We studied the effect of pre-existing inflammation on the genesis of brain metastases using MRI. Molecular MRI enhanced with VCAM-1 targeted microparticles of iron oxide enable the assessment of pre-existing vascular inflammation and then of metastasis implantation. The occurrence of brain metastases was increased by 2 fold when inflammation was present; this could be blocked with VCAM-1 antibody. This underlies the key role of VCAM-1 in tumor seeding to the brain, and suggests that preventive inflammation targeted treatment in cancer patients may minimize risks of brain metastasis.
Imaging VCAM-1. Inflammation was induced in Balb/c mouse using stereotaxic lipopolysaccharide (lipopolysaccharide (LPS), 1 µg in 1 µl saline, n = 3) injection into the right hemisphere (RH) and evaluated 24 h later using the distribution of VCAM-1 as a marker of vascular inflammation. Mice injected with saline or sham-operated animals were used as controls. VCAM-1 distribution was visualized and quantified using molecular magnetic resonance imaging (MRI) with VCAM-1 antibody conjugated microparticles of iron oxide (MPIOs) as first described by McAteer et al3.
Imaging metastases. VCAM-1 targeted MPIOs can also be used to detect metastases with high sensitivity, because VCAM-1 is expressed on vessels associated with metastases and by tumor cells4. Hence, to study the impact of pre-existing inflammation on the distribution of tumor cells, mice injected with LPS (or controls, as above) were injected intracardially with 4T1 breast cancer cells (105 cells in 100 µl phosphate buffered saline) 24 h later. Metastasis imaging was performed on days 10 and 18 post cell injection.
Exploring the role of VCAM-1 on metastases formation. An extra group of mice received an intravenous injection of VCAM-1-MPIOs 24h after intracerebral LPS-injection, putatively blocking VCAM-1 binding sites, followed by 4T1 cell injection 4 hours later. In this group MRI was performed 3.5 h after the injection of VCAM-1-MPIO (to detect inflammation), as well as on day 18 post tumor cell injection (to detect metastases). All MR experiments were conducted on a small animal 7T scanner (Varian Inc., Palo Alto, CA) with a dedicated mouse head-coil (RAPID MR International, OH) using a T2*-weighted sequence (TR = 50 ms, TE = 25 ms, flip angle = 15°, data matrix = 256x256x96, field of view = 20 x 20 x 10 mm3, 2 averages). Brains were extracted following the final imaging session for histological analysis.
Figure 1A - Inflammation detection - Representative T2* weighted axial image of VCAM-1 in a mouse injected with VCAM-1-MPIOs 24h post LPS injection in the RH. Specific retention of the contrast agent on activated vasculature is visible as signal voids (white arrow) in the inflamed hemisphere but was hardly detected in the contralateral hemisphere. Figure 1B - Metastasis detection - Representative T2* weighted axial image of brain metastases on day 18 post cancer cells injection in an animal treated on day 0 with LPS. MR detection of brain metastases is confirmed by histological analysis (Fig. 1C). Metastases are detected through VCAM-1 expression in vessels associated with metastases as shown by VCAM-1 immunostaining (Fig. 1D, black arrow: VCAM-1-MPIOs bound to an activated vessel inside a tumor in Fig. 1C).
The number of brain metastases was markedly increased (2 fold higher) in the inflamed hemisphere (Fig. 2A) compared to the contralateral hemisphere and to control conditions (Fig 2B and C). The distribution of metastases co-localize with inflammation (Fig. 1A, B, C white arrow). Evidence that VCAM-1 contributes to tumor cell adhesion was demonstrated by functionally blocking VCAM-1 with anti-VCAM-1 MPIOs pre-treatment in LPS treated animals. This blocking reduces the number of metastases in the inflamed hemisphere back to control values (Fig. 2D). MRI results were confirmed by tumor quantification on histological sections (data not shown).