[Pro re nata anti-VEGF treatment results for neovascular age-related macular degeneration in routine clinical treatment: comparison of single with triple injections]

Visual acuity prediction on real-life patient data using a machine learning based multistage system

In ophthalmology, intravitreal operative medication therapy (IVOM) is a widespread treatment for diseases related to the age-related macular degeneration (AMD), the diabetic macular edema, as well as the retinal vein occlusion. However, in real-world settings, patients often suffer from loss of vision on time scales of years despite therapy, whereas the prediction of the visual acuity (VA) and the earliest possible detection of deterioration under real-life conditions is challenging due to heterogeneous and incomplete data. In this contribution, we present a workflow for the development of a research-compatible data corpus fusing different IT systems of the department of ophthalmology of a German maximum care hospital. The extensive data corpus allows predictive statements of the expected progression of a patient and his or her VA in each of the three diseases. For the disease AMD, we found out a significant deterioration of the visual acuity over time. Within our proposed multistage system, we subsequently classify the VA progression into the three groups of therapy "winners", "stabilizers", and "losers" (WSL classification scheme). Our OCT biomarker classification using an ensemble of deep neural networks results in a classification accuracy (F1-score) of over 98%, enabling us to complete incomplete OCT documentations while allowing us to exploit them for a more precise VA modelling process. Our VA prediction requires at least four VA examinations and optionally OCT biomarkers from the same time period to predict the VA progression within a forecasted time frame, whereas our prediction is currently restricted to IVOM/no therapy. We achieve a final prediction accuracy of 69% in macro average F1-score, while being in the same range as the ophthalmologists with 57.8 and 50 ± 10.7 % F1-score.

[Influence of CNV vascular morphology in exudative age-related macular degeneration on development of visual acuity and need for anti-VEGF therapy after 1 year]

BACKGROUND

The aim of this pilot study was to investigate vascular morphological characteristics of choroidal neovascularization (CNV) at the time of the initial diagnosis of exudative age-related macular degeneration (nAMD), which enable a prognosis for the development of visual acuity and the necessity for treatment in the first year.

METHODS

In 57 patients with the initial diagnosis of nAMD, CNV was detected by optical coherence tomography angiography (OCT-A) and an automated quantitative vessel analysis was performed with respect to area, total vessel length, flow value and average vessel caliber of the CNV. After 12 months patients were divided into 2 groups based on visual acuity (visual loss vs. visual gain) and necessity of anti-VEGF therapy (<7 intravitreal injections, IVOM vs. ≥7 IVOM).

RESULTS

The mean CNV area was 0.95 mm² ± 1.07 mm² (group visual loss 1.56 mm² ± 1.54 mm²; group visual gain 0.65 mm² ± 0.53 mm²; p = 0.002/<7 IVOM 1.05 mm² ± 1.40 mm²; ≥7 IVOM 0.98 mm² ± 0.94 mm², p = 0.60). The average total vessel length of the CNV was 9.84 mm ± 11.35 mm (visual loss 16.00 mm ± 16.58 mm; visual gain 6.74 mm ± 5.42 mm; p < 0.003/<7 IVOM 11.21 mm ± 15.10; ≥7 IVOM 9.90 mm ± 9.68 mm; p = 0.68). The mean flow value of the CNV was 0.40 ± 0.06 (visual loss 0.37 ± 0.04; visual gain 0.41 ± 0.07; p = 0.004/<7 IVOM 0.42 ± 0.08; ≥7 IVOM 0.38 ± 0.06; p = 0.02). The average vessel caliber was 28.86 µm ± 2.93 µm (visual loss 28.39 µm ± 2.97 mm; visual gain 29.32 µm ± 3.05 µm; p = 0.24/<7 IVOM 30.26 µm ± 3.49 µm; ≥7 IVOM 28.23 µm ± 2.25 µm; p = 0.02).

