INTRODUCTION

Lung cancer is the second most common cancer in men and women, and the leading cause of cancer deaths. Current standards for detection include tissue biopsy, an invasive procedure for which the detection of cancer lesions is highly dependent on the biopsy location, and chest CT scans which expose patients to potentially harmful radiation and for which soft tissue contrast is generally low, making detection of small metastatic lesions challenging. Safe and specific MRI probes are needed to enable early detection of lesion presence and therapeutic response. As a hallmark of cancer, inflammation is a key component in the function of the tumor microenvironment. Perfluorocarbon (PFC) nanoemulsion injected intravenously is taken up by the reticuloendothelial system which is comprised of mononuclear, phagocytic cells, including macrophages. PFC uptake by tumor-associated macrophages (TAMs) can therefore be used as an in vivo biomarker for lesion detection^1,2. In a metastatic lung cancer mouse model, we show that PFC is effectively taken up by TAMs and vividly displays lung metastasis using ¹⁹F MRI. Validation assays using in vivo bioluminescence (BLI) and histology support the MRI findings. Preliminary data suggests that ¹⁹F hot-spot imaging offers a highly-specific marker of tumor burden in lung parenchyma.

METHODS

SCID beige female mice (N=10) of 6-7 week age were inoculated via tail vein with 5x10⁶ A549-luc-C8 cells (PerkinElmer). The presence, location and progression of tumors was initially assessed by BLI weekly for three weeks. BLI scans were acquired using an IVIS Spectrum (PerkinElmer) in vivo imaging system 10 min post-IP injection of luciferin (150 mg/kg) using large pixel binning with F/Stop=1, and the scan time was scaled to acquire 2-3x10⁵ total counts. Seventy percent (7/10) mice developed lung metastases by three weeks post-cell injection, five of which were selected for injection with PFC nanoemulsion (~150 nm mean particle size, 30% v/v, V-Sense, Celsense) 48 hour prior to MRI. Mice were anesthetized with 1-2% isoflurane in oxygen and kept warm using a water-heated imaging bed. ¹⁹F RARE MRI scans were acquired using a Bruker 11.7T system with TR/TE=825/18 ms, RARE factor 8, 3×3 cm FOV, 32x32 matrix and 2 mm-thick axial slices covering lung and liver regions. The number of averages was scaled to acquire a 30-45 min scan. Co-registered ¹H RARE scans were acquired with TR/TE= 950/18 ms, RARE factor 8, 3×3 cm FOV, 192x192 matrix and 1 mm-thick axial and coronal slices. The number of slices was doubled to obtain the same coverage as the 19F RARE scan. At the experimental endpoint animals were sacked for histology. Images were analyzed using Living Image 4.1 for BLI and VivoQuant 3.5 (Invicro) for ¹⁹F/¹H composite image rendering.

RESULTS/DISCUSSION

The ¹⁹F MRI data show localized uptake of PFC in lung, presumably from A549-luc-C8 lung metastases; representative data is displayed in Fig. 1a. The relative location of metastases was confirmed by BLI in Fig. 1b, however, anatomical MRI scans were not able to resolve the lesions (not shown). Interestingly, in some cases, the ¹⁹F MRI shows distinct signals throughout the lungs indicating multiple lesions compared to BLI detection which shows a single tumor signal in the thoracic region. The intensity of ¹⁹F signals also correlates to BLI signal intensity, suggesting increased pro-inflammatory activity in larger tumor lesions. The results demonstrate the feasibility of PFC as a biomarker for ¹⁹F MRI detection of lung metastases. The inflammatory sensitivity of this biomarker also suggests a potential use in immuno-oncology. Therapies in these studies focus on modulating immune response to treat cancer, and thus a biomarker for tumor-related inflammation may be useful in early evaluation of treatment response. ¹⁹F MRI additionally overcomes challenges associated with ¹H MRI’s low specificity and resolution limitations due to motion artifacts when imaging the thoracic region.

Acknowledgements

This work was supported by the NIH grants R01-EB017271, R01 CA139579, R01 EB024015 and CIRM grant LA1-C12-06919.