Faculty Bibliography

Background/Purpose : Osteoarthritis (OA) is a highly heterogeneous disease, which suggest that multiple endotypes exist. Identification and characterization of such endotypes may assist in precision medicine for identification of faster progressors whom may benefit from a given type of intervention. Recent published data have shown that SNPs in growth factors such as TGFbeta and GDF are associated with OA, which indicate that cartilage formation and repair play an important role in progression of OA. The aim was to determine whether a biomarker of type II collagen formation measured in serum, as a potential surrogate measure of cartilage formation, could predict radiographic progression in knee OA population. Subsequently, we investigated if such a proposed low cartilage formation/repair endotype was more responsive to a potential treatment of OA. Methods : hsPRO-C2, a measurement of the type II collagen pro-peptide, was measured in blood samples of two independent knee OA cohorts: 106 recruited at New York University (NYU cohort) and 147 from the phase III OA trial SMC021-2301 (clinicaltrial.gov: NCT00486434) evaluating the efficacy and safety of oral salmon calcitonin. Patients were dichotomized based on their baseline level of hsPRO-C2 and the mean difference in two-year radiographic progression (joint space narrowing (JSN)) was analyzed using ANCOVA adjusting for baseline demographics and clinical characteristics. Results : In the NYU cohort, baseline plasma hsPRO-C2 levels were negatively correlated with the progression of radiographic JSN (r = -0.26, p = 0.009). Quartile analysis demonstrated a significant difference in mean JSN from quartile 1 to 4 (0.51 mm versus -0.07 mm, p = 0.036, fig. 1). Knee OA patients with low hsPRO-C2 levels (<= 1.48 ng/mL) revealed significantly larger JSN compared to the individuals with high hsPRO-C2 levels ( > 1.48 ng/ mL) at 24 months (0.37 mm vs 0.02 mm, p = 0.042). In the SMC cohort, there was no significant treatment effect on the medial JSN over 2 years before stratification by hsPRO-C2; however, as observed in the NYU cohort, JSN was on average higher in the low hsPRO-C2 (<= 1.96 ng/mL) group compared to the high group ( > 1.96 ng/ mL). Furthermore, in the low baseline hsPRO-C2 subgroup, sCT-treated patients on average had a lower JSN compared to placebo patients (p < 0.05, fig. 2). The opposite trend was observed in patients with high baseline hsPRO-C2. Conclusion : Here we show that low levels of cartilage formation, measured by PRO-C2, were associated with radiographic progression and greater likelihood of response to a salmon calcitonin. Low PRO-C2 may provide a measure of an OA endotype with low background cartilage formation (at baseline) and higher capacity for repair when treated with a potential cartilage anabolic drug

Background: There is a lack of objective diagnostic modalities that identify patients at risk for severe osteoarthritis (OA), which complicates the development of disease-modifying OA drugs. The biochemical marker, high-sensitive PRO-C2 (hsPRO-C2)¹, is a measure of the propeptide of type IIB collagen and a blood measure of cartilage formation.
Objective(s): The aim of this study was to determine whether hsPRO-C2 could predict radiographic progression in a knee OA population and stratify patients into high and low risk for joint destruction.
Method(s): Subjects with varying degrees of symptomatic knee OA (n=106) were included from a New York University (NYU) progression cohort. Radiographic progression was assessed by medial joint space narrowing (JSN), based on the change in joint space width (JSW), of the signal knee at baseline and at 24months. Baseline plasma type II collagen formation biomarker (hsPRO-C2) levels were measured. Association between baseline hsPRO-C2 and JSN was analyzed by Pearson's correlation, corrected for age, sex, BMI, race, baseline JSW, and non-steroids antiinflammatory drugs (NSAID) use. Subjects were divided into quartiles of equal size depending on the hsPRO-C2 levels, and the difference in JSN was investigated. The median level of baseline hsPRO-C2 (1.48 ng/ml) was used as a cut-off for stratifying all the subjects. The difference in JSN over 24 months was investigated in patients dichotomized based on median level. The values were compared with two-way analysis of covariates (ANCOVA).
Result(s): Baseline plasma hsPRO-C2 levels were negatively correlated with the progression of radiographic joint space narrowing over 24 months (r =-0.26, p = 0.009) after adjustment for confounders (Figure 1A). Quartile analysis demonstrated a decreasing trend of hsPRO-C2 in the radiographic progression from quartile 1 to 4 (Figure 1B). One-way ANOVA revealed a significant difference in mean JSN between quartiles 1 and 4 (0.5073 mm versus-0.0691 mm, p = 0.036, Figure 1B). JSN was significantly larger in the low hsPRO-C2 patients (0.3710 mm) com pared to the high hsPRO-C2 patients (0.0195 mm) (Figure 2).
Conclusion(s): These data suggest that symptomatic knee OA subjects with lower levels of hsPRO-C2 at baseline presented more radiographic medial JSN progression as compared to the subjects with higher levels of hsPRO-C2. The biomarker hsPRO-C2 may be useful for predicting OA progression

