A Study to Understand the Usage of an Artificial Intelligence Technology Called DARWEN for Eye Disorders and to Find Out the Factors Influencing Treatment Frequencies in People With nAMD and RVO (AIRSPEED)

Establishing the Use of DARWEN AI Technology in Ophthalmology and Identifying Predictors of Treatment Intervals in nAMD and RVO

This is an observational study in which data already collected from people with the eye disorders below are studied. In observational studies, only observations are made without participants receiving any advice or any changes to healthcare.

Neovascular age-related macular edema (nAMD): an eye disorder caused by the lack of oxygen in the retina. The lack of oxygen leads to the increase of a protein called vascular endothelial growth factor (VEGF). VEGF causes new, weak blood vessels to grow. These vessels can leak fluid or blood into the central part of the retina at the back of the eye (macula). This leads to blurring or a blind spot in the central (straight ahead) vision needed for reading or threading a needle.

Retinal vein occlusion (RVO): an eye disorder where a blood vessel that carries blood away from the retina (vein) becomes blocked. The blocked vein causes a lack of oxygen in the retina which leads to the increase of VEGF and then vision disturbances.

These eye disorders can be treated with a type of medicine called anti-vascular endothelial growth factor (anti-VEGF). Anti-VEGF treatment helps control the growth of new blood vessels in the eye and is given via injection into the eye.

DARWEN AI is a system using artificial intelligence (AI) technology. It has been proven that DARWEN AI can be used to extract data accurately and efficiently in multiple disease areas, including breast cancer, lung cancer, ambulatory care diseases and skin disorders.

The main purpose of this study is to validate if DARWEN AI can extract and sort through information from participants' health records accurately to identify those who need more frequent injections and those who do not.

The secondary purpose of this study is to use DARWEN AI to describe the factors influencing treatment frequencies in participants with nAMD or influencing treatment discontinuation in participants with RVO.

The data will come from the participants' information stored in EyeDoc Electronic Medical Records (EMR) at St. Joseph Health Care in London. Data collected are from January 1, 2009, to May 31, 2023.

Study Overview

• Status: Completed
• Conditions: Retinal Vein Occlusion; Neovascular Age-related Macular Degeneration
• Intervention / Treatment: Drug: anti-VEGF treatment

• Study Type: Observational
• Enrollment (Actual): 1350

Contacts and Locations

Study Locations

• Canada
  • Mississauga, Canada
    • Bayer

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult; Older Adult
• Accepts Healthy Volunteers: No

• Sampling Method: Non-Probability Sample

Study Population

All patients with the diagnosis of nAMD or RVO (Branch or Central) who are patients of Dr. Tom Sheidow listed in the EyeDoc Electronic Health Record (EHR) with available patient records from Jan 1, 2009 to May 31, 2023.

Description

Inclusion Criteria:

• All patients with the diagnosis of nAMD or RVO (Branch or Central) who are patients of Dr. Tom Sheidow listed in the EyeDoc EHR with available patient records from Jan 1, 2009 to May 31, 2023.

Exclusion Criteria:

• Not applicable.

Study Plan

How is the study designed?

Number of groups / cohorts

5

Cohorts and Interventions

Group / Cohort	Intervention / Treatment
nAMD patients, anti-VEGF naïve on treatment after July 2015; Cluster A: Likely to require frequent injections; Cluster B: Probable to extend treatment interval; Cluster C: Likely to extend treatment interval	Drug: anti-VEGF treatment; Follow clinical practice/administration
nAMD patients, anti-VEGF naïve on treatment before July 2015; Cluster D: Likely to require frequent injections; Cluster E: Probable to extend treatment interval; Cluster F: Likely to extend treatment interval	Drug: anti-VEGF treatment; Follow clinical practice/administration
nAMD patients, anti-VEGF experienced on treatment after July 2015; Cluster G: Likely to require frequent injections; Cluster H: Likely to extend treatment interval	Drug: anti-VEGF treatment; Follow clinical practice/administration
CRVO patients, anti-VEGF naïve on treatment after July 2015; Cluster A: Likely to require continued anti-VEGF treatment; Cluster B: Likely to be off anti-VEGF treatment	Drug: anti-VEGF treatment; Follow clinical practice/administration
BRVO patients, anti-VEGF naïve on treatment after March 2017; Cluster C: Likely to require continued anti-VEGF treatment; Cluster D: Likely to be off anti-VEGF treatment	Drug: anti-VEGF treatment; Follow clinical practice/administration

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Validation of DARWEN™ AI.	Based on defined objective clinical eligibility criteria, the EMR data set was divided into four cohorts using Pentavere's proprietary AI engine DARWEN™. The following ophthalmology-specific variables were validated against a manual review: diagnosis of nAMD, diagnosis of RVO including CRVO and BRVO, treatment intervals of anti-VEGF treatment grouped as < Q8 weeks or ≥ Q8 weeks for nAMD patients, discontinuation of anti-VEGF treatment for RVO patients, and the final outputted cohorts. The manual review was performed on a random selection of 25 patients from each of the four cohorts (100 patients total). Two trained manual abstractors classified these patients according to the rules outlined in the objective cohort criteria. They aligned on a consensus for each patient and compared with the DARWEN™cohorts. Pentavere then provided a success rate for the accuracy of the classification.	Retrospective data analysis from January 1, 2009 to May 31, 2023.

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Summary of baseline demographics reported as number of participants with different categories.	Age, sex.	Retrospective data analysis from January 1, 2009 to May 31, 2023.
Summary of clinical characteristics reported as number of participants with different categories.	Baseline ophthalmology characteristics, ocular comorbidities, other comorbidities, family history, etc.	Retrospective data analysis from January 1, 2009 to May 31, 2023.

Collaborators and Investigators

• Sponsor: Bayer

Publications and helpful links

Helpful Links

• Click here for access to information about Bayer's transparency standards and Bayer studies.

Study record dates

Study Major Dates

• Study Start (Actual): November 2, 2023
• Primary Completion (Actual): February 29, 2024
• Study Completion (Actual): February 29, 2024

Study Registration Dates

• First Submitted: July 3, 2024
• First Submitted That Met QC Criteria: July 9, 2024
• First Posted (Actual): July 10, 2024

Study Record Updates

• Last Update Posted (Actual): July 10, 2024
• Last Update Submitted That Met QC Criteria: July 9, 2024
• Last Verified: July 1, 2024

More Information

Terms related to this study

• Keywords: CRVO; BRVO; Branch retinal vein occlusion; RVO; nAMD; Central retinal vein occlusion
• Additional Relevant MeSH Terms: Cardiovascular Diseases; Vascular Diseases; Eye Diseases; Retinal Degeneration; Retinal Diseases; Embolism and Thrombosis; Venous Thrombosis; Thrombosis; Macular Degeneration; Retinal Vein Occlusion
• Other Study ID Numbers: 22642

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO
• IPD Plan Description: Availability of this study's data will later be determined according to Bayer's commitment to the EFPIA/PhRMA "Principles for responsible clinical trial data sharing". This pertains to scope, timepoint and process of data access. As such, Bayer commits to sharing upon request from qualified researchers patient-level clinical trial data, study-level clinical trial data, and protocols from clinical trials in patients for medicines and indications approved in the US and EU as necessary for conducting legitimate research. This applies to data on new medicines and indications that have been approved by the EU and US regulatory agencies on or after January 01, 2014. Interested researchers can use www.vivli.org to request access to anonymized patient-level data and supporting documents from clinical studies to conduct research. Information on the Bayer criteria for listing studies and other relevant information is provided in the member section of the portal.

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No