The use of clinical MRI scanners to conduct preclinical research facilitates a more direct or matched comparison with clinical studies, and provides evidence supporting clinical relevance of functional MRI data/biomarkers (1). Most clinical scanners operate between 1.5 and 3T, and thus have lower SNR than preclinical systems, with reduced image quality if conventional clinical imaging coils are used. One approach for increasing SNR is to use small, dedicated receiver coils designed to fit closely to the object of interest, giving a better coupling between the object and coil, an increase in signal and consequently improved image quality.

Orthotopically propagated xenografts and transgenic mouse models of cancer, which more closely emulate clinical disease, are increasingly being used in preclinical research. Such models demand non-invasive methods to longitudinally and accurately assess the progression and in vivo treatment response of tumours that typically arise within deep-seated anatomical locations (2).

The purpose of this study was to evaluate the utility and sensitivity of anatomical and functional MRI parameters acquired from transgenic mouse models of neuroblastoma using a non-bespoke asymmetric high resolution RF coil on a clinical 3T scanner, and to cross-reference to previously published multiparametric data, with a particular focus on native T₁ and R₂*, acquired from the same transgenic models of neuroblastoma on a dedicated preclinical 7T system (3, 4).

Tumours spontaneously arising within genetically modified mouse models of high-risk neuroblastoma, including Th-ALK^F1174L/Th-MYCN amplified tumours harbouring the ALK^F1174L mutation (5), and Th-MYCN amplified tumours (6), were imaged on a Philips 3T Achieva scanner using a dedicated asymmetric high resolution 3-channel/3-animal RF coil (“Mouse Hotel”, Philips), enabling simultaneous data acquisition from up to three mice. The acquisition parameters used to quantify tumour volume and native T₁ and R₂* are shown in Table 1. Transgenic mice were imaged at both 3 and 7T within 24 hours for comparison of tumour volume determination. To assess treatment response, transgenic mice were imaged at 3T prior to and 24 hours post-treatment with either saline or 25mg/kg cyclophosphamide (CPM). Tumour volume was determined using OsiriX, and T₁ and R₂* quantified using in-house software (7, 8). Statistical significance was identified using Student’s 2-tailed t-tests with a 5% level of significance.Figure 1 shows T₂-weighted images of three mice acquired simultaneously at 3T. No significant difference was found between volumetric measurements acquired at 3T and 7T (819±153mm³ @ 3T versus 889±191mm³ @ 7T; p=0.26) and were significantly correlated (R²=0.96; p<0.0001) (Figure 2). There was no significant difference in native T₁ between the Th-ALK^F1174L/Th-MYCN and the Th-MYCN mice (1.15±0.12s and 1.1±0.07s; p=0.58). However, R₂* was significantly slower in tumours in the Th-ALK^F1174L/Th-MYCN mice when compared to the Th-MYCN cohort (27.7±3s^-1 and 49.7±4s^-1; p=0.0007) (Figure 3). Treatment with CPM elicited a significant (p=0.0015) decrease in tumour burden, measured at 3T, after 24 hours. This was associated with a significant decrease in native T₁ (p=0.0085). There was no significant change in R₂* 24 hours after treatment (Figure 4).

Use of the asymmetric high resolution 3-channel/3-animal RF coil on a clinical 3T platform yielded good quality T₂-weighted images of up to three mice simultaneously, with sufficient resolution to accurately define and quantify the volume of neuroblastomas arising within the abdomen of transgenic mice. Furthermore this approach could also be used for whole mouse body imaging for the detection of distant metastasis.

The coil arrangement also has sufficient sensitivity to acquire and quantify native tumour T₁ and R₂* in the transgenic mice. The significantly slower R₂* associated with the ALK^F1174L mutated neuroblastomas is wholly consistent with their impaired functional vascular phenotype relative to that in the Th-MYCN mice, previously established using intrinsic susceptibility MRI at 7T (3). Similarly, the significant reduction in native tumour T₁ following successful treatment with CPM, the current standard of care for children with neuroblastoma, is in agreement with similar intervention data acquired at 7T (4).

There are several methodological/logistic limitations of pursuing animal work on a clinical system in this way. Reliable and repeatable shimming over a small volume of interest poses a challenge. Furthermore, significant effort has to be made to minimise risk of cross-contamination, and the limited access to clinical scanners during normal hospital operating hours should be recognised.

Anatomical and multiparametric imaging data acquired on a clinical 3T system with the asymmetric high resolution 3-channel/3-animal RF coil can be reliably used for preclinical studies utilising more complex orthotopic and transgenic mouse models of cancer. Simultaneous data acquisition in up to three animals can significantly reduce overall scanning times, and enhance throughput for imaging-embedded preclinical trials of novel therapeutics.