Histology-informed microstructural diffusion simulations for MRI cancer characterisation (Histo-µSim): ex vivo mouse data

The Histo-μSim diffusion MRI framework: data from fixed mouse tissue scanned ex vivo

This record contains diffusion MRI and hematoxylin-eosin (HE) histological data from fixed mouse tissue, scanned ex vivo on a 9.4T system.Data have been obtained from MMTV-PyMT transgenic mice and from the folic acid-induced kidney injury model.

- Samples are illustrated in sample_illustration.pdf and sample_illustration.pptx- Animal models and MRI/histology acquisitions are detailed in acquisition_information.txt- Data are stored in the exvivo_mouse.zip folder.

The data has been used in the following pre-print:

"Histology-informed microstructural diffusion simulations for MRI cancer characterisation - the Histo-microSim framework"Athanasios Grigoriou, Carlos Macarro, Marco Palombo, Daniel Navarro-Garcia, Anna Voronova, Kinga Bernatowicz, Ignasi Barba, Alba Escriche, Emanuela Greco, Maria Abad, Sara Simonetti, Garazi Serna, Richard Mast, Xavier Merino, Nuria Roson, Manuel Escobar, Maria Vieito, Paolo Nuciforo, Rodrigo Toledo, Elena Garralda, Roser Sala-Llonch, Els Fieremans, Dmitry S. Novikov, Raquel Perez-Lopez and Francesco Grussu. medrxiv 2024, DOI: 10.1101/2024.07.15.24310280.https://doi.org/10.1101/2024.07.15.24310280 

Animal models

Data were acquired on 8 formalin-fixed ex vivo mousetissue specimens, namely: a non-cancerous breast sample; 3 breast tumours from the MMTV PyMT model [Guy et al., 1992,Attalla et al., 2021], obtained at weeks 9, 11 and 14; a normal spleen and a spleen suffering from splenomegaly, secondary toadvanced breast cancer in one MMTV mouse; two kidneys from C57BL/6 WT male mice (9 weeks old), one normal and onewith folic acid-induced injury [Yan, 2021]. Mice were housed at the Specific Pathogen-Free barrier area of the Vall d’HebronInstitute of Oncology (VHIO). All animal procedures were approved by the Animal Care unit and the Ethics Committee forAnimal Experimentation (CEEA) of the Vall d’Hebron Research Institute (VHIR) and the Generalitat de Catalunya, andwere performed according to the European legal framework for research animal use and bioethics. Animals were monitoreddaily and euthanised upon signs of humane endpoints. Two mouse models were used, generating breast, spleen and kidneysamples. These were processed for further histological analyses, as part of ongoing studies at VHIO. A dMRI scan of thetissue was performed at room temperature before inclusion in paraffin for histology.

MMTV-PyMT transgenic mouse modelThe MMTV-PyMT FVB/NJ mouse strain [Guy et al., 1992] is commonly employed to mimic human breast cancer progres-sion [Attalla et al., 2021]. The model relies on the MMTV long terminal repeat promoter, which drives the expression of theantigen of PyMT, a potent oncogene. These transgenic mice are viable despite loss of lactational ability, which is coincidentwith the transgene expression. Breast tumours arise in virgin and breeder females as well as in males starting from 9 weeksof age. Splenomegaly is also observed at the latter stages of the tumour growth. For this study, we used 4 MMTV-PyMTFVB/NJ female mice, which were euthanised by CO2 asphyxiation at different time points to collect the following samples:non-cancerous breast and non-pathological spleen (2 weeks); a breast tumour at weeks 9, 11 and 14; an enlarged spleen(splenomegaly) at late stage cancer (14 weeks).

Folic acid-induced kidney injuryThe folic acid-induced kidney injury mouse model is based on the fact that high doses of folic acid are toxic, despite being thesame substance beneficial at low doses [Yan, 2021]. For this study, we used two male mice (C57BL/6 WT, approximately9 weeks old), which were intra-peritoneally injected with a single dose of vehicle (300 mM NaHCO3) or with folic acid(250 mg/kg). 30 days after the injection, mice were euthanised by CO2 asphyxiation and the kidneys were collected fordownstream processing.

References[Attalla et al., 2021] Attalla, S., Taifour, T., Bui, T., and Muller, W. (2021). Insights from transgenic mouse models of pymt-induced breast cancer: recapitulating human breast cancer progression in vivo. Oncogene, 40(3):475–491.

[Guy et al., 1992] Guy, C. T., Cardiff, R. D., and Muller, W. J. (1992). Induction of mammary tumors by expression ofpolyomavirus middle t oncogene: a transgenic mouse model for metastatic disease. Molecular and cellular biology,12(3):954–961.

[Yan, 2021] Yan, L.-J. (2021). Folic acid-induced animal model of kidney disease. Animal models and experimentalmedicine, 4(4):329–342.

