Peripheral blood DNA methylation models as predictors of knee osteoarthritis radiographic progression
Inventors
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Assignees
Member
Oklahoma Medical Research FoundationFounded in 1946, this independent nonprofit biomedical research institute conducts basic, translational, and clinical research in critical areas such as heart disease, cancer, autoimmune, and neurodegenerative diseases. Its mission focuses on understanding biological mechanisms and advancing diagnostics and therapeutics. Activities include conducting clinical trials, managing a patent portfolio, commercializing biotechnologies, and supporting the biotech community. Research efforts are funded by grants and philanthropy, and the institute hosts advanced facilities, interdisciplinary research teams, and collaborations with academia and industry.
Founded in 1946, this independent nonprofit biomedical research institute conducts basic, translational, and clinical research in critical areas such as heart disease, cancer, autoimmune, and neurodegenerative diseases. Its mission focuses on understanding biological mechanisms and advancing diagnostics and therapeutics. Activities include conducting clinical trials, managing a patent portfolio, commercializing biotechnologies, and supporting the biotech community. Research efforts are funded by grants and philanthropy, and the institute hosts advanced facilities, interdisciplinary research teams, and collaborations with academia and industry.
Publication Number
US-12467092-B2
Publication Date
2025-11-11
Expiration Date
Abstract
The present invention includes a method for determining whether a subject diagnosed with osteoarthritis is at increased risk for progression to severe osteoarthritis within 12-24 months, the method comprising the steps of: (a) preparing a blood sample; (b) measuring in the blood sample prepared in step (a) a level of DNA methylation; (c) comparing the DNA methylation level(s) measured in step (b) to a control blood sample level(s) of DNA methylation, and (d) identifying the subject as being at increased risk for progression to severe osteoarthritis within 12-24 months when the DNA methylation is increased by at least 2 fold as compared to the control blood level(s) of DNA methylation.
Core Innovation
The invention provides a predictive method that uses peripheral blood mononuclear cell (PBMC) DNA methylation to identify knee osteoarthritis (OA) patients at increased risk of radiographic progression within 12–24 months from a single baseline blood sample. DNA methylation in PBMCs is measured and compared to predetermined control methylation levels, with risk called based on a defined 13‑CpG panel. The approach addresses stratification of early knee OA patients.
Machine‑learning discriminant models based on M values outperform beta‑value models in a nested case‑control cohort from the Osteoarthritis Initiative. A supervised reduced model using a defined 13‑CpG panel achieves high predictive performance, reporting AUC‑ROC and accuracy, and validates across independent cohorts. Modeling and validation procedures reduce overfitting, and analyses include correction for cell‑composition effects.
The PBMC DNA methylation signature maps to genes and pathways including antigen presentation, AMPK signaling, and sonic hedgehog, and to upstream regulators such as PITX2, histone H3/H4, miR‑141/9/137, and BMP2. Proposed clinical uses include prognostic testing to stratify patients, guide treatment selection, and enrich disease‑modifying osteoarthritis drug trials. Preprocessing, array‑based methylation measurement, and statistical modeling methods are described [procedural detail omitted for safety].
Claims Coverage
Overview: One independent claim is covered. The main inventive features are five elements related to PBMC sample preparation, methylation measurement, comparison to predetermined control standards, detection of ≥2‑fold higher methylation at a defined 13‑CpG panel with identification of increased risk within 12–24 months, and administration of specified OA treatments.
Preparing a PBMC-containing blood sample
(a) preparing a subject blood sample, the sample comprising PBMCs from the subject;
Measuring DNA methylation levels in PBMCs
(b) measuring the levels of DNA methylation in the PBMCs of the sample prepared in step (a);
Comparing methylation levels to predetermined control standards
(c) comparing the DNA methylation levels measured in step (b) to control levels of DNA methylation in PBMCs from a control human subject or a group of control human subjects with knee OA that did not progress to severe OA, wherein the control levels are predetermined standards;
Detecting at least twofold higher methylation at a defined 13-CpG panel and identifying increased risk within 12–24 months
(d) detecting DNA methylation levels higher by at least 2 fold in the PBMCs in the subject sample compared to the control levels of DNA methylation at CpG markers comprising cg04195161, cg22064129, cg04985016, cg12692919, cg04043957, cg08872579, cg02019955, cg01333532, cg23705082, cg05042110, cg00715363, cg01307007, and cg09239099, and identifying the subject as being at increased risk for progression to severe knee OA within 12–24 months;
Administering a selected OA treatment to identified subjects
(e) administering to the human subject who is identified as being at increased risk for progression to severe OA within 12–24 months in step (d) at least one treatment of OA selected from: acetaminophen, steroids, hyaluronic acid, nonsteroidal anti-inflammatory drugs (NSAIDs), or Duloxetine.
The independent claim covers a PBMC‑based DNA methylation measurement and comparison workflow that detects at least 2‑fold higher methylation at a defined panel of 13 CpG markers to identify subjects at increased risk of progression to severe knee OA within 12–24 months, coupled with administration of specified OA treatments to identified subjects.
Stated Advantages
High predictive performance (radiographic AUC ≈0.94; accuracy 84–89% across comparisons).
Validation across independent cohorts.
Use of modeling and validation procedures to reduce overfitting.
Predictive identification from a single baseline blood sample.
M-value models outperform beta-value models (ROC-AUC 0.81 vs 0.75; accuracy ~73% vs 69%).
Enables prognostic testing to stratify early knee OA patients, guide treatment selection, and enrich disease-modifying osteoarthritis drug (DMOAD) trials.
Documented Applications
Predicting radiographic knee osteoarthritis progression and identifying subjects at increased risk within 12–24 months.
Prognostic testing to stratify early knee osteoarthritis patients (Kellgren-Lawrence score 1–3).
Guiding treatment selection with at least one treatment selected from acetaminophen, steroids, hyaluronic acid, nonsteroidal anti-inflammatory drugs (NSAIDs), or Duloxetine.
Enriching disease-modifying osteoarthritis drug (DMOAD) clinical trials.
Interested in licensing this patent?