The objective of the proposed research is to better understand how peripheral blood epigenetic patterns are associated with knee osteoarthritis (OA). A great deal of work has already been demonstrated widespread epigenetic changes within articular tissues in both knee and hip OA. Others have described serum and urine protein biomarkers as predictors of future knee OA progression.
Our first Aim i s to evaluate peripheral blood cell DNA epigenetic patterns in baseline blood samples from patients who will go on to have rapid radiographc and/or pain progression in the subsequent 24 months. We will then use these data to develop develop and evaluate the performance of epigenetic algorithmic models to discriminate these groups. Patient samples will parallel the National Institutes of Health OA Biomarkers Consortium (OABC-FNIH) study. DNA methylation will be evaluated using a next-generation bisulfite sequencing approach (methylSeq), and algorithms developed using cutting-edge machine learning techniques. We will then translate our findings into a more high- throughput, inexpensive, and clinically relevant form by developing and validating a targeted capture sequencing system to interrogate these specific epigenetic locations.
Our second Aim i s to evaluate the peripheral blood DNA methylation patterns that precede the development of OA, using samples from 48-, 24-, 12-, and 0-months before incident OA. We will again develop algorithms to predict future OA development using similar techniques as Aim 1 and translate this to a targeted capture sequencing system. This unique longitudinal approach which will allow us not only to determine whether and when epigenetic patterns develop preceding OA development, but also track longitudinal epigenetic changes as OA develops. The proposed work is important, as there are no FDA approved biomarkers for OA diagnosis or prognosis. Our work is quite innovative both in its combination of big data epigenetic analysis and cutting-edge machine learning techniques applied to a specific clinical problem, as well as in its examination of PBMC epigenetics in OA, which has not yet been described. Moreover, we tackle the problem of translation of big-data research by aiming specifically to develop high-throughput methods to translate our findings into a clinically-relevant and accessible form. Success in our proposal will produce both algorithmic models with direct clinical impact to predict future OA development and progression, as well as broaden our understanding of epigenetic changes in peripheral blood cells from OA patients.
The proposed research is relevant to public health because a better understanding of blood epigenetic patterns in osteoarthritis (OA) is likely not only to give insights into the causes of OA, but also to give rise to new biomarkers to assist physicians and researchers diagnose OA patients and predict how rapidly they will worsen over time. Thus, this proposal is relevant to the part of the NIH's mission that pertains to fostering fundamental creative discoveries, innovative research strategies, and their applications as a basis for ultimately improving health.
