“Using these inclusion criteria would reduce the ability to generalize the results and would also make recruitment more difficult,” Dr. Weinreb added. “However, since change is more likely in these patients, the likelihood of detecting benefit with a neuroprotective agent would be improved.”
The sample size will depend on the rate of worsening of the outcome measures in untreated patients and the desired level of improvement to be provided by the treatment. Dr. Weinreb pointed out it might be possible to design trials enrolling just a few hundred patients treated for less than 1 year, and perhaps even just 6 months.
However, endpoint selection is critical.
“The detection of progressive glaucomatous injury and the definition of study endpoints have been problematic,” Dr. Weinreb said. “Regulatory agencies equate glaucoma progression with standard achromatic visual field loss and have not embraced structural alterations in optic nerve as primary endpoints.”
He acknowledged that preservation of visual function and vision-related quality of life is the ultimate goal in the clinical management of glaucoma patients. However, these are subjective endpoints.
Visual field thoughts
Speaking specifically about visual field loss, Dr. Weinreb said it is an insensitive measure early in the disease, variable (particularly late in disease), and not patient-friendly. He suggested that the ability to detect change with standard perimetry could be improved by increasing the frequency of testing, i.e., performing a cluster of tests at baseline and the end of the trial. Otherwise, demonstrating preservation of visual function would depend on the development of a more objective test.
Use of a structure-related endpoint is more attractive for assessing efficacy in glaucoma neuroprotection studies–since structural change is an objective measure and often occurs before functional change. Furthermore, with current spectral-domain optical coherence tomography (SD-OCT) systems, the imaging is patient-friendly.
“OCT imaging is done through an undilated pupil and data acquisition is very fast, which enables repeat testing,” Dr. Weinreb said. “The ability to easily collect a sufficient number of endpoints would allow a study that has a smaller sample size, shorter duration, and lower cost.”
Using an endpoint that combines structure and function testing is another approach that is promising. The combined structure-function index (CSFI), developed by Felipe A. Medeiros, MD, PhD, and colleagues at the Hamilton Glaucoma Center and Shiley Eye Institute, estimates retinal ganglion cell loss using data from SD-OCT imaging of the retinal nerve fiber layer, functional results from standard perimetry, and takes into account age-related change.