19F MRI with a PFC contrast agent is emerging as an effective approach to evaluate the onset of inflammation in both acute and chronic diseases [28–34]. Our results extend these findings to the detection and evaluation of CIA, a model with a quantifiable clinical surrogate of disease severity, enabling a direct comparison of disease activity with the 19F signal. Numerous studies have co-located the perfluorocarbon reagent within macrophage at the site of inflammation, allowing 19F MRI to image a general characteristic of inflammation at the cell-function level [33, 37]. The detection of the 19F signal in diseased animals in this and other studies [28–34] and the lack of 19F signal in naïve animals indicates the specificity of this imaging approach. However, it was not clear whether the intensity of the signal could be used as an independent measure of disease activity. This study is the first to extend previous findings of the presence of inflammation to show the potential of 19F MRI to reveal the severity of inflammation. More importantly, serial 19F MRI monitoring could effectively be used to evaluate the persistence of inflammatory responses, progression of disease, and longitudinal study of the response to therapy. Limitations of the present study include the lack of methods to detect PFC within individual phagocytes at the site of inflammation histologically, and the relative insensitivity of MRI to detect very low amounts of 19F which might be present at sites of minimal but potentially relevant inflammation, leading to a false negative. We have recently developed a dual mode fluorescent version of the PFC contrast agent which will facilitate the evaluation of specific cells containing the contrast agent in future studies. The data reported here indicate the utility of PFC contrast agent with 19F MRI for monitoring the course of disease to assess the efficacy of a therapeutic.
Early in the disease process, a marked difference between individual animals was found, both in disease severity as well as in the accumulation of contrast agent. A linear relationship was observed between the amount of contrast agent at the site of inflammation and a clinical measurement of the severity of the experimental disease. In two subjects a high accumulation of contrast agent appeared in the tail (Figure 3B), with a lower level of 19F signal relative to ankle swelling measurement. Arthritis in the CIA model is typically restricted to the fore and hind limbs without axial involvement , and no clinical signs of disease were noted in the tails of any subjects. We surmised that signal in the tail could be a consequence of failure to completely deliver the contrast agent into the bloodstream, and the resulting misadministration enabled the local accumulation of the PFC emulsion at the site of injection, effectively reducing the amount systemically available to label circulating phagocytes. These results indicate that while the intensity of the 19F signal in the lesioned paws correlated with disease severity, care in administration of the contrast agent is necessary for the most reliable readout.
In the serial imaging studies, a difference in the pattern of 19F accumulation over time was found between the vehicle control and prednisolone treated cohorts. In the control cohort, 19F signal in the diseased limbs continued to accumulate upon repeated administration, whereas in the treated cohort, the signal remained stable over time, even after repeat administration of contrast agent. Histological and caliper measurements of ankle swelling point to continued infiltration of macrophage in the vehicle control cohort, consistent with 19F measurements. Histological endpoints and caliper measurements show fewer inflammatory cells and less swelling in the treated group, and a stable 19F signal. While 19F MRI did accurately reflect the abatement of macrophage infiltration to the site of inflammation (i.e., no increases in 19F were observed in treated animals), the persistence of signal after the departure of disease points to the need for future study and characterization of tissue clearance mechanisms and timescales of the 19F reagent. While a simple linear correlation between ankle swelling and 19F signals was not observed in the context of repeated administration of the contrast agent at days 22 and day 29 (data not shown), the 19F results nonetheless reflected the clinical responses, in which increases in 19F reflected disease progression and the inhibition of further 19F accumulation in animals undergoing successful therapy with a measureable clinical response. This data points to the utility of 19F imaging as a surrogate biomarker for evaluating therapeutic efficacy in RA.
While the CIA model in rats is largely restricted to the fore and hind limbs, and can be clinically assessed by measuring changes in ankle size, not all inflammatory diseases provide for a facile, rapid measurement of a response to a therapeutic drug . Arthritis which affects the axial skeleton, such as ankylosing spondylitis or spondyloarthropathy, does not present simple external measurements for disease severity in preclinical models  and MRI is a standard clinical practice in the diagnosis of the disease . In this case, the 19F MRI method of precisely measuring site-specific inflammation in vivo could enable an opportunity to facilitate study and treatment of disease, aiding the clinical development of therapeutics for ankylosing spondylitis and other inflammatory conditions [29, 33, 34].
As a preclinical tool, 19F MRI may have advantages over histological evaluation of tissues, given that a single, live animal may be imaged in less than one hour. In contrast, histology requires biopsy or necropsy of the particular tissues of interest, followed by fixation, preparation of frozen tissue blocks or paraffin embedding, slicing and mounting tissue sections, then staining and cover-slipping slides before the tissue is evaluated microscopically. 19F is taken up by macrophages in situ, and the signal intensity at sites of inflammation is directly related to the degree of cellular macrophage infiltration [29, 31, 33, 34], providing a rapid means of assessing inflammatory infiltration. Further, MR methods provide more comprehensive information of the extent and location of inflammation compared with selected representative tissue sections evaluated by histology for phagocytic cells, although it may not replace detailed evaluation of cell subsets or subcellular biomarkers. MRI also allows longitudinal studies in the same animal over time, without biopsy or other invasive procedures, such as synovial aspiration . Ultimately, this may speed the screening of inflammatory drugs against disease, particularly for those diseases where an external measurement on a live animal is unavailable. In the absence of imaging equipment, excised tissues may also be evaluated by 19F NMR spectrometers  for a more high throughput approach to quantitatively evaluate inflammatory lesions, with tissue potentially amenable to histology following NMR analysis.
While the goal of this study was to evaluate the imaging potential, there were several incidental findings. The detection of high amounts of 19F in the proximity of the injection site suggests that tail vein injection was less successful then one might have predicted, and that inclusion of the contrast agent could enable one to quantify misadministration. It was also noted that administration of multiple large doses of the PFC contrast agent occurred in the absence of anaphylaxis or adverse clinical effects. While more extensive preclinical toxicological safety testing are necessary prior to drawing conclusions, the results here contribute to the accumulating data regarding the safety of systemic PFC administration for imaging and other applications [45–47].