Contrast-enhanced small-animal PET/CT in cancer research: strong improvement of diagnostic accuracy without significant alteration of quantitative accuracy and NEMA NU 4–2008 image quality parameters

SA-PET/CT examinations were performed on an Inveon SA-PET/CT (Siemens Medical Solutions, Knoxville, TN, USA). First, SA-CT images were acquired in approximately 10 min using 80 keV and 500 μA. Then PET acquisitions were performed using energy and coincidence timing windows of 350 to 650 keV and 3.4 ns, respectively. Emission scan duration was 20 min for phantom acquisitions, as recommended by the NEMA NU 4–2008 standards, and 15 min for animal scans. When the animal was imaged more than once, the duration of the subsequent acquisition was increased to account for ¹⁸F decay.

In the first experiment, immediately after the PET examination, tumors and organs were harvested. Tissue samples were weighed with a precision scale (±0.01 mg). Tumor radioactivity was counted for 2 min in a cylinder-well counter (Cobra II, Packard, GMI, Inc., MN, USA) and corrected for instrument efficiency and decay. Counts per minute were converted to becquerel and normalized for sample weight, assuming a density of 1 g/mL.

The SA-PET/CT system was calibrated according to the manufacturer's guidelines, by imaging a ⁶⁸Ge cylinder phantom. A cross calibration among the SA-PET/CT system, the dose calibrator, and the gamma counter was performed. A 10 MBq ¹⁸F-FDG solution (as assessed by the dose calibrator) was used to fill a vial with an exact known volume, which resulted in a solution with an exactly known concentration. This solution was used to fill a cylinder phantom that was scanned for 20 min on the SA-PET/CT scanner. Three samples of this solution (0.5 mL) were also drawn up with a calibrated pipette to be counted in the gamma counter. A large volume of interest (VOI) was used to determine the mean activity concentration as assessed by the SA-PET/CT scanner. Cross-calibration factors were then derived and used to synchronize counts/measurements for the three pieces of equipment.

The relationship between the radioactivity in animal tumors as determined by SA-PET/CT and by gamma counter was estimated using linear regression analysis. In addition to regression analysis, Bland-Altman plots were produced [23]. The same type of analysis was used to compare quantitative values extracted from _UEPET/CT and _CEPET/CT. Areas under the receiver operating characteristic curve for _CEPET/CT versus _UEPET/CT were compared according to the methodology of DeLong et al. [22].

Figures 5 and 6 illustrate how the intravenous injection of a mixture of ¹⁸F-FDG and iodinated contrast agent combined with intraperitoneal iohexol injection immediately before the start of the SA-PET/CT acquisition improved visual delineation of the intraperitoneal tumor from the surrounding organs in rats and mice. In our experience, this was of particular value in the evaluation of tumors with a heterogeneous tracer uptake that could easily be mistaken for physiological uptake. In addition, this technique facilitates and improves the visual evaluation of the intra-abdominal organs. The kidneys were well visualized because of the urinary elimination of the intravenous iodinated contrast agent and because of an early peritoneal absorption after intraperitoneal injection, leading to the presence of contrast media in the renal collecting systems and bladder. Therefore, the bladder was very well defined both on its inner side by the intravesical accumulation of contrast media eliminated in the urine and on its outer side by the remaining intraperitoneal iodinated agent, allowing for excellent delineation of the bladder wall. The injection of long-lasting contrast agents was also useful in providing additional anatomical landmarks through liver and spleen enhancement. In addition, the contrast-enhanced CT images showed small volume tumoral deposits in one rat and one mouse that were not visualized on the PET images, as shown in Figure 7.

Receiver operating characteristic (ROC) analyses were performed to objectively compare the results of _UEPET/CT and _CEPET/CT data interpretation according to a 5-point scale (Figure 8). A total of 69 ¹⁸F-FDG foci were indentified during the review process. Based on the results of the necropsy, 51 abdominal foci corresponded to tumors, while 18 of them were located in animals or regions of the abdomen that were free of disease. For the senior observer, the area under the receiver operating characteristic curve was 0.77 (95% CI 0.65 to 0.86) for _UEPET/CT and 0.96 (95% CI 0.89 to 0.99) for _CEPET/CT (P = 0.004). For the junior observer, the area under the receiver operating characteristic curve was 0.58 (95% CI 0.45 to 0.70) for _UEPET/CT and 0.78 (95% CI 0.66 to 0.87) for _CEPET/CT (P = 0.004). When selecting the third level of the scale as the diagnostic threshold, sensitivity and specificity for the senior observer were respectively 65% (95% CI 40 to 78) and 83% (95% CI 59 to 96) for _UEPET/CT, and 78% (95% CI 63 to 88) and 94% (95% CI 73 to 100) for _CEPET/CT. Sensitivity and specificity for the junior observer with the same level of scale were respectively 34% (95% CI 21 to 41) and 67% (95% CI 41 to 87) for _UEPET/CT, and 71% (95% CI 57 to 83) and 72% (95% CI 47 to 90) for _CEPET/CT.

