Magnetic resonance imaging of melanoma metastases in a clinical relevant human melanoma xenograft scid mouse model

Magnetic resonance imaging of metastases in xenograft mouse models of cancer

Magnetic resonance imaging (MRI) of small animals has emerged as a valuable tool to noninvasively monitor tumor growth in mouse models of cancer. However, imaging of metastases in mouse models is difficult due to the need for high spatial resolution. We have demonstrated MRI of metastases in the liver, brain, adrenal glands, and lymph nodes in different xenograft mouse models of cancer. MRI of mice was performed with a clinical 3.0 T magnetic resonance scanner and a commercially available small-animal receiver coil. The imaging protocol consisted of T1- and T2-weighted fat-saturated spin echo sequences with a spatial resolution of 200 μm × 200 μm × 500 μm. Total acquisition time was 30 min per mouse. The technique allowed for repetitive examinations of larger animal cohorts to observe the development of metastases.

Characterization and longitudinal monitoring of melanoma growth in ret-transgenic mice using a single-sequence MRI protocol

Spontaneous melanoma models in transgenic mice are increasingly used in preclinical research as they most closely match the progression of melanoma in humans. While optical inspection only allows analysis of tumors located on the skin, the accurate measurement and growth of subcutaneous tumors have not been adequately assessed. To improve the measurement accuracy of melanoma tumors, we used a fast single-sequence MRI protocol at 9.4 Tesla for longitudinal characterization of a ret-transgenic mouse model. Repeated MRI (average acquisition time 30 min per animal) of the trunk (excluding head and distal limbs) in six siblings revealed an increase in the mean total tumor volume (TTV) from 102.0 ± 80.5 mm(3) at 35 days of age to 434.8 ± 154.9 mm(3) by 77 days. The main tumor load was located within the pelvis (>40%), followed by the proximal hind limbs and groins (>30%). The smallest detectable tumor measured 0.07 mm(3). Inter-rater reliability between a radiologist and a veterinarian analysing MRI data was 0.993 for TTV and 0.840 for number of tumors (both p < 0.001). We thus conclude that because of the high variance of TTV of same-aged mice, MRI should be used (i) to establish treatment groups matched for TTV and (ii) for longitudinal examination of the TTV in mice over the course of treatments.

Selectin-deficiency reduces the number of spontaneous metastases in a xenograft model of human breast cancer

Metastasis formation is a complex process still poorly understood. Previous work in a colon cancer xenograft model showed that E(ndothelial) and P(latelet) selectins mediate spontaneous metastasis to the lungs. To investigate the functional role of selectins in breast cancer, human DU4475 breast cancer cells were injected subcutaneously into pfp-/-rag2-/- mice and in all their selectin-deficient variants (EP-/-, E-/- and P-/-). Pfp-/-rag2-/- mice as well as all their selectin-deficient variants developed primary tumours and spontaneous metastases. Compared with the wild-type mice, disseminated tumours cells were significantly lower (74% reduction, P=0.046) in the bone marrow of selectin-deficient mice. Pfp-/-rag2-/- mice developed significantly higher numbers of lung metastases (6644.83±741.77) than the E-/- (4053.33±112.58; P=0.002) and the EP-/- pfp-/-rag2-/- mice (4665.65±754.50; P<0.001). The results indicate that E- and P-selectins play a role in spontaneous metastasis formation both into bone marrow and lungs. However, spontaneous metastasis was not completely abrogated, hence additional cell adhesion molecules must be involved in the metastatic spread.