Increased serum AXL is associated with radiographic knee osteoarthritis severity

Abstract Objective To investigate the expression and clinical significance of serum soluble AXL in patients with radiographic knee osteoarthritis (KOA). Methods There were 183 patients with KOA who were selected and divided based on the Kellgren‐Lawrence (KL) score into KL 0 subgroups (n = 42), KL I‐II subgroups (n = 90), and KL III‐IV subgroups (n = 51). Healthy volunteers (n = 170) in our hospital were selected with matched age and gender as the control group. AXL level in serum was detected by enzyme‐linked immunosorbent assay. The correlation between serum AXL with severity and clinical indicators of osteoarthritis was analyzed. Results The level of serum AXL was significantly higher in the osteoarthritis group than that in the control group (P < .05). In the osteoarthritis patients, serum AXL level was increased with the increase of KL score. Serum AXL level was positively correlated with age, body mass index, erythrocyte sedimentation rate, serum C‐reactive protein, cartilage oligomeric protein, matrix metalloproteinase‐13, and transforming growth factor‐β1 levels. The cut‐off value for serum AXL was determined as 33.375 ng/mL by receiver operating curve analysis. Conclusion The level of serum AXL in patients with osteoarthritis is significantly higher than in healthy controls, and is closely related to the severity of radiographic osteoarthritis.

Japanese population was reported as up to 42.0% in men and 62.4% in women aged over 40 years. 10 Therefore, KOA is a global issue in the elderly population and early diagnosis is required to begin possible treatment.
The radiography imaging technique is viewed as a gold standard method for diagnosis of KOA, but the current imaging technique is suffering from sensitivity and specificity. 11 Imaging technique allows detection of OA of the knee joint space narrowing, presence of osteophytes, subchondral sclerosis, and cysts. Due to the lack of sensitivity and specificity of radiographic imaging techniques, there is an urgency to develop a potential alternative tool for the diagnosis of KOA. Body fluid serum is routinely tested in clinics for diagnosis, and treatment of different diseases. This is a very decisive medium and harbors plenty of biomarkers for the monitoring of our health.
Several biomarkers are known to be correlated with the extent of OA on radiography of the knee and are being proposed as diagnostic tools. [12][13][14][15] However, the currently used biomarkers are inadequate for the prognosis of radiographic KOA.
The receptor tyrosine kinase AXL is a 140 kDa protein that belongs to a tyrosine kinase receptor (TAM) subfamily, together with Tyro3 and Mer. The TAM receptors (AXL, Tyro3, and Mer) play a critical role in innate immune homeostasis and vitamin K-dependent ligand growth arrest-specific protein 6 (GAS6) can bind all 3 receptors with the highest affinity for AXL. 16,17 Transmembrane protein AXL can be cleaved proteolytically at its extracellular membrane domain and subsequently released as soluble AXL, which can be detected in serum or plasma. 18,19 Furthermore, studies revealed that targeted delivery of TAM receptor ligand genes Gas6 diminishes the arthritis pathology effectively but the endogenous role of AXL in arthritis development is not fully understood. 20

| Subjects
This study included 183 patients with KOA who were admitted to our hospital from July 2019 to December 2020. There were 102 males and 81 females, with a median age of 68 years. Patient inclusion criteria were as follows: (a) all patients were diagnosed with KOA with duration of disease >6 months; (b) all patients had radiological evidence of osteoarthritis with a KL score of 0-IV. All patients were graded as KL according to the X-ray pictures of bone and joint: grade 0, no change; grade I, slight osteophyte; grade II, obvious osteophytes, no joint space involved; grade III, moderate narrowing of joint space; grade IV, joint space narrowing, subchondral osteosclerosis. 21  All experimental protocols were approved by the ethics committee of the hospital.

| X-ray measurement of knee joint
All subjects were in the weight-bearing standing position, and the Xray films of both knees in the anterior-posterior position and lateral position were taken respectively. The evaluation of KL grade and double-blind measurement of joint structure were carried out by 2 experienced rheumatologists. When the evaluation of the 2 specialists was different, the joint KL grade and joint structure were evaluated by senior specialists in the Department of Radiology.

| Specimen collection and preparation
Whole blood was collected from fasting participants in the morning.
Blood samples were centrifuged (15 000 × g for 10 minutes at 4°C) to separate serum. The processed serum supernatant was then aliquoted into a 1.5 mL Eppendorf tube and stored at −80°C for further analysis.