Acquisition details

Diffusion MRICollected tissues were fixed for 24 hours in buffered 4% formaldehyde, transferred to phosphate-buffered saline (PBS) solution and embedded in 1% agarose gel dissolved in PBS, within a histological cassette. Embedded samples were kept in PBS solution, and scanned at room temperature on a 9.4T Bruker Avance system, with 200 mT/m gradient insert and a RX/TX birdcage coil. The protocol consisted  of 2 b = 0 and 6 DW measurements with: b = {0, 500, 2000, 4500} s/mm2 acquired for each of Δ = {16.5, 37.0} ms, with δ = 12 ms.  Other salient dMRI scan parameters were: fat suppression with a frequency-selective 90 degree gauss512 pulse (bandwidth: 1400.10 Hz);  resolution 0.2 x 0.2 x 0.57 mm3, TE = 55.1 ms, TR = 2250 ms, 3 mutually-orthogonal direction for each gradient timing and b-value.  Scans were denoised and Gibbs ringing mitigated.

HistologyAfter MRI, samples were transferred to 70% ethanol for 24 hours and then embedded in paraffin. 3-μm sections were obtained on a manual microtome and stained with HE, using a robust carousel tissue stainer (Slee Medical) according to common methods. Digital images of the HE-stained sections were acquired on a Hamamatsu C9600-12 scanner (resolution: 0.45 μm).

Content of the unzipped exvivo_mouse.zip folder

./exvivo_mouse/acquisition_information.txt                              --> details on the mouse models and on the MRI and histology acquisition

./exvivo_mouse/histology:                                                                --> HE histological images of 4 breast samples, 2 kidneys, and 2 spleens 

./exvivo_mouse/scans:                                                                       --> diffusion MRI scans

Content of ./exvivo_mouse/histology./exvivo_mouse/histology/histology_images                             --> raw histological images in .ndpi format  ./exvivo_mouse/histology/manual_segmentations                  --> manual segmentations of cells across multiples regions-of-interest                                                                                                                       in SVG format (see medrxiv 2024, DOI: 10.1101/2024.07.15.24310280)                                                    

+Content of ./exvivo_mouse/histology/histology_images./exvivo_mouse/histology/histology_images/breast                --> breast histology. It contains two files:                                                                                                                        fslslice4_BC_week11_week14.ndpi: breast tumours at week 11 and 14                                                                                                                          fslslice4_NC_breast_BC_week9.ndpi: breast tumour at week 9 and                                                                                                                         non-cancerous breast./exvivo_mouse/histology/histology_images/kidney_spleen  --> histology of spleens and kidneys. It contains one file                                                                                                                           with all specimens (fslslice6_KID_SPLEEN_2_HE.ndpi)  

Content of ./exvivo_mouse/scans./exvivo_mouse/scans/breast                                                            --> diffusion MRI (dMRI) of the breast specimens (pulsed gradient spin echo)../exvivo_mouse/scans/kidney_spleens                                           --> dMRI of the kidney and spleen specimens.

Each of these two folders contains the following files:dwi.nii --> raw dMRI scandwi.bval --> b-values in s/mm2 of dwi.nii (FSL format)dwi.bvec --> gradient directions of dwi.nii (FSL format)dwi.gdur --> gradient duration of dwi.nii in ms (FSL format)dwi.gsep --> gradient separation of dwi.nii in ms (FSL format)dwi_denoise_unring.nii --> preprocessed dMRI scan (MP-PCA denoising and Gibbs unringing)dwi_denoise_unring_sphmean.nii  --> spherical mean of dwi_denoise_unring.nii (average of images for x, y and z gradients)dwi_denoise_unring_sphmean.bval --> approximated b-value of each volume of dwi_denoise_unring_sphmean.nii, in s/mm2 (FSL format)dwi_denoise_unring_sphmean.gdur  --> gradient duration of each volume of dwi_denoise_unring_sphmean.nii, in ms (FSL format)dwi_denoise_unring_sphmean.gsep  --> gradient separation of each volume of dwi_denoise_unring_sphmean.nii, in ms (FSL format)dwi_denoise_unring_sphmean.scheme --> b-values, gradient duration and gradient separation as one scheme fileROI_masks.nii.gz                  --> location of the regions-of-interest (ROIs) used in medrxiv 2024, DOI: 10.1101/2024.07.15.24310280.                                      The ROIs have been drawn in the MRI slice from which the HE histological images where collectedmaps_analyticalmodel              --> dMRI parametric maps from fitting an analytical signal model of intra-cellular, restricted diffusion                                      within spherical cells, and hindered, extra-cellular Gaussian diffusion.                                      See medrxiv 2024, DOI: 10.1101/2024.07.15.24310280.maps_HistouSim                    --> dMRI parametric maps from the proposed Histo-microSim technique.                                       See medrxiv 2024, DOI: 10.1101/2024.07.15.24310280.                                                                            

