It’s Time for a Change – Potential New Wound Healing Endpoints Highlight Need for Better Data Collection Tools
By Brian McManus and Travis Smith
Wound healing studies are on the brink of an exciting change. The FDA has long held “complete wound closure” as the single acceptable primary endpoint for efficacy in wound healing trials. But thanks to the efforts of the FDA Inter‐Center Wound Healing Working Group, a bevy of new clinical outcome assessments (COAs) have been proposed to the FDA for use as primary endpoints (reference). If the FDA expands the parameters for valid, reliable primary endpoints in wound healing studies, serious consideration must be given to the quality and accuracy of data-collection tools and methods. Though these COAs are frequently used for secondary endpoints, the scientific rigor of the data is sub optimal. Data collected manually (using calipers) or estimated (using linear, contact planimetry) are given the same consideration as measurements calculated using 3D digital planimetry systems. With the advancements being made in 3D digital planimetry, it’s time to reevaluate.
Where clinical research is concerned, it’s terrifying that there’s no standard protocol for measuring wound area. This makes it impossible to objectively distinguish ‘good’ data from ‘bad’ data. When it’s time for data analysis, data collected manually by measuring the length and width of a wound and multiplying them to get the area (of a rectangle) is given the same weight as data generated using 3D cameras designed to provide highly accurate wound measurements for area, length, width, depth and volume. It’s acceptable today for such an amalgamation of methods to co-exist because they don’t play a big hand in the binary open/closed assessment related to the current primary endpoint. But if new COAs are implemented as primary endpoints, like wound area reduction in 4-8 weeks, the quality and accuracy of the data connecting the dots between a wound at Baseline and the End of Treatment, particularly for wounds that don’t close, becomes vital.
Even if there was a standard protocol for collecting wound data, wound photography is difficult. Therefore, so is getting consistent, quality data. First, all the basic elements of taking a photo impact data quality in wound photography – distance between the camera lens and the wound, lighting, and focus. Also, in 2D digital planimetry, the wound must be captured from straight-on and there usually must be a reference marker present in the photo, and in the same plane as the wound, to calibrate the image or the data will be skewed or uninterpretable. Subjects with different skin tones and similar wounds photographed under the same lighting conditions can have significantly differing results. The photographer is also limited by what the camera can see so undermining and tunneling will never likely be accurately represented in a photo. For these reasons, and despite site staff training, data points are frequently lost during the conduct of the study through pure human nature.
The variability of data collected using differing methods is further compounded by the subjectivity of assessing and measuring a wound. Based on this, consider how many ‘opinions’ make it into the final wound measurement data for a study. Typically, there are 2-5 site staff at each study site trained and approved to collect wound data. Assume that different staff are photographing each subject wound from visit to visit and if wound tracings are applied by staff, the wounds are traced by different staff from visit to visit. Perhaps it’s these assumptions that lead to the FDA’s guidance Clinical Trial Imaging Endpoint Process Standards Guidance for Industry (reference) detailing how to optimize the quality of imaging data in clinical trials. But, while using a core lab can help mitigate data variability by having a limited number of readers measuring wounds and confirming wound closure, it doesn’t solve the issue of site staff collecting wound images in a multitude of ways. Systems that use artificial intelligence (AI) or proprietary algorithms to calculate wound measurements once the image has been collected will yield more consistent data than those where wound measurements are generated manually by users.
The future is in 3D wound photography systems. These systems capture wound images in a controlled manner by instantly providing bio-feedback to the user on camera positioning, lighting, focus and/or wound orientation. They instantly and automatically calculate wound measurements once the image has been collected. And even though volume isn’t on the list of new COAs, 3D cameras yield more consistent and accurate data for area, length and width. Subjects should no longer be enrolled in studies based on estimation of wound area while pending official measurement data from their vendor or core lab. A Principal Investigator’s designation of wound closure can be reviewed and confirmed in real time, allowing the subject to continue treatment unaffected if the wound closure designation is overturned by a core lab. It‘s time to decide whether the technical accuracy of wound measurement from a 3D device can be sacrificed for the lower cost of a 2D device.
Raising the bar for wound data collection methods and overall data quality is critical but it will not be easy. It’s time to eliminate the more archaic, manual measurement methods like calipers and acetate tracings. By making digital 3D planimetry the new standard, it may be possible to reduce the number of ways a user’s individual perspective can impact data collection. Using AI or other technology to increase the consistency of wound measurement will further abate variability. If the FDA Inter‐Center Wound Healing Working Group is successful in adding endpoints, the importance of accurate wound data collection should become pivotal.
Brian McManus has worked in clinical research for nearly 20 years, working with sponsors at a Clinical Research Organization on dozens of trials with a focus on collecting high quality endpoint data. As the former manager of a wound imaging core lab, he is well versed in the nuances of running wound healing trials. Working as the Director of Clinical Operations for eKare, Inc., he is invested in helping research clients collect wound healing data optimally.
Travis Smith started his career working in clinical research, with increasing managerial responsibility at companies like PPD and Chiltern. Travis later ran a global department at a CRO, specializing in projects requiring complex solutions for data collection. For the last two years prior to joining eKare, Travis was responsible for business development globally at a leading 3D wound imaging company.