| Measurement of serum proteins and cytokines concentration
The serum concentrations of cytokines and proteins were determined using enzyme-linked immunosorbent assay (ELISA) kits, including serum C-reactive protein (CRP) (DCRP00, R&D Systems), cartilage oligomeric protein (COMP) (DCMP0), matrix metalloproteinase-13 (MMP-13) (DY511), transforming growth factor-β1 (TGF-β1) (DY240), Gas6 (DY885B), and AXL (DAXL00), according to the manufacturer's protocol. In brief, serum samples were directly transferred or diluted to the wells of the ELISA plate, and the absorbance was measured in a microplate reader. The protein levels were quantified by their corresponding standard curve. SPSS 20.0 software was used for statistical analysis. The measurement data were expressed as median (interquartile range), and analyzed by one-way analysis of variance or Mann-Whitney U test. The correlation between serum AXL level and other clinical indicators was analyzed by Spearman correlation. Receiver operating curve (ROC) analysis was carried out to determine the diagnostic potential of AXL for KOA; the difference was statistically significant at P < .05.

| Baseline demographic and clinical characteristics of subjects
A total of 183 OA patients were enrolled in this study. The baseline demographic and clinical characteristics are shown in Table 1. There was no significant difference in age and gender between the control group and OA group. The mean body mass index (BMI) and erythrocyte sedimentation rate (ESR) were significantly higher in OA patients with respect to healthy controls (both P < .05). Analysis of biochemical parameters revealed significantly higher serum CRP, COMP, MMP-13, TGF-β1 and Gas6 levels in the OA group compared to control group (all P < .05).

| Serum AXL concentrations were elevated in KOA patients
The serum AXL concentrations were analyzed in KOA patients and healthy controls by non-parametric tests. Serum AXL levels were analyzed between males and females. The serum AXL level was sig-  [24.25-29.82] pg/mL, CV: 0.155) (P < .001; Figure 1A).
Also, serum AXL levels were significantly higher in KOA compared to healthy controls, after BMI adjustments (Table S1). Moreover, among OA patients, serum AXL levels were significantly higher in subjects with KL I-II grade compared to subjects with KL 0 grade, higher in subjects with KL III-IV compared to subjects with KL I-II grade (both P < .001; Figure 1B). Additionally, significant difference was found in serum AXL and different KL grades after BMI adjustment (Table S2). A Mann-Whitney test was performed to compare the serum AXL levels between males and females. There was no significant difference observed between male and female groups (Table S3).

| D ISCUSS I ON
Clinical findings are more important than radiographs in KOA to decide further course of management. 22 (Table 1).
These findings are in line with several other studies that indicate that these biochemical parameters are involved in osteoarthritis development and treatments. [25][26][27][28][29] In addition, serum AXL level in OA patients was significantly higher compared with that in the healthy controls ( Figure 1A). Thus, correlation between serum AXL level and biochemical indicators can distinguish KOA patients from healthy controls. Furthermore, Spearman correlation analysis revealed that serum AXL level is positively correlated with the clinical biochemical indicators BMI, ESR, CRP, COMP, MMP-13, and TGFβ (Table 2).
Moreover, a significant positive correlation was observed between serum level AXL and Gas6 in OA patient groups and no correlation among control groups (Figure 2 cut-off value of serum AXL level was obtained by ROC curve as 33.375 ng/mL, which was used to distinguish healthy controls and all osteoarthritis (including KL 0 to IV) ( Figure 3). Thus, based on our biochemical analysis, we revealed that the AXL cut-off value could be useful during screening for radiographic KOA. Serum AXL level might be an indicator of radiographic KOA diagnosis and severity assessments.

CO N FLI C T O F I NTE R E S T
The authors declare no competing financial interests.

E TH I C S A PPROVA L
Our study has been approved by the medical ethics committee of the Shanghai Kaiyuan Orthopedic Hospital and written informed consents were obtained from all included patients.

DATA AVA I L A B I L I T Y S TAT E M E N T
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

F I G U R E 2
The correlation between serum AXL and clinical parameters in osteoarthritis (OA) patients. Spearman correlation test was performed. ***P < .001

F I G U R E 3
Receiver operating curve analysis of serum AXL levels discriminating between patients with knee osteoarthritis (OA) and healthy subjects