CONCLUSION

The results show that a mathematical quantification of the CNV in nAMD is possible using OCT‑A. This analysis confirmed again that the size of the CNV (area and total vessel length) is decisive for the prognosis of visual acuity. It also shows that a larger flow value as a sign of a well-differentiated CNV is associated with a better functional prognosis. The number of IVOMs required, however, depends primarily on the composition of the CNV (flow value and vascular caliber). More precise imaging and larger examination cohorts could possibly reveal further relevant biomarkers.

[IVOM quality assurance in Westfalen-Lippe : Structure of quality assurance and results of the pilot study Q-VERA]

BACKGROUND

The KVWL-QS assists ophthalmologists in the transfer for good clinical praxis into real life. In addition the QS-commission initiated a pilotstudy "Qualitäts-Versorgung bei AMD" (Q‑VERA) in order to test new instruments for improvements. It was analized, if Reading Center (RC) based controls in combination with specific case-management modules can improve the results of IVOM treatment.

PATIENTS AND METHODS

In 5 treatment centers 878 consecutive patients with newly diagnosed AMD (Neu-Patienten) were included, who were treated with the IVAN-scheme. Initial FA and OCT images were transferred electronically to the RC. Also 781 retreatment patients (mean 20.7 IVOM before) with retreatment due to lesion activity were observed.

RESULTS

In 5% of the 878 newly treated patients a discrepancy between RC and treatment center was recorded. In this group the 481 patients, who finished up to the analysis date the 12-month follow-up, the visual function (increase in BCVA) and SD-OCT (reduction in central retinal thickness) results were comparable with large prospective cohorts. This was achieved with 6.5 injections and 10.6 visits over 12 months. In the group of 781 patients with repeated injections the number of injections over 12 months was 7.7 and the number of visits 11.6.

CONCLUSION

Quality assessment can improve the efficacy of IVOM therapy for AMD patients in real life. In addition to existing structures, electronical exchange by a RC assisted evaluation can further improve the quality by reducing the number of unnecessary treatment visits. The case-management with adherence control, re-call-system and specific information for patients and relatives can specifically increase the long-term adherence and thus the success of the therapy.

Changes in the OCT angiographic appearance of type 1 and type 2 CNV in exudative AMD during anti-VEGF treatment

Objective

Optical coherence tomography angiography (OCT-A) enables detailed visualisation of the vascular structure of choroidal neovascularisation (CNV). The aim of this study was to determine whether mathematically ascertained OCT-A vascular parameters of type 1 and type 2 CNV in exudative age-related macular degeneration (AMD) change during antivascular endothelial growth factor (anti-VEGF) treatment. The OCT-A vascular parameters were also compared with previously obtained activity parameters (fluid distribution on spectral domain OCT (SD-OCT)) to establish whether they could potentially be used as further ‘activity parameters’ for assessment of anti-VEGF treatment.

Methods and Analysis

We evaluated 27 eyes of 27 patients (mean follow-up 9.8 months) with type 1, type 2 or mixed CNV who had received anti-VEGF treatment (IVAN scheme). The parameters analysed were area (aCNV), total length of all vessels (tlCNV), overall number of vascular segments (nsCNV) and fractal dimension (FD) of the CNV. The changes in each of these parameters were correlated with the central foveal thickness (CFT).

Results

Regression and renewed perfusion of the CNV corresponded with the decrease or increase, respectively, of macular fluid distribution on SD-OCT. The increase and decrease of CFT during anti-VEGF treatment were highly significantly correlated with changes in FD (p<0.00001), aCNV (p<0.00001), tlCNV (p<0.00001) and nsCNV (p<0.00001).

Conclusion

OCT-A enables detailed analysis of AMD with regard to FD, aCNV, tlCNV and nsCNV. As the changes in these parameters correlate closely with changes on SD-OCT, they can be used as new activity parameters, alongside fluid distribution, for assessment of treatment effect and as parameters of stabilisation or the need for repeated treatment.