Biomarkers Musculoskeletal
Musculoskeletal
Update: April 2025
| Biomarker | Units of Measurement | Acquisition Modality | Data acquisition requirements | Extracting biomarker (Reading/Algorithm) | Pathophysiological Process |
Use of biomarker Diagnosis Prognosis Prediction Treatment evaluation |
Accuracy/Agreement | Evidence level | References | Issues/Limitations | |
| Knee joint space width on radiography | mm | Radiography | Weight-bearing extended or semi-flexed knee radiographs, sometimes using an positioning frame | Manual / semi-automatic / automatic segmentation | Reflects cartilage thickness loss (in osteoarthritis) | Research only (diagnosis, prognosis, prediction, treatment evaluation) | Test/retest precision: ICC 0.98, RMSE 0.23 mm, RMSE CV 7.4% | 4 (Diagnosis and prognosis) | [1-3!] | lack of standardization, indirect assessment of cartilage loss, captures only moderate to advanced stage OA and insentive to change | |
| Cartilage thickness on MRI | mm | MRI | Usually high-field MRI, regions of interest or 2D maps, usually on DESS sequences | Manual / semi-automatic / automatic segmentation | Reflects structural integrity and cartilage substance loss | In research only: Diagnosis, Prediction, Treatment evaluation | Test/retest precision: 69micrometer | 4 (Diagnosis and prognosis) | [4-7!] | interindividual variability / intraindividual variabilty, lack of standardization of aquisition protocol, segmentation may be challenging | |
| T1Gd relaxation time (dGEMRIC) | ms | MRI | Usually high-field MRI, administration of double dose gadolinium-based contrast agent, acquisition after 90-120min delay, T1-weighted sequences with inversion recovery to measure T1 relaxation times | Fitting data to an exponential recovery curve to obtain T₁ values from variable inversion time images | Proteoglycan content | In research only: Diagnosis, Prediction, Treatment evaluation | Correlation between dGEMRIC and proteglycan concentration (meta-analysis): r=0.59 [0.41, 0.73] | 2a (Diagnosis) | [8!] | lack of standardization, requires intravenous injection of a contrast agent with potential side effects, delay time necessary between contrast injection and imaging needs, ariability in contrast uptake and distribution due to individual patient factors | |
| T2 mapping | ms | MRI | Techniques vary, high-field MRI (usually 3T) Gradient-Echo or Spin-Echo, Echo Times from 0 to 80ms | Fitting data to an exponential decay curve to obtain T2 values from multi-echo images | Water and collagen content, and orientation of collagen fibers | In research only: Diagnosis, Prediction, Treatment evaluation | Correlation between T2 mapping and collagen anistropy: R2=0.44 (at 9.4T, from 3 species) | 2a (Diagnosis) | [9-14!] | lack of standardization (QIBA initiative under work), interindividual variability | |
| T1rho mapping | ms | MRI | Techniques vary, high-field MRI (usually 3T), Gradient-Echo or Spin-Echo, Echo Times from 0 to 80ms | Fitting data to an exponential decay curve to obtain T₁ρ values from spin-lock images | Proteoglycan content | In research only: Diagnosis, Prediction, Treatment evaluation | Correlation between T1 rho and proteglycan concentration (meta-analysis): r=-0.54 [-0.73, -0.29] | 2a (Diagnosis) | [15-20!] | lack of standardization (QIBA initiative under work), interindividual variability | |
| Quantitative DCE-MRI Perfusion parameter Ktrans (for soft tissue sarcoma differentiation) | min-1 | MRI | DCE-MRI acquisiton: dynamic imaging through fast or ultrafast sequences after intravenous contrast agent administration.Protocols vary wrt temporal and spatial resolution | Extraction of quantitative parameters (e.g. Ktrans representing contrast transit from the vascular compart ment to the interstitial compartment) after pharmacokinetic modeling of exchanges between the plasmatic compartment and the tumor interstitium. | Capillary permeability | Mainly clinical research: Diagnosis | Sensitvity 79-81%, varying specificity (27-77%) | 4 (Diagnosis) | [21!] | limited evidence for clinical implementation, lack of standardization | |
