CP #6: Novel imaging biomarkers for early osteoarthritis

Novel imaging biomarkers for early osteoarthritis


Osteoarthritis (OA) is a degenerative joint disease that affects more than 27 million people in the United States alone [1-3] and by 2030 approximately 70 million people will be affected. Early OA is manifested by biochemical changes in articular cartilage (e.g. proteoglycan loss), which eventually lead to irreversible morphologic changes and cartilage loss. While the morphologic changes are permanent, the biochemical degradation can be reversed. Conventional imaging methods, including radiography and magnetic resonance imaging (MRI), are limited to depicting late-stage OA, but early detection and intervention is imperative to prevent irreversible cartilage loss. There is therefore a significant need for reliable, non-invasive biomarkers that can detect the early biochemical changes in cartilage before irreversible morphological changes occur. The overarching goal of this project is to develop and characterize novel fluid suppressed-3D-23Na-UTE- techniques for in-vivo knee MRI on high- and ultra-high-field systems (3T and 7T). High-resolution 23Na (aggrecan) and 1H (morphology, collagen) imaging of cartilage, along with improved pulse sequences, image reconstructions, and visualization methods will significantly impact the objective assessment of OA pathology. This project will establish a powerful non-invasive imaging biomarker that is useful for clinical staging of disease severity, predicting risk for progression and eventually monitoring disease modifying therapies for OA.


In this study, we will implement and evaluate fluid suppressed-3D-23Na-UTE techniques for in-vivo knee imaging and quantify 23Na concentration in the knee joint of healthy controls and age matched early OA patients. We will then measure the 23Na concentration in early OA patients in a longitudinal study to compare with previously established metrics. In order to make this OA screen broadly available, we will also translate our methodology from 7T to a clinically relevant 3T platform and will correlate with T1rho.


The fluid-suppressed 3D 23Na-MRI method is able to reduce partial volume artifacts in the knee joint, ultimately enhancing the quantification of cartilage 23Na concentration. Figure 1 shows in vivo 7T 23Na MR images of the knee joint (using single-tuned 23Na and 1H knee coils) with visible suppression of both external saline (axial plane) and internal fluids (popliteal artery in sagittal plane) by inversion recovery techniques (NaIR). As we continue improving these techniques, we will then implement them in healthy controls (n=40) and OA patients of the same age range to establish the distribution of baseline 23Na concentration values in healthy volunteers as a function of age and correlate results with those of early OA patients. Once the baselines are established, we will be able to measure the 23Na concentration in early OA patients in a longitudinal fashion and correlate the results with previously established 1H-metrics and clinical scoring. Ultimately, we will translate this methodology from 7T field strength to 3T (see preliminary data in Figure 2) and will correlate with T1rho.

Push-Pull Interaction with TR&D 1 & 2:

TR&D#1:The sodium imaging methodology used thus far relies on the use of radial k-space trajectories. Working with BTRC staff including Drs Madelin and Otazo, we have already reported promising preliminary studies using compressed sensing reconstructions with these trajectories [4]. The use of combined radial compressed sensing and parallel imaging reconstructions developed as part of TR&D 1 will offer new avenues for acceleration, SNR-enhancement, and/or resolution enhancement in what is widely accepted as a challenging and SNR-constrained application. We will also explore 3D golden-angle radial acquisitions as potential motion-robust alternatives to our current radial trajectories.

TR&D#2: This project has already benefitted from improved dual-tuned (23Na /1H) RF coil arrays developed in collaboration with BTRC staff Drs Brown, Madelin, Lattanzi, Sodickson, and Wiggins [5]. We will explore alternative coil structures designed with the acquisitions and reconstructions from TR&D 1 in mind. We will also provide ongoing feedback to the BTRC regarding important practical issues affecting coil designs, such as low 23Na signal intensity in bone which has complicated the generation of coil sensitivity calibration maps.

Principal Investigator: 


Latest Updates

05/10/2019 - 21:50
05/10/2019 - 21:23

Philanthropic Support

We gratefully acknowledge generous support for radiology research at NYU Langone Health from:
• The Big George Foundation
• Raymond and Beverly Sackler
• Bernard and Irene Schwartz

Go to top