The role of radiography and MRI for eligibility assessment in DMOAD trials of knee OA

Abstract

Currently, no disease-modifying osteoarthritis drugs (DMOADs) have been approved. Past clinical trials have failed for several reasons, including the commonly applied definition of eligibility based on radiographic assessment of joint structure. In the context of precision medicine, finding the appropriate patient for a specific treatment approach will be of increasing relevance. Phenotypic stratification by use of imaging at the time of determining eligibility for clinical trials will be paramount and cannot be achieved using radiography alone. Furthermore, identification of joints at high risk of rapid progression of osteoarthritis is needed in order to enable a more efficient DMOAD trial design. In addition, joints at high risk of collapse need to be excluded at screening. The use of MRI might offer advantages over radiography in this context. Technological advances and simplified image assessment address many of the commonly perceived barriers to the application of MRI to assessment of eligibility for DMOAD clinical trials.

Figure 1

Radiographically ‘normal’ knees (i.e. Kellgren & Lawrence Grade 0) are structurally not ‘normal’. A. Anterior-posterior radiograph of the knee shows a radiographically normal knee without any signs of osteoarthritis (OA). The medial and lateral joint space is not narrowed and there are no osteophytes present. B. Corresponding sagittal intermediate-weighted fat-suppressed MRI acquired on the same day shows a large bone marrow lesion in the central subregion of the medial femur (thin white arrows), representing an OA feature that is associated with pain and structural progression. Note additional concomitant MRI features of OA including effusion-synovitis (asterisk) and superficial focal cartilage damage in the central part of the medial femur (thick white arrow).

Figure 2

Positioning. Serial radiographs of the same knee acquired consecutively. A. Anterior posterior radiograph of the knee acquired with a positioning frame with 11 degrees flexion shows a normal medial joint space width without signs of narrowing (arrows). B. Radiograph of the same knee acquired with 12 degrees flexion shows definite, flas positive, decrease in joint space width (arrows). C. Radiograph obtained with 13 degrees flexion shows severe, false positive joint space narrowing compared to the image obtained with 11 degrees flexion (arrows). Knee flexion has marked influence on joint space width and may lead to false positive or negative findings particularly in longitudinal studies.

Figure 3

Differential diagnosis of a painful knee with osteoarthritis. A. Anterior-posterior X-ray of a left knee shows marked joint space narrowing laterally (large arrow) and a definite osteophyte at the lateral tibial margin (small arrow) representing Kellgren & Lawrence grade 3 knee osteoarthritis. At the time of the image the patient was complaining about severe lateral joint pain without a history of trauma. B. Coronal T1-weighted MRI shows a stress fracture of the lateral tibial plateau (arrows). The knee also shows osteoarthritic features such as lateral tibiofemoral cartilage loss and moderate lateral meniscus extrusion.

Figure 4

Progression of meniscal extrusion and cartilage damage leading to increase in joint space narrowing on X-ray. A. Anterior-posterior radiograph of the knee shows no medial joint space narrowing (arrows) in a patient with previous anterior cruciate ligament repair and bilateral osteoarthritis as characterized by moderate marginal osteophytes at the medial and lateral joint margins. B. Follow-up X-ray 2 years later shows marked increase in joint space narrowing medially (arrows). C. Corresponding baseline MRI shows a normal medial meniscal body that is aligned with the medial margin of the tibia (arrowhead). There is discrete superficial tibial cartilage damage (arrows). D. MRI at 2 years shows marked extrusion of the meniscal body beyond the medial joint margin (arrowheads). In addition, there is progression of medial cartilage damage (arrows). Joint space narrowing on X-ray may be a result of cartilage damage, meniscal damage or meniscal extrusion, or a combination of these features.

Figure 5

Structural phenotypes as defined by MRI. A. Atrophic phenotype. Coronal intermediate-weighted fat-suppressed MRI shows marked bone marrow edema at the medial femur (small arrows). In addition there is meniscal extrusion (dashed arrow) and full-thickness cartilage loss at the medial femur (large arrow). No marginal osteophytes are seen at the medial or lateral joint line defining this knee as exhibiting an atrophic phenotype. B. Hypertrophic phenotype. Coronal dual echo at steady state (DESS) MRI shows large marginal ostephytes medially and laterally (arrows) characteristic of the hypertrophic phenotype of knee OA. In addition there are large osteophytes at the femoral notch (arrowheads). In addition there is superficial cartilage damage in the medial and lateral femur and tibia. C. Meniscal phenotye. Coronal intermediate-weighted MRI shows diffuse cartilage loss at the medial femur (arrow) and corresponding tibia. In addition there is marked menscal extrusion beyond the medial joint line (white bars) characteristic for the meniscal phenotype of knee OA. D. Coronal DESS MRI of another patient shows only a tiny residuum of the medial meniscal body (large arrow) following extensive partial meniscectomy. There is diffuse cartilage denudation at the medial tibia (small arrows). Furhermore there is a completely missing lateral meniscal body following meniscal resection also on the lateral side (dotted arrow). Note additional cartilage loss at the lateral compartment. Diffuse meniscal damage predisposes the joint for rapid cartilage loss and may be labeled as the meniscal phenotype. E. Inflammatory phenotype. Axial DESS MRI shows marked intraarticular joint effusion distending the joint capsule (asterisk). There is superficial cartilage damage at the medial patella facet. F. Sagittal intermediate-weighted fat-suppressed MRI of the same knee shows diffuse hyperintensity within Hoffa’s fat pad (grade 3 according the MOAKS scoring system), a commonly used imaging surrogate on non contrast-enhanced sequences for whole joint synovitis (arrrows). The combination of these MRI findings of joint effusion and Hoffa-synovitis is characteristic of the inflammatory phenotype on MRI.

Figure 6

Diagnoses of exclusion. A. Coronal intermediate-weighted fat-suppressed MRI shows a complete tear of the posterior meniscal root (arrow) leading to marked extrusion of the medial meniscal body beyond the joint line (arrowheads). Meniscal root tear represents a functional meniscectomy and will lead to rapid joint deterioration. B. Sagittal intermediate-weighted fat-suppressed MRI in another patient shows a subchondral insufficiency fracture of the medial femoral condyle reflected as a hypointense subchondral line on MRI (thin arrows). There is marked surrounding bone marrow edema (thick arrows). C. Sagittal MRI of the same patient as in B. obtained 12 months later show delamination of an osteochondral fragment (blue arrows) that appears to be instable as shown by fluid between the fragment and the subchondral bone (small white arrow). In addition there are areas of bone marrow edema at the medial femur and tibia (large white arrows). Severe progression of cartilage loss is observed in the weight-bearing aspect of the medial tibia. D. Another patient exhibits marked medial joint space narrowing (arrow) and large medial osteophytes defining this knee as Kellgren & Lawrence grade 3. The patient had severe pain at the time of image acquisition. E. The corresponding coronal intermediate-weighted fat-suppressed MRI shows a subchondral insufficiency fracture of the medial femur (grey arrow) and marked surrounding bone marrow edema (white arrows). F. The follow-up X-ray 6 months later shows no relevant increase in joint space narrowing (arrow) but an unclear radiolucency at the medial femur now also on X-ray suggestive of a subchondral insufficiency fracture (arrowheads). G. The corresponding MRI at 6 months follow-up acquired at the same time as the X-ray shows increasing deformity of the articular surface and diffuse concomitant bone marrow edema at the medial femur and tibia (arrows). Note that there is additional marrow reconversion with red bone marrow extending into the distal metaphyseal femur and tibia (asterisk).