Morphological assessment of non-human primate models of osteoarthritis using HR-MRI and µCT arthrography

(1)

HAL Id: hal-01984459

https://hal.archives-ouvertes.fr/hal-01984459

Submitted on 17 Jan 2019

HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

Morphological assessment of non-human primate models of osteoarthritis using HR-MRI and

µCT arthrography

Emmanuel Chereul, Denis Grenier, Anne-Laure Perrier, F. Taborik, L.

Mahieu-Williame, Kevin Tse-Ve-Koon, T. Chuzel, S Martin, L Magnier, X.

Pesesse, et al.

To cite this version:

Emmanuel Chereul, Denis Grenier, Anne-Laure Perrier, F. Taborik, L. Mahieu-Williame, et al.. Mor- phological assessment of non-human primate models of osteoarthritis using HR-MRI andµCT arthrography. ISMRM, 2013, Salt Lake City, United States. �hal-01984459�

(2)

Morphological assessment of non-human primate models of osteoarthritis using HR-MRI and µCT arthrography

E. Chereul ¹ , D. Grenier ² , A-L. Perrier ² , F. Taborik ³ , L. Mahieu-Williame ² , K. Tse Ve Koon ² , T. Chuzel ¹ , S. Martin ¹ , L. Magnier ¹ , X. Pesesse ⁴ , S. Pietri ⁵ , H. Contamin ³ and O. Beuf ²

ISMRM, Salt Lake City, USA, 2013

References:

1. R. Bolbos et al., Osteoarthritis Cartilage 15:656-65 (2007).

2. A. Rengle et al., IEEE Trans Biomed Eng 56:2891-2897 (2009).

3. JC. Goebel et al., Rheumatology 49:1654-1664 (2010).

4. T. Hildebrand and P. Ruegsegger, J Microsc 185:67–75 (1997).

1

VOXCAN, Marcy l’Etoile, France ;

²

Université de Lyon, CREATIS CNRS UMR 5220; Inserm U1044; INSA-Lyon; Université Lyon 1, Villeurbanne, France ;

³

Cynbiose, Marcy l’Etoile, France ;

⁴

Bone Therapeuthics, Grosselies, Belgium;

⁵

Laboratoire de Rhumatologie, Hôpital Erasme, ULB, Bruxelles, Belgium.

Introduction

Conclusion

 Small animal models of osteoarthritis (OA) do not mimic perfectly the complex conditions occurring in human OA.

 OA that closely resembles the human condition occurs naturally in primate. Non-human primates (NHP) could be a useful model for human OA.

 Non-invasive techniques such as 3D HR-MRI have been validated to directly assess the cartilage thickness on guinea pigs (1) and different cartilage compartment volumes on rat models of OA (2-3).

 Nonetheless, spatial resolution is limited compared to µCT scanner that however needs contrast agent injected in the joint to depict cartilage limits.

Acknowledgements: This work was funded by OSEO with grant E! 5671 from the Eurostars European program and performed within the framework of the LABEX PRIMES (ANR-11-LABX-0063) of Université de Lyon, within the program "Investissements d'Avenir" (ANR-11-IDEX-0007) operated by the French National Research Agency (ANR).

Results

MRI acquisition protocol

Material and method

[email protected]

 No differences were found with MRI examination between non injected and injected knees (required for the µCTA protocol) over the time.

 No difference was found between µCTA- and MRI-based cartilage thickness methods on control groups (G1&G3) since no residual measurement between both methods were found above the resolution of these techniques. Differences shown with G2 have to be further investigated.

 NHP model of OA was characterized by both imaging methods showing a monotone progression of the cartilage thinning up to -24.6±5.7% on D90 and -27.2%±5.2 on D180.

Perspectives

 MRI and µCTA modalities are valuable to measure cartilage morphology.

 Additional information can be obtained:

 Indirectly about cartilage structure (T2, T1rho…) with MRI

 Subchondral bone density with µCTA

 The aim of this work, based on morphological parameters assessed on MRI and µCT arthrography (µCTA) acquisitions, was:

 To assess the potential µCTA protocol impact on the model follow-up

 To compare quantifications results based on both imaging modalities

 To characterize an induced model of OA by transection of the anterior cruciate ligament (ACL).

Objectives

Susceptibility tissue-matched foam

Prof. Conolly, UC Berkeley.

5 mm

 1.5T Siemens Sonata system

 3D water excitation FLASH sequence: 25°

flip angle, 27 ms TR, 11.7 ms TE, 70 Hz/Pixel receiver bandwidth

 A pair of homemade two-channel array coil

 In-plane pixel: 112x131 µm2, partition thickness: 220 µm

 scan time/knee : 20 min

 Total examination time: 90 min

 2mL HexabrixTM (320mg/mL) with 40/60 dilution in PBS was injected in the synovial capsule with 23G needle.

 GE Locus µ-CT (standard voltage and amperage)

 Isotropic voxel of 90 µm

 Scan time for both knees : 15 min

 Total examination time: 30 min

µCTA acquisition protocol

Image processing protocol and cartilage quantification parameters

 The ethical guidelines for animal experimental investigations were followed and the experimental protocol was approved by the Animal Ethics Committee from Ecole nationale vétérinaire de Lyon (VetAgro Sup), Marcy l’Etoile.