| Areal bone mineral density (BMD) | gr/m2 | DXA | Specific bones, usually spine/hip/wrist | Bone density compared with average index based on age, sex, and size, expressed as T-score and Z-score | Indirect assessment of amount of bone mineral (calcium hydroxyapatite) in bone tissue. | Clinical and research (diagnosis, prognosis, prediction, treatment evaluation) | Short-term precision in-vivo: SD RMS 0.008-0.012, %CV RMS 0.55-1.60 (depending on site) | 1a | [22!] | prone to errors (positioning, ROI, analysis) and artefacts (OA, calcifications), measures bone quantity rather than quality | |
| Volumetric bone mineral density | gr/m3 | CT (QCT) | Typically non-contrast CT acquisition of lumbar spine | Average bone mineral density based on HU compared to age and sex matched controls (abnormality determined against guideline thresholds) | Indirect assessment of amount of bone mineral (calcium hydroxyapatite) in bone tissue. | Mostly research but increasing clinical implementation (opportunistic finding) | Coefficient of variation (CV) of 0.8% from repeated examinations with repositioning | 2b | [23!] | typically requires calibration, analysis methods not widely available | |
| Trabecular bone score (TBS) | no units | DXA | Lumbar spine DXA | Grey-level bone texture analysis | Bone microarchitecture | Clinical and research (diagnosis, prognosis, prediction, treatment evaluation) | Short-term reproducibility 95%/CoV 1.9% | 1a | [24-25!] | reliant on one specific commercial software |
- Hellio Le Graverand M-P, Mazzuca S, Duryea J, Brett A. Radiographic-based grading methods and radiographic measurement of joint space width in osteoarthritis. Radiol Clin N Am. 2009;47:567–579.
- Nevitt et al. Arthritis Rheum. 2007 May;56(5):1512-20
- Oak SR Osteoarthritis Cartilage. 2013 21(9):1185-90.
- Eckstein F et al., Felson DT. Osteoarthritis Cartilage (2010) 18:1275-1283
- Favre J et al., Andriacchi TP. J Orthop Res (2017) 35:2442-2451
- Wirth W et al., Jansen M. Osteoarthritis Cartilage (2023) 31:238-248
- Eckstein. Arthritis Rheumatol 2017 7(12):3184-9
- Emanuel KS et al., Emans PJ. Osteoarthritis Cartilage (2022) 30:650-662
- Chalian M et al., SNA QIBAMSKBCM. Radiology (2021) 301:423-432
- MacKay JW et al., Gilbert FJ. Osteoarthritis Cartilage (2018) 26:1140-1152
- Kim J et al., Li X. Osteoarthritis Cartilage (2020) 28:1539-1550
- Li X et al., Majumdar S. J Magn Reson Imaging (2014) 39:1287-1293
- Li X et al., Majumdar S. Osteoarthritis Cartilage (2015) 23:2214-2223
- Verschueren J et al., Oei EHG. Quant Imaging Med Surg (2021) 11:1247-1255
- Emanuel KS et al., Emans PJ. Osteoarthritis Cartilage (2022) 30:650-662
- Chalian M et al., SNA QIBAMSKBCM. Radiology (2021) 301:423-432
- MacKay JW et al., Gilbert FJ. Osteoarthritis Cartilage (2018) 26:1140-1152
- Kim J et al., Li X. Osteoarthritis Cartilage (2020) 28:1539-1550
- Li X et al., Majumdar S. J Magn Reson Imaging (2014) 39:1287-1293
- Li X et al., Majumdar S. Osteoarthritis Cartilage (2015) 23:2214-2223
- Shomal Zadeh F et el.. Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) in differentiation of soft tissue sarcoma from benign lesions: a systematic review of literature. Skeletal Radiol. 2024 Jul;53(7):1343-1357.
- Mesinovic J et al. Bone imaging modality precision and agreement between DXA, pQCT, and HR-pQCT. JBMR Plus. 2024 Dec 3;9(2):ziae158. doi: 10.1093/jbmrpl/ziae158.
- Adams JE. Quantitative computed tomography. Eur J Radiol. 2009 Sep;71(3):415-24.
- Shevroja E. et al. Clinical Performance of the Updated Trabecular Bone Score (TBS) Algorithm, Which Accounts for the Soft Tissue Thickness: The OsteoLaus Study. J Bone Miner Res. 2019 Dec;34(12):2229-2237
- Bandirali M et al. Short-term precision assessment of trabecular bone score and bone mineral density using dual-energy X-ray absorptiometry with different scan modes: an in vivo study. Eur Radiol. 2015 Jul;25(7):2194-8.