Now that disease-modifying osteoarthritis drugs are under development, such as sprifermin and TissueGene-C, there’s a need to identify osteoarthritis (OA) at very early, potentially reversible stages. No current MRI technique can do that.
Mark Brezinski, MD, PhD, is a pioneer of optical coherence tomography (OCT), including polarization-sensitive OCT (PS-OCT), which assesses double reflection of light from body tissues. He has worked with Scott Martin, MD, director of the Joint Preservation Service in the Department of Orthopaedic Surgery at Massachusetts General Hospital and Brigham and Women’s, on OCT imaging osteoarthritis and musculoskeletal disease for over 25 years. Together with their colleagues, they previously determined PS-OCT can identify pre-OA. Healthy cartilage displays “banding” on PS-OCT, but when collagen becomes disorganized—one of the earliest changes in OA—banding is disrupted or lost.
Most recently, Dr. Brezinski, Dr. Martin, and colleagues showed PS-OCT can be performed minimally invasively. In Osteoarthritis and Cartilage Open, they report results from a pilot study funded by the arthritis division of NIAMS that will inform a large clinical trial to determine whether PS-OCT can predict the development of OA.
PS-OCT was conducted in 15 patients undergoing arthroscopic partial medial meniscectomy who had no evidence of OA on preoperative knee MRI. Medial meniscectomy was selected because a high percentage of that patient population progresses rapidly to OA.
No endocatheter is specifically designed for intra-articular imaging, so the researchers used a commercially available catheter for cardiovascular OCT. They used PS-OCT to completely image each condyle/plateau along six adjacent sagittal planes. The highest score of the six sagittal planes was taken as the grade for the entire medial tibial plateau.
PS-OCT images were successfully obtained in all cartilage sections of all 15 patients. However, several challenges arose with the cardiovascular endocatheter:
- Four catheters broke during the first three cases when the tip struck a solid surface
- Getting into each sagittal plane added about 15 to 20 minutes to the first five cases until the operator became skilled in handling the pliable catheter
- The catheters image radially to identify atherosclerosis in circular coronary arteries, and in the knee, only small areas of cartilage could be assessed at one time, and it wasn’t possible to image exclusively perpendicular to cartilage surfaces
At a mean of 2.65 years after arthroscopy, 11 of the 15 patients underwent repeat MRI. When the scoring method created for the study was applied, five patients were found to have had mild structural damage beyond pre-OA at baseline. They were excluded from the next analysis.
Ability of PS-OCT to Predict OA
The PS-OCT scores of the remaining six patients were categorized as being normal or abnormal, using 1.0 as the threshold:
- Three patients had normal scores (mean, 0.3)
- The other three had abnormal scores (mean, 4.3)
On repeat MRI, the mean MRI scores were:
- Normal group—0.3 (unchanged from baseline)
- Abnormal group—3.0 (severe OA)
The difference was significant (P<0.008), but the study was not powered to diagnose OA.
Information for Designing a Large Trial
The objective of a large clinical trial will be to determine whether abnormal PS-OCT scores predict progression to frank OA, as the Brigham team has noted in animals. This pilot confirmed that patients undergoing medial meniscectomy are suitable, but substantial endocatheter redesign will be needed. The team expects to have to build catheters specifically for the knee.
A minor point is that patient eligibility should be determined by grading preoperative MRIs with the method to be used in the study.