MRI offers three fundamental advantages in osteoarthritis:

• Tomographic viewing perspective.
  • Facilitates dimensional measurements by eliminating magnification and morphological distortion caused by projectional radiography
  • Increases sensitivity for detecting bone erosions by eliminating superimposition of overlapping structures, which can obscure abnormalities on radiographs
• Unparalleled tissue contrast.
  • Allows direct visualization of marrow edema / inflammation, menisci synovial tissue and effusion,articular cartilage, tendons and ligaments, which cannot be seen with X-rays. This makes whole-organ assessment of the joint possible.
  • Allows assessment of compositional (collagen, proteoglycan, water) and physiological (vascular perfusion) features in addition to morphology
• Digital image format.
  • Compared to hard copy films, electronic images offer safer and more economical archival, rapid recovery and distribution to multiple readers, computer aided analysis and facilitated regulatory submission and auditing

These advantages can be leveraged in clinical trials of osteoarthritis to extend the scope of structural assessments and increase discriminative power so as to reduce the number of patients and study sites, and shorten measurement intervals required to demonstrate efficacy. This can reduce uncertainty and cost in drug development dramatically, and also potentially accelerate market entry and revenues realized over the patent life of a new drug. With over 9,000 MRI systems available worldwide (over 4,000 systems in the U.S. alone), MR image acquisition is no longer restricted to specialized research facilities, and can be supported in large multi-center and multi-national clinical trials. Recent technological innovations in this area promise to markedly decrease the cost and increase the versatility and convenience of this remarkable technology for osteoarthritis in the near future.

MRI is particularly useful for internal decision-making, as it can provide faster readouts in proof-of-concept, dose-selection and interval-selection studies than is possible with more conservative methods, which are required for pivotal therapeutic confirmatory studies seeking regulatory approval. Proper application of innovative methods in these early Phase II and III studies can shave several months off a drug development program.

Whole-Organ Evaluation in Osteoarthritis

The structural determinants of mechanical dysfunction and pain in arthritis are not well understood, but probably involve a multitude of interactive pathways. Accordingly, osteoarthritis is best modeled as a disease of organ failure. The current practice of monitoring only a few of these features (usually radiographic joint-space narrowing and osteophytes), provides only a keyhole view of the disease process and limits the utility of such assessments. A broader panel of imaging markers – i.e., a whole-organ evaluation -is needed to properly appraise structural integrity in osteoarthritis. MRI is uniquely suited to providing such a comprehensive assessment of joint status.

	Multiple disease pathways in arthritis call for a whole-organ view of the joint. MRI can evaluate all components of the joint simultaneously, allowing whole-organ assessment of osteoarthritic joints.

Dr. Peterfy, co-founder and Chief Medical Officer of Synarc, was the first to develop whole-organ MRI scoring (WORMS) for the knee. WORMS combines semi-quantitative assessments of a total of 11 structural features, including the articular cartilage, subarticular marrow-edema, cysts and bone attrition in 8 different locations in the knee; osteophytes along 16 articular margins; the medial and lateral menisci; the anterior and posterior cruciate ligaments; and the medial and lateral collateral ligaments. This whole-organ scoring method provides high inter-reader reproducibility and has been used to evaluate over 3000 knees in various clinical trials and epidemiological studies over the last several years.

SynaVu™ MRI Reading System

Centralized analysis of serially acquired MRI examinations requires the use of a specialized workstation in order to deal with the enormous number of individual images produced in each examination. Using the SynaVu™ MR reading system, individual sections from serial MRI scans acquired at multiple imaging sites are stripped of identifying information, anatomically registered, and stacked in separate windows on the workstation monitor. Each window represent a serial time point in a single patient series. Images in adjacent windows are anatomically aligned and viewed side-by-side to maximize our readers’ ability to detect small changes.

Serial MRI reading / quantification. Four windows containing stacked sagittal MRI images of the same knee from three time points in a study. The bottom right window contains an image from one of the three data sets in which the articular cartilage of the patella has been segmented for volume quantification.

The system allows the reader to scroll back and forth through the anatomy in all the windows simultaneously or individually in order to gain a 3D perspective of the joint destruction. The SynaVu™ reading system also provides numerous tools to aid in image analysis, including image windowing and level adjustment, zoom and panning functions, and dimensional measuring tools. Workstations are also provided with validated electronic score sheets with reader sign-off and automatic databasing of results.

The system is capable of more sophisticated image processing and analyses as well, including multi-modality fusion (e.g., MRI and scintigraphy) to expand the scope of imaging evaluations; spectral and temporal data fusion to facilitate longitudinal assessments, and various image segmentation and parametric mapping algorithms for quantitative analyses.

MRI of Articular Cartilage

MRI is unparalleled in its utility for delineating articular cartilage morphology and composition, particularly in large joints, such as the knee. In contrast to conventional radiography, MRI is able to evaluate all the articular surfaces in a joint not just those regions that are in direct contact, as is the case for radiographic joint-space width measurements. This allows MRI to disclose cartilage defects and thinning in regions of the joint not visible with radiography and gives MRI greater sensitivity to change. Using conventional MRI techniques available in most clinical centers around the world, high-quality images of the cartilage can be generated within only a few minutes in a highly standardized fashion and transferred to our facilities for centralized analysis. Morphological defects in different regions cartilage can be assessed by semi-quantitative scoring, and volume or thickness changes can be quantified using specially designed image processing and analysis software. Moreover, early degeneration of the cartilage matrix can be detected before morphological defects arise.