BACKGROUND:To investigate the expression of vascular adhesion protein-1 (VAP-1) in joint tissues and serum in symptomatic knee osteoarthritis (SKOA) patients and examine whether VAP-1 levels predict increased risk of disease severity in a cross-sectional study. METHODS:Baseline VAP-1 expression and soluble VAP-1 (sVAP-1) levels were assessed in the synovium synovial fluid and in the serum in cohorts of patients with tibiofemoral medial knee OA and healthy subjects. Standardized fixed-flexion poster anterior knee radiographs scored for Kellgren-Lawrence (KL) grade (0-4) and medial joint space width (JSW). KL1/2 vs. KL3/4 scores defined early and advanced radiographic severity, respectively. Biochemical markers assessed in serum or synovial fluids (SF) comprised sVAP-1, interleukin 1 receptor antagonist (IL-1Ra), interleukin 6 (IL-6), soluble receptor for advanced glycation end-products (sRAGE), C-C motif chemokine ligand 2 (CCL2), C-C motif chemokine ligand 4 (CCL4), cluster of differentiation 163 (CD163), high sensitivity C-reactive protein (hsCRP), and matrix metalloproteinases (MMPs)-1,-3,-9. Associations between biomarkers and radiographic severity KL1/2 vs. KL3/4 (logistic regression controlling for covariates) and pain (Spearman correlation) were evaluated. RESULTS:Elevated levels of sVAP-1 observed in OA synovial fluid and VAP-1 expression in synovium based on immunohistochemical, microarray, and real-time quantitative polymerase chain reaction (qRT-PCR) analyses. However, serum sVAP-1 levels in OA patients were lower than in controls and inversely correlated with pain and inflammation markers (hsCRP and soluble RAGE). Soluble VAP-1 levels in serum were also lower in radiographically advanced (KL3/4) compared with early KL1/2 knee SKOA patients. CONCLUSION/CONCLUSIONS:Local (synovial fluid) semicarbazide-sensitive amine oxidase (SSAO)/sVAP-1 levels were elevated in OA and correlated with radiographic severity. However, systemic (serum) sVAP-1 levels were lower in SKOA patients than normal and inversely correlated with pain and inflammation markers. Serum sVAP-1 levels were higher in early (KL1/2) compared with advanced (KL3/4) SKOA patients.

OBJECTIVE:To validate a radial imaging spin-echo diffusion tensor (RAISED) sequence for high-resolution diffusion tensor imaging (DTI) of articular cartilage at 3 T. METHODS:The RAISED sequence implementation is described, including the used non-linear motion correction algorithm. The robustness to eddy currents was tested on phantoms, and accuracy of measurement was assessed with measurements of temperature-dependent diffusion of free water. Motion correction was validated by comparing RAISED with single-shot diffusion-weighted echo-planar imaging (EPI) measures. DTI was acquired in asymptomatic subjects (n = 6) and subjects with doubtful (Kellgren-Lawrence [KL] grade 1, n = 9) and mild (KL = 2, n = 9) symptomatic knee osteoarthritis (OA). MD and FA values without correction, and after all corrections, were calculated. A test-retest evaluation of the DTI acquisition on three asymptomatic and three OA subjects was also performed. RESULTS:The root mean squared coefficient of variation of the global test-restest reproducibility was 3.54% for MD and 5.34% for FA. MD was significantly increased in both femoral condyles (7-9%) of KL 1 and in the medial (11-17%) and lateral (10-12%) compartments of KL 2 subjects. Averaged FA presented a trend of lower values with increasing KL grade, which was significant for the medial femoral condyle (-11%) of KL 1 and all three compartments in KL 2 subjects (-18 to -11%). Group differences in MD and FA were only significant after motion correction. CONCLUSION/CONCLUSIONS:The RAISED sequence with the proposed reconstruction framework provides reproducible assessment of DTI parameters in vivo at 3 T and potentially the early stages of the disease in large regions of interest. KEY POINTS/CONCLUSIONS:• DTI of articular cartilage is feasible at 3T with a multi-shot RAISED sequence with non-linear motion correction. • RAISED sequence allows estimation of the diffusion indices MD and FA with test-retest errors below 4% (MD) and 6% (FA). • RAISED-based measurement of DTI of articular cartilage with non-linear motion correction holds potential to differentiate healthy from OA subjects.