 Group 1&3 (n=3+3): control animals – only the right knee was injected with HexabrixTM for µCTA imaging. MR imaging of both knees.

 Group 2 (n=6): ACL transection of the right knee only. Multi-modal imaging of both knees.

 Longitudinal follow-up using HR-MRI and µCTA of 4 year old female cynomolgus at baseline, D15 (µCTA only), D30, D60, D90, D180.

 Binarization performed in two steps:

 Manual segmentation

 Global thresholding

 Thickness distribution (4)

G1 & G3 - MRI-based quantification of left and right tibial plateaus

0.960.97 0.961.05 1.011.03 0.961.02 1.021.01

0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40

D00 D30 D60 D90 D180

Mean of bone cartilage average thickness (mm)

Time step

Groups 1&3 - Mean of medial bone cartilage average thickness of right and left paw obtained by MRI.

Left paw Right paw

0.420.42 0.420.45 0.430.47 0.420.46 0.430.46

0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40

D00 D30 D60 D90 D180

Time step

Groups 1&3 - Mean of lateral bone cartilage thickness of right and left paw obtained by MRI.

Left paw Right paw

0%

5%

10%

15%

20%

25%

0 0.5 1 1.5 2 2.5

Mean of moving average of occurrence (%)

Thickness of bone cartilage (mm)

Groups 1&3 - Average distribution of medial bone cartilage thickness of right and left paw, obtained by MRI.

Right paw Left paw

0%

5%

10%

15%

20%

25%

0 0.5 1 1.5 2 2.5

Mean of moving average of occurrence (%)

Thickness of bone cartilage (mm)

Groups 1&3 - Average distribution of lateral bone cartilage thickness of right and left paw, obtained by MRI.

Right paw Left paw

Repeated injections for µCTA have no impact on cartilage quantification

G1&G3 vs G2 - Mean of tibial cartilage average thicknesses of right knee obtained by µCTA and MRI

0.97

1.05 1.03 1.02 1.01

1.00 1.00 1.02 1.02 1.03

0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40

D00 D15 D30 D60 D90 D180

Time step

Groups 1&3 - Mean of medial bone cartilage average thickness of right paw obtained by 2 methods: MRI and µ-

CT. MRI

µ-CT

0.42 0.45 0.47 0.46 0.46

0.43 0.42 0.42 0.43 0.43

0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40

D00 D15 D30 D60 D90 D180

Time step

Groups 1&3 - Mean of lateral bone cartilage average thickness of right paw obtained by 2 methods: MRI and µ-

CT. MRI

µ-CT

1.01 1.02

0.90 0.88

0.84

1.01 0.99

0.88

0.83

0.76 0.78

0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40

D00 D15 D30 D60 D90 D180

Time step

Group 2 - Mean of medial bone cartilage average thickness of right paw obtained by 2 methods: MRI and µ-CT.

MRI µ-CT

0.54 0.54 0.51

0.44 0.42

0.45 0.44 0.40 0.38

0.37 0.35

0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40

D00 D15 D30 D60 D90 D180

Time step

Group 2 - Mean of lateral bone cartilage average thickness of right paw obtained by 2 methods: MRI and µ-CT.

MRI µ-CT

Similar trend between μCTA and MRI with systematic thickness overestimation by MR

G1&G3 – control group

G2 – ACL transection of right knee

(*) P=0.03

(**)

P=0.003 (*)

P=0.031 (*)

P=0.017 (*)

P=0.047

G1&G3 vs G2 - Mean of tibial cartilage average thicknesses of right knee obtained by µCTA

1.00 1.00 1.02 1.02 1.03

1.01 0.99

0.88

0.83

0.76 0.78

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4

D00 D15 D30 D60 D90 D180

Mean of bone cartilage average thickness (mm)

Time step

Groups 1&3 and Group 2 - Mean of medial bone cartilage average thickness of right paw obtained by µ-CT.

Groups 1&3 Group 2

0.430.45 0.44 0.420.40 0.420.38 0.430.37 0.43 0.35

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4

D00 D15 D30 D60 D90 D180

Mean of bone cartilage average thickness (mm)

Time step

Groups 1&3 and Group 2 - Mean of lateral bone cartilage average thickness of right paw obtained by µ-CT

Groups 1&3 Group 2

Morphological assessment of non-human primate models of osteoarthritis using HR-MRI and µCT arthrography

Morphological assessment of non-human primate models of osteoarthritis using HR-MRI and

Morphological assessment of non-human primate models of osteoarthritis using HR-MRI and µCT arthrography

E. Chereul 1 , D. Grenier 2 , A-L. Perrier 2 , F. Taborik 3 , L. Mahieu-Williame 2 , K. Tse Ve Koon 2 , T. Chuzel 1 , S. Martin 1 , L. Magnier 1 , X. Pesesse 4 , S. Pietri 5 , H. Contamin 3 and O. Beuf 2

