the treatment outcomes of choroidal neovascular membrane [1].” The purpose of this study was to assess the efficacy of optical coherence tomography (OCT) imaging in monitoring the treatment of choroidal neovascularization, a topic of great therapeutic interest. However, on a thorough evaluation of the study methodology, we found some flaws that must be addressed in order to improve the study quality in the future. This investigation used a prospective interventional study design that was appropriate for tracking the treatment outcomes. However, it lacked a control group to compare the used treatment to other treatments or no treatment at all, making it impossible to draw causal conclusions about the changes in best-corrected visual acuity and central retinal thickness caused by ranibizumab injection alone. Furthermore, choosing patients with “clear retinal imaging” may result in selection bias, as patients with hazy vision or ocular comorbidities may be omitted. The stated means, SDs, and P values (<0.001) were statistically significant. However, no mention has been made of checking the distribution assumptions (normality test) or employing data-specific statistics, such as repeated measures analysis of variance or mixed modeling, which would reduce the data errors caused by repeated measurements in the same individual. Furthermore, the usage of “decimal best-corrected visual acuity” values may not be as widely recognized as the LogMAR conversion, which is more precise and accurate. In addition, multivariate regression data should contain information about the controlled factors and regression coefficients to further highlight the impact of OCT markers on the visual outcomes. In the discussion section, the researchers can elaborate by linking the study findings to pathophysiological mechanisms, such as the relationship between subretinal fluid depletion and photoreceptor integrity, or the effect of pigment epithelial detachment on anti-VEGF drug response. Furthermore, the discussion should address variations in outcomes compared to studies utilizing other medications, such as aflibercept or bevacizumab, and should take into account the frequency of injections at various phases to improve the comparative views and practical use in clinical practice. Future research directions should include randomized controlled trials comparing the OCT biomarkers in patients receiving different medications or treatment durations, as well as the use of artificial intelligence and machine learning technologies to thoroughly analyze the OCT images and create more accurate predictive models for treatment response in each report. Integrating biochemical data, such as blood VEGF levels, with retinal imaging data may aid in the development of customized medicine approaches for the choroidal neovascularization patients in the future, allowing for more targeted and effective outcome assessments and treatment planning. Financial support and sponsorship Nil. Conflicts of interest There are no conflicts of interest. Reply: We would like to thank Kleebayoon and Wiwanitkit for their letter containing comments about our published work (Biomarkers of optical coherence tomography in evaluating the treatment outcomes of choroidal neovascular membrane) [1]. We would like to comment at these points of criticism. 1. Our work was a single center study whose primary target was evaluating the effect of the detected OCT biomarkers on the treatment outcome after loading doses of ranibizumab only. We did not aim to compare ranibizumab to other anti-VEGF drugs because a larger number of patients would be required to conduct a reliable comparative study which was beyond the scope of our study. In addition, all our patients were looking for treatment, therefore we found it legally and ethically unacceptable to leave them without therapy just as a control group. 2. As inclusion criteria, our patients must have clear retinal imaging without other comorbidities to get reliable OCT data. Furthermore, patients with hazy vision were included as long as they have clear media. 3. To reduce the data errors that might be induced by repeated measurements of the same individual, we used Heidelberg Spectralis OCT machine which has an accurate tracking system that provides a reliable follow up of all patients. Also, the Pearson moment correlation equation was used to determine the correlation between the different variables. Additionally, the connection between a dependent variable and multiple independent factors was estimated using a multivariate regression analysis model to reliably detect the anatomical and visual outcome at the final follow up. 4. We preferred to use decimal BCVA as it showed easily non-confusing data in correlation with other parameters. 5. The poor BCVA at the final follow up was linked to PED, outer retinal tabulation (ORT), and hyperreflective foci and was explained by the associated sub-foveal photoreceptor damage. 6. We could not reliably compare our results to other studies using different anti-VEGF drugs due to the different inclusion criteria, regimens, and genetic characteristics. 7. We agree that future researches should include randomized controlled trials of larger number of cases, comparing the OCT biomarkers in patients receiving different medications with various treatment regimens and durations, as well as the use of AI and machine learning technologies to thoroughly analyze the OCT images and create more accurate predictive models for the treatment response in each report. Integrating the biochemical data with retinal imaging markers may aid in the development of customized medicine approaches for CNV patients in the future, allowing for more targeted and effective outcome assessments and treatment planning. References 1 Fadel AZ, Elkoumy HA, Aboelenine MA, Eldesoky EI. Biomarkers of optical coherence tomography in evaluating the treatment outcomes of choroidal neovascular membrane. Delta J Ophthalmol 2025; 26:184–191. Asmaa Z. Fadel, Hamdy A. Elkoumy, Mostafa A. Aboelenine, Elsaeed I. Eldesoky Department of Ophthalmology, Faculty of Medicine, Menoufia University, Gamal Abdel Naser Street, Shebinelkom, Menoufia 32951, Egypt. Tel: +20 100 870 4628. e-mail: [email protected]