OBJECTIVE:Phenotypic changes of chondrocytes toward hypertrophy might be fundamental in the pathogenesis of OA, of which type X collagen (Col10) is a well-known marker. The purpose was to develop a specific immunoassay for blood quantification of a newly identified neo-epitope of type X collagen to assess its diagnostic value for radiographic knee osteoarthritis (OA). METHODS:A neo-epitope of Col10 was identified in urine samples from OA patients. A monoclonal antibody against the neo-epitope was produced in Balb/C mice. The enzyme responsible for the cleavage was identified. Immunohistochemical detection of this neo-epitope was performed on human OA cartilage. An immunoassay (Col10neo) was developed and quantified in two clinical studies: the C4Pain-003 and the NYU OA progression study. ROC curve analysis was carried out to evaluate the discriminative power of Col10neo between OA and RA. RESULTS:A neo-epitope specific mAb was produced. The Cathepsin K-generated neo-epitope was localized to the pericellular matrix of chondrocytes, while its presence was extended and more prominent in superficial fibrillation in the cartilage with advanced degradation. In the C4Pain study, a higher level of Col10neo was seen in subjects with greater KL grade. The group of the highest tertile of Col10neo included more subjects with KL3-4. In the NYU study, Col10neo was statistically higher in OA than control or RA. ROC curve analysis revealed area under the curve was 0.88 (95% CI 0.81-0.94). CONCLUSION/CONCLUSIONS:Our findings indicate that Col10neo linked to hypertrophic chondrocytes could be used as a diagnostic biochemical marker for knee OA.

Purpose: To obtain knee ultrasound (KUS) imaging and assessments on an ongoing community-based cohort study to provide additional information regarding features of knee OA beyond what is found on radiographs. Method(s): A radiology technologist was trained to obtain standardized bilateral KUS images of consecutive individuals attending the 4^th follow-up visit (T4) of the Johnston County OA Project in 2017-18 per a written protocol. A scoring atlas was developed and subsequently revised to allow for semi-quantitative grading of key features by four US-trained rheumatologists, all of whom had input into the training and development activities. Weighted kappa statistics were generated for each feature, by side and combined, along with Kendall coefficients of concordance and percentage agreement. This was first done on initial random selection of 50 participants, followed by discussion via web conference and second round of scoring on 15 participants (11 women/4 men; 11 white/4 African American) selected to represent a range of scores by feature; the latter is presented here. The mean, SD, median, and scoring range was calculated for each feature across the four readers. Result(s): The protocol includes 7 views per knee: Longitudinal and transverse suprapatellar in 30 degrees flexion (for effusion, synovitis by gray scale and color power Doppler (CPD)), medial and lateral longitudinal (for osteophytes, meniscal damage), suprapatellar transverse in maximal flexion (for cartilage damage) and posterior transverse (for cysts). The feature scores (11 score features per knee, see Table) were consistent with mild to moderate pathology in these 30 knees that were selected to represent a range of disease. In particular, mean and median scores for effusion/synovitis and medial cartilage damage demonstrated moderate disease severity. The overall reliability was in the fair to moderate (0.2-0.6) range, and was better for features of synovitis by CPD, osteophytes, and cartilage damage compared with effusion/synovitis by gray scale and meniscal damage (Table). Reliability as assessed simultaneously for all readers by the Kendall coefficient was somewhat better than as assessed by pairwise weighted kappa statistics. The next steps for this project will be to score the full set of 200 participants' US images (400 knees) from T4 and to assess these (likely again as a mean or median) in conjunction with radiographic and symptomatic data in these same knees. [Figure presented] Conclusion(s): US of the knee has been added to the protocol of the Johnston County OA Project and will provide additional information beyond the already extensive radiographic, symptomatic, and clinical data on these participants in a feasible and sustainable way. Further work will characterize the cross-sectional and longitudinal associations among US, radiographic and other features in the parent cohort as well as the new Johnston County Health Study cohort (a new enrollment cohort beginning at age 35 and including Hispanic individuals).