Phantom acquisitions involving contrast media were performed at an iodine concentration of 100 mg I/mL, as this concentration results in CT densities much higher than those obtained with long-lasting contrast agents and is thus more likely to alter the attenuation map derived from segmented CT images. NEMA NU 4–2008 image quality phantom data showed that changes in %SD_unif and SORs when attenuation or attenuation plus scatter corrections were applied were slightly different when the NEMA NU 4–2008 image quality phantom was filled either with ¹⁸F-FDG or with ¹⁸F-FDG and contrast media. This finding, which was confirmed by imaging the NEMA NU 4–2008 image quality phantom twice for each situation, demonstrates that the use of contrast material minimally affected the accuracy of attenuation and scatter corrections. Noticeably, despite slight changes due to the use of iodinated contrast media, observed %SD_unif and SORs were in the range of those reported by other labs running the Inveon system [17, 24] and can be compared to NEMA NU 4–2008 evaluation of other modern SA-PET systems that have been recently described in details by Goertzen et al. [25]. The homemade phantom acquisitions demonstrated that, on average, a 10.3% overestimation could be expected when imaging a ¹⁸F-FDG source that mimics a pertinent-sized tumor surrounded by contrast media, whereas measured activity was almost equal to true activity when the source was surrounded by water. This finding was confirmed by animal experiments comparing _CEPET/CT and _UEPET/CT quantitative data, for which Bland-Altman analysis showed that the overestimation in a large cohort of 55 abdominal lesions never exceeded 10%. Furthermore, our results are in the range of those observed with clinical PET/CT systems where a 20 ± 1.8% overestimation was reported when filling a NEMA phantom with a contrast concentration usually used in clinical settings (3% solution of Gastrografin™ Bristol-Myers Squibb, Princeton, NJ, USA; 370 mg I/mL) [26]. Regarding clinical studies, oral contrast media have been shown to induce an average error of 4.4 ± 2.8% (max CT value of 520 HU) [27] and stasis of intravenous contrast media in the subclavian vein has been reported to increase maximum standardized uptake value (SUVmax) by 27.1% [14]. Of note are the similar ratios between PET quantitative data and ex vivo counting that we observed for the different types of intravenous contrast media. This shows that the overestimation is driven by the intraperitoneal fluid (100 mg I/mL) and that intravenous contrast media can be used without concern on quantitative accuracy.

As shown in Figures 5 and 6, intraperitoneal injection of an iodinated contrast agent dramatically improved delineation of an intraperitoneal tumor from the surrounding organs and detection of carcinomatosis involving the abdominal wall or the diaphragm. It is noteworthy that in one mouse and one rat, contrast media also enabled detection of small-volume carcinomatosis on CT that would have been overlooked by PET-only images, as shown in Figure 7. However, this finding was rare and did not discriminate between the presence or absence of carcinomatosis in both animals, as ¹⁸F-FDG-positive carcinomatosis was also present. The intravenous injection of long-lasting contrast agents was also useful in providing additional anatomical landmarks, thanks to liver and spleen enhancement. This better delineation of tumors was particularly useful in the case of lesions harboring heterogeneous tracer uptake, as shown in Figure 6 in a mouse with a necrotic tumor near the bladder. A new insight from this study is that the fast absorption through the peritoneum leads to the presence of contrast media in the renal collecting systems and in the bladder at higher densities than in the remaining intraperitoneal contrast media, allowing for a clear visualization of the bladder wall. This finding offers new opportunities for imaging bladder cancer models.

An objective assessment of _CEPET/CT versus _UEPET/CT images was performed by means of receiver operating characteristic analysis, which showed that _CEPET/CT performs better than _UEPET/CT for lesion detection within the abdomen, although in some cases, lesions were correctly identified on _UEPET/CT images (Figure 8). It is noteworthy that the comparison between _CEPET/CT and _UEPET/CT images was made in the group of animals that received only intraperitoneal contrast media; because animals had to be killed after the tracer uptake period to stop the uptake that could have occurred between the two SA-PET examinations, intravenous injection of long-lasting contrast medium was not feasible. Given that liver and spleen enhancement provided by these intravenous contrast media results in useful additional anatomical landmarks, one can assume that the difference in the area under the curve for receiver operating characteristic analysis would have been even more important if both intraperitoneal and intravenous contrast media had been used.

As shown by receiver operating characteristic analysis, intraperitoneal contrast enhancement with clinical contrast media (iohexol), easily available at low cost, significantly improved diagnostic accuracy and could therefore be used alone. The additional application of long-lasting contrast agents provides further valuable anatomical landmarks but involves a larger volume for intravenous injections and supplemental costs. The choice between intraperitoneal contrast media or a dual-contrast media protocol including intraperitoneal and intravenous contrast agents will be a compromise among cost, potential difficulties for intravenous injection, morbidity, and image quality. Regarding morbidity, to the best of our knowledge, no study has reported side effects due to an intravenous bolus with long-lasting contrast agents in mice or rats.

SA-PET and SA-PET/CT are being used increasingly in cancer research, with a major interest in the ability to monitor repeatedly and non-invasively the effect of chemotherapy or molecularly targeted therapies. As opposed to clinical PET images that can be interpreted both visually and semi-quantitatively, SA-PET images, when used for therapy monitoring, are interpreted only by means of changes in semi-quantitative values like SUVs or percentage injected dose per gram of tissue (%ID/g). The overestimation of quantitative values due to contrast media, which never exceeded 10% in our study, could be a source of inter-animal variability, previously demonstrated to be 15% on average in mice-bearing tumors imaged twice on the same day, after injection and reinjection of ¹⁸F-FDG [28], ¹⁸F-FLT [29], or ¹⁸F-labeled RGD [30]. Another implication of the use of contrast media is the possible interaction between intraperitoneal contrast media and chemotherapy, molecularly targeted therapies, or vehicle administered intraperitoneally the day of the PET examination. Although the peritoneum is known to allow fast absorption of fluids, further studies are required to investigate the delay required to reach full absorption of intraperitoneal contrast media, as well as potential renal toxicity of repeated intraperitoneal injections, having in mind that some antineoplastic treatments induce kidney damage. The morbidity that could arise from long-lasting agents and that relates to viscosity, volume, and iodine content will also be the topic of a future work.