ISMRM, Salt Lake City, USA, 2013

References:

1. R. Bolbos et al., Osteoarthritis Cartilage 15:656-65 (2007).

2. A. Rengle et al., IEEE Trans Biomed Eng 56:2891-2897 (2009).

3. JC. Goebel et al., Rheumatology 49:1654-1664 (2010).

4. T. Hildebrand and P. Ruegsegger, J Microsc 185:67–75 (1997).

VOXCAN, Marcy l’Etoile, France ;

Université de Lyon, CREATIS CNRS UMR 5220; Inserm U1044; INSA-Lyon; Université Lyon 1, Villeurbanne, France ;

Cynbiose, Marcy l’Etoile, France ;

Bone Therapeuthics, Grosselies, Belgium;

Laboratoire de Rhumatologie, Hôpital Erasme, ULB, Bruxelles, Belgium.

Introduction

Conclusion

 Small animal models of osteoarthritis (OA) do not mimic perfectly the complex conditions occurring in human OA.

 OA that closely resembles the human condition occurs naturally in primate. Non-human primates (NHP) could be a useful model for human OA.

 Non-invasive techniques such as 3D HR-MRI have been validated to directly assess the cartilage thickness on guinea pigs (1) and different cartilage compartment volumes on rat models of OA (2-3).

 Nonetheless, spatial resolution is limited compared to µCT scanner that however needs contrast agent injected in the joint to depict cartilage limits.

Results

MRI acquisition protocol

Material and method

[email protected]

 No differences were found with MRI examination between non injected and injected knees (required for the µCTA protocol) over the time.

 No difference was found between µCTA- and MRI-based cartilage thickness methods on control groups (G1&G3) since no residual measurement between both methods were found above the resolution of these techniques. Differences shown with G2 have to be further investigated.

 NHP model of OA was characterized by both imaging methods showing a monotone progression of the cartilage thinning up to -24.6±5.7% on D90 and -27.2%±5.2 on D180.

Perspectives

 MRI and µCTA modalities are valuable to measure cartilage morphology.

 Additional information can be obtained:

 Indirectly about cartilage structure (T2, T1rho…) with MRI

 Subchondral bone density with µCTA

 The aim of this work, based on morphological parameters assessed on MRI and µCT arthrography (µCTA) acquisitions, was:

 To assess the potential µCTA protocol impact on the model follow-up

 To compare quantifications results based on both imaging modalities

 To characterize an induced model of OA by transection of the anterior cruciate ligament (ACL).

Objectives

Susceptibility tissue-matched foam

Prof. Conolly, UC Berkeley.

5 mm

 1.5T Siemens Sonata system

 3D water excitation FLASH sequence: 25°

flip angle, 27 ms TR, 11.7 ms TE, 70 Hz/Pixel receiver bandwidth

 A pair of homemade two-channel array coil

 In-plane pixel: 112x131 µm2, partition thickness: 220 µm

 scan time/knee : 20 min

 Total examination time: 90 min

 2mL HexabrixTM (320mg/mL) with 40/60 dilution in PBS was injected in the synovial capsule with 23G needle.

 GE Locus µ-CT (standard voltage and amperage)

 Isotropic voxel of 90 µm

 Scan time for both knees : 15 min

 Total examination time: 30 min

µCTA acquisition protocol

Image processing protocol and cartilage quantification parameters

 The ethical guidelines for animal experimental investigations were followed and the experimental protocol was approved by the Animal Ethics Committee from Ecole nationale vétérinaire de Lyon (VetAgro Sup), Marcy l’Etoile.

 Group 1&3 (n=3+3): control animals – only the right knee was injected with HexabrixTM for µCTA imaging. MR imaging of both knees.

 Group 2 (n=6): ACL transection of the right knee only. Multi-modal imaging of both knees.

 Longitudinal follow-up using HR-MRI and µCTA of 4 year old female cynomolgus at baseline, D15 (µCTA only), D30, D60, D90, D180.

 Binarization performed in two steps:

 Manual segmentation

 Global thresholding

 Thickness distribution (4)

G1 & G3 - MRI-based quantification of left and right tibial plateaus

Repeated injections for µCTA have no impact on cartilage quantification

G1&G3 vs G2 - Mean of tibial cartilage average thicknesses of right knee obtained by µCTA and MRI

Similar trend between μCTA and MRI with systematic thickness overestimation by MR

G1&G3 – control group

G2 – ACL transection of right knee

G1&G3 vs G2 - Mean of tibial cartilage average thicknesses of right knee obtained by µCTA

Significative decrease of cartilage thickness starting at D90 and D180 on NPH model of OA

Medial Lateral

Medial Lateral

Medial Lateral

E. Chereul ¹ , D. Grenier ² , A-L. Perrier ² , F. Taborik ³ , L. Mahieu-Williame ² , K. Tse Ve Koon ² , T. Chuzel ¹ , S. Martin ¹ , L. Magnier ¹ , X. Pesesse ⁴ , S. Pietri ⁵ , H. Contamin ³ and O. Beuf ²