INTRODUCTION/BACKGROUND:Bariatric surgery reduces obesity and knee osteoarthritis (OA) pain, but some patients improve more than others. We aimed to identify characteristics that predict this knee pain improvement. METHODS:We reviewed NYU Langone Health bariatrics records (2002-2015) and called eligible patients reporting pre-operative knee pain. Patients were asked to rate their pain on a 10-point scale at three time points: before surgery, one year post-surgery, and time of survey administration. Subjects were asked about pre-operative knee injuries and surgeries, presence of OA in other joints, and OA family history. Data were analyzed using paired t-tests and ANOVA. RESULTS:Of 125 eligible patients reporting knee pain, we analyzed the 120 patients who had laparoscopic gastric band (LAGB) surgery. The cohort was 78.3% female, with an average age at surgery of 49.7 ± 10.2 years. There was no correlation between pre-operative body mass index (BMI) and knee pain reduction at one year post-LAGB, but the subgroup with the most BMI improvement reported the most knee improvement (p = 0.043). We found significantly better pain reduction after one year in younger patients (p = 0.009). Those with prior knee injuries improved less than those who were injury-free (p = 0.044), but a history of prior knee surgery was not similarly significant. Patients with multifocal OA improved less (p = 0.001). CONCLUSION/CONCLUSIONS:Younger knee OA patients and those without prior knee injury or other OA involvement, experience more knee pain relief from LAGB weight loss surgery. LAGB may be a viable treatment option for knee OA pain, irrespective of the degree of obesity.

OBJECTIVE:The aims of this study were to systematically review clinimetrics of commonly assessed ultrasound pathologies in knee, hip and hand osteoarthritis (OA), and to conduct a meta-analysis for each clinimetric. METHODS:Medline, Embase, and Cochrane Library databases were searched from their inceptions to September 2016. According to the Outcome Measures in Rheumatology (OMERACT) Instrument Selection Algorithm, data extraction focused on ultrasound technical features and performance metrics. Methodological quality was assessed with modified 19-item Downs and Black score and 11-item Quality Appraisal of Diagnostic Reliability (QAREL) score. Separate meta-analyses were performed for clinimetrics: (1) inter-rater/intra-rater reliability; (2) construct validity; (3) criteria validity; and (4) internal/external responsiveness. Statistical Package for the Social Sciences (SPSS), Excel and Comprehensive Meta-analysis were used. RESULT/RESULTS:Our search identified 1126 records; of these, 100 were eligible, including a total of 8542 patients and 32,373 joints. The average Downs and Black score was 13.01, and average QAREL was 5.93. The stratified meta-analysis was performed only for knee OA, which demonstrated moderate to substantial reliability [minimum kappa > 0.44(0.15,0.74), minimum intraclass correlation coefficient (ICC) > 0.82(0.73-0.89)], weak construct validity against pain (r = 0.12 to 0.27), function (r = 0.15 to 0.23), and blood biomarkers (r = 0.01 to 0.21), but weak to strong correlation with plain radiography (r = 0.13 to 0.60), strong association with Magnetic Resonance Imaging (MRI) [minimum r = 0.60(0.52,0.67)] and strong discrimination against symptomatic patients (OR = 3.08 to 7.46). There was strong criterion validity against cartilage histology [r = 0.66(-0.05,0.93)], and small to moderate internal [standardized mean difference(SMD) = 0.20 to 0.58] and external (r = 0.35 to 0.43) responsiveness to interventions. CONCLUSION/CONCLUSIONS:Ultrasound demonstrated strong criterion validity with cartilage histology, poor to strong correlation with patient findings and MRI, moderate reliability, and low responsiveness to interventions. PROSPERO REGISTRATION NO/UNASSIGNED:CRD42016039954.