Personalized Forecasting of Disease Trajectory for Patients with Open Angle Glaucoma

项目总结

Project Summary The primary objective of this project is to integrate principles and methods of operations research and systems engineering with data from NEI-funded clinical trials to develop an innovative approach to personalize the care of patients with open-angle glaucoma (OAG) and ocular hypertension (OHTN) to prevent avoidable blindness and vision loss. The outcomes sought will assist clinicians by (a) producing personalized forecasts of the probability of progressing from OHTN to OAG and less severe to more advanced disease states, (b) determining the optimal timing of specific diagnostic tests to monitor for glaucomatous progression for each patient, (c) identifying those at highest risk for irreversible vision loss from OAG (i.e., “fast progressors”), and (d) generating recommended target treatment goals of intraocular pressure (IOP). To achieve these objectives, this project integrates an understanding of glaucoma progression trajectory from NEI-funded clinical trials (including OHTS, CIGTS, and AGIS) with an individual patient's past and current test results from perimetry, tonometry and optical coherence tomography to generate personalized forecasts of glaucoma progression dynamics. Prior work by our group has shown that a preliminary forecasting tool we developed could accurately identify instances of OAG progression 57% sooner (p=0.02) and 29% more efficiently (p<0.0001), compared with the current practice for many patients of fixed 1-year intervals for patient assessment and testing. In this proposal, we look to greatly enhance the forecasting tool in several ways. In Aim 1 we will develop, parameterize, calibrate, and validate an advanced tool using data from the OHTS trial, to forecast if a patient with OHTN will develop OAG and the timing of progression to OAG. While our sophisticated state space Kalman filtering methodology appears to perform very well on patients with moderate to severe OAG, in this aim we plan to apply this methodology to study disease progression dynamics for patients with OHTN using data from OHTS. In Aim 2 we plan to extend the inputs to the forecasting tool beyond data from tonometry and perimetry to now also include data from structural testing including optical coherence tomography. In addition, this aim proposes to expand the output of the tool to include personalized predictions of which patients will become fast progressors, allowing clinicians to intervene before vision is irreversibly lost. In this aim, we also plan to forecast and graphically display the patient's likely OAG progression trajectory given a menu of different possible levels of IOP control, aiding the eye-care provider and patient in choosing how aggressive the treatment should be. By fulfilling the aims of this proposal, we hope to develop an advanced forecasting tool that will provide clinicians and patients with personalized, dynamically-updated, real time forecasts of OAG progression dynamics for each eye, which will greatly aid with decreasing avoidable vision loss and blindness from OAG.