MRI shows articular cartilage defects directly. Conventional T2-weighted MR image of the knee that can be acquired in less than 5 min using the majority of clinical MR systems available today clearly shows a small focal partial-thickness defect (arrow) in the cartilage over the central medial tibial plateau. This represents a 2.0-Grade defect using the 14-point scale developed by Peterfy, et al.

Synarc's founding scientists have pioneered many of the MRI methods currently used to evaluate osteoarthritis in clinical trials. Charles Peterfy, Medical Director of Musculoskeletal at Synarc, developed a 14-region, 14-point MRI cartilage scoring method which has been used in more clinical trials and epidemiological studies than any other method. Dr. Peterfy was also the first to develop and validate a MRI method for quantifying the volume of articular cartilage in the knee and hands. Additionally, he developed the first whole-organ MRI scoring method (WORMS) for the knee. These methods have been used in more than 3,000 knee MRI examinations to date.

Longitudinal studies continue to provide information about the validity and performance of these methods, but considerable knowledge about how to implement these techniques in clinical trials already exists.

		Segmentation of articular cartilage. Using specialized image processing algorithms, individual structures, such as the articular cartilage, can be isolated from a MR image and quantified in a variety of ways, including total volume measurement (A) and thickness mapping (B). The precision error for measuring cartilage volume is 1% - 4% (coefficient of variation), while that for cartilage thickness mapping is 0.3 mm.

		Crescent™ phantom system for cartilage MRI. MR acquisition of articular cartilage in the knee and hip in longitudinal multi-center clinical trials is aided by the use of our specially designed Crescent™ phantom system. These phantoms provide a variety of internal and external standard references for documenting and correcting longitudinal fluctuations in MRI hardware performance that may affect cartilage volume, thickness and T2 measurements over time. Using these phantoms, spatial drift of <1% over time is attainable.

MRI markers of matrix damage predict cartilage loss. a. Conventional T2-weighted MRI images shows a focus of abnormal signal (arrow) in the femoral cartilage adjacent to the lateral meniscus. B. Fat-suppressed 3D gradient-echo images confirm the absence of any cartilage defects at this site. Follow-up MRI 1 year later (C, D) shows a focal cartilage defect (Grade 2.0) at this exact site.

MRI of Bone Changes

Osteophytes

Because of its tomographic viewing perspective, MRI can delineate osteophytes in locations that are invisible on x-ray radiographs of the knee as a result of projectional superimposition. This increases the sensitivity of MRI for this classic feature of osteoarthritis.

		MRI of osteophytes in osteoarthritis. MRI can delineate osteophytes that are not visible with conventional radiography.

Bone Edema

One of the most intriguing potential MRI endpoints in osteoarthritis is bone marrow edema-like change. Whether this feature represents pulsion of joint fluid through breaks in the articular surface, localized inflammation, or changes associated with trauma (microfracture or osteonecrosis) due to biomechanical incompetence of the articular surface is not known. However, since this feature can change very rapidly (less than 3 months) and correlates with pain and scintigraphic uptake of radiolabeled bone tracers, it may represent a very useful predictive imaging marker for patient selection and longitudinal assessment of efficacy and safety of putative therapies for osteoarthritis, particularly in studies aimed at internal decision making.

MRI of bone edema and cyst formation. MRI can also delineate subarticular cysts and areas of bone attrition better than radiography; although, these are relatively late findings in osteoarthritis. The three columns represent baseline, 6-month and 12-month follow-up images of a knee. The top row contains transverse T2-weighted images that show the gradual development of a subarticular cyst beneath abnormal articular cartilage over the posterior surface of the lateral femoral condyle. The middle row contains sagittal fat-suppressed, T1-weighted 3D gradient-echo images through the same femoral condyle. This MRI technique is insensitive to marrow edema, but delineates bone and cartilage very well. The images show the development of the bone cyst during primarily the second 6-month interval, and a focal defect in the overlying cartilage. The bottom row contains fat-suppressed T2-weighted images, which are highly sensitive to marrow edema. These images show extensive focal marrow edema at 6 months predicting the cyst formation evident at 12 months.

MRI of Synovitis

MRI is uniquely able to detect and quantify synovitis. Most of the interest in this imaging endpoint has been focused on rheumatoid arthritis. However, synovitis is frequently present in osteoarthritis and has the valuable attribute of rapid rate of change. Better understanding of how this feature predicts other structural changes in osteoarthritis and correlates with pain and other clinical outcomes would therefore be useful. Quantitative MRI markers of synovitis include the volume of synovial tissue and fluid and the rate of synovial enhancement following intravenous injection of contrast material.

MRI of synovitis. MRI is highly sensitive for fluid and synovial thickening in the synovial cavity. The volume of effusion ± hypertrophic synovial tissue can be measured in a manner similar to that of cartilage volume with high precision.

MRI of the Menisci and Ligaments

MRI has been the clinical imaging method of choice for evaluating the menisci, cruciate ligaments and collateral ligaments in the knee for almost two decades, and there is considerable knowledge about its diagnostic accuracy. Integrating these articular elements into the whole-organ assessment provides a richer picture of the structural integrity of the joint.

		MRI of the menisci and ligaments. a. Non-displaced tear (arrow) of the posterior horn of the medial meniscus. B. Lateral mensical tear associated with a mensical cyst (arrow). The medial and lateral collateral ligaments are intact. C. Torn anterior cruciate ligament (arrow).