The folder maps_analyticalmodel contains maps from fitting a two-compartment analytical signal model, accounting for restrictedintra-cellular diffusion within spherical cells of equal size, and hindered, Gaussian extra-cellular diffusion. The model has beenfitted using the pgse2sphereinex.py script, released within the BodyMRITools python package (https://github.com/fragrussu/bodymritools).The output parametric maps are:InExAnalysis_AIC.nii: Akaike information criterion (quality of fit)InExAnalysis_BIC.nii: Bayesian information criterion (quality of fit)InExAnalysis_cellsmm-2.nii: cell density per unit area, in cells/mm2InExAnalysis_cellsmm-3.nii: cell density per unit volume, in cells/mm3InExAnalysis_D0um2ms-1.nii: intrinsic intra-cellular diffusivity, in um2/msInExAnalysis_Dexinfum2ms-1.nii: extra-cellular apparent diffusion coefficient (ADCex), in um2/msInExAnalysis_exit.nii: fitting exit code (0 background, 1 success, -1 failure)InExAnalysis_fin.nii: intra-cellular signal fractionInExAnalysis_fobj.nii: fitting objective functionInExAnalysis_logL.nii: fitting log-likelihoodInExAnalysis_Lum.nii: cell diameter in umInExAnalysis_S0.nii: b = 0 signal level estimate

The folder maps_HistouSim contains maps from fitting the proposed Histo-microSim signal model, which has been developed by learninga forward model from synthetic diffusion MRI signals form virtual cancer environments reconstructed directed from histology. Theenvironments used to build Histo-microSim are already available online at the following, permanent address:Grussu, F., Grigoriou, A., Macarro, C., & Perez-Lopez, R. (2024). "Histology-informed microstructural diffusion simulations for MRI cancer characterisation (Histo-µSim): histology substrates"(1.1.0) [Data set]. Zenodo. https://doi.org/10.5281/zenodo.14559104. 

Histo-microSim has been fitted using the mri2micro_dictml.py script, released within the BodyMRITools python package (https://github.com/fragrussu/bodymritools). The maps contained in the maps_HistouSim folder are:InExAnalysis_AIC.nii: Akaike information criterion (quality of fit)InExAnalysis_BIC.nii: Bayesian information criterion (quality of fit)InExAnalysis_exit.nii: fitting exit code (0 background, 1 success, -1 failure)InExAnalysis_fobj.nii: fitting objective functionInExAnalysis_logL.nii: fitting log-likelihoodInExAnalysis_par1.nii: intra-cellular signal fractionInExAnalysis_par2.nii: mean cell diameter in umInExAnalysis_par3.nii: variance of cell diameters, in um2InExAnalysis_par4.nii: skewness of the cell diameter distributionInExAnalysis_par5.nii: intrinsic intra-cellular diffusivity, in um2/ms InExAnalysis_par6.nii: intrinsic extra-cellular diffusivity, in um2/msAcknowledgementsVHIO would like to acknowledge: the State Agency for Research (Agencia Estatal de Investigacion) for the financial support as a Center of Excellence Severo Ochoa (CEX2020-001024-S / AEI / 10.13039 / 501100011033), the Cellex Foundation for providing research facilities and equipment and the CERCA Programme from the Generalitat de Catalunya for their support on this research. This study has been funded by Instituto de Salud Carlos III (ISCIII) through the project "PI21/01019" and co-funded by the European Union. Part of the data acquisitionhas been supported by PREdICT, sponsored by AstraZeneca. R.P.L is supported by the "la Caixa" Foundation CaixaResearch Advanced Oncology Research Program, the Prostate Cancer Foundation (18YOUN19), a CRIS Foundation Talent Award (TALENT19-05), the FERO Foundation through the XVIII Fero Fellowship for Oncological Research, the Instituto de Salud Carlos III-Investigacion en Salud (PI18/01395 and PI21/01019), the Asociacion Espanola Contra el Cancer (AECC) (PRYCO211023SERR) and the Generalitat de Catalunya Agency for Management of University and Research Grants of Catalonia (AGAUR) (2023PROD00178). The project that gave rise to these results received the support of a fellowship from "la Caixa" Foundation (ID 100010434). The fellowship code is "LCF/BQ/PR22/11920010" (funding F.G.). A.G. is supported by a Severo Ochoa PhD fellowship (PRE2022-102586). C.M. is funded by the Asociacion Espanola Contra el Cancer (AECC) (PRYCO211023SERR).

Copyright noticeCopyright (c) 2024, 2025. Fundació Privada Institut d’Investigació Oncològica de Vall d’Hebron. All rights reserved. Copyright (c) 2024, 2025. Vall d'Hebron Institute of Oncology (VHIO). Cellex Center, Carrer de Natzaret 115-117, 08035, Barcelona, Spain. Tel: [phone]. email: [e-mail]. Data released under a Attribution-NonCommercial-ShareAlike 4.0 International License (CC BY-NC-SA 4.0).