In this study we examined the tumour growth kinetics of SPC-raf transgenic mice by serial micro-CT examinations. Small animal imaging allows assessment using each animal as its own control in follow-up examinations. Given the relevant inter-individual spread it has the potential to optimize studies. To prevent intra-individual spread sophisticated imaging and post-processing techniques have to be established as elaborated below. An advantage of imaging especially in diffuse or multifocal pathologies is that the entire volume can be assessed additional to circumscribed areas of sectional histopathology obtained.
Very few studies on follow-up micro-CT examination have been performed in transgenic murine models of lung cancer (mainly K-ras transgenic) [12–14]. Other groups performed follow-up examination in single lesions caused by intrapulmonary injection of tumour cells or several/multiple lesions initiated by intraperitoneal injection of urethane [15–17]. To the best of our knowledge, no report on micro-CT assessment of tumour growth kinetics in the SPC-raf transgenic lung tumour mouse model has been published so far. Furthermore, the follow-up exams reported did usually include only a limited number of imaging time points as compared to up to 15 time points in this study, allowing a more detailed assessment of growth kinetics. Further studies have shown the use of micro-CT for the detection of primary lung tumours or pulmonary metastases without a follow-up being performed [7, 18].
All the various imaging approaches of murine animal models of human lung tumour have different advantages and disadvantages. Some animal models are highly artificial with regard to the mechanism of carcinogenesis or the immunological competence of the tumour bearing organism. But they may provide imaging characteristics similar to the situation in humans with a single comparatively well-defined lesion. The multifocal tumour growth in the SPC-raf transgenic animal model examined in this study limits direct application of established radiological imaging techniques in humans. However, it has already been reported that this animal model allows examination of the potential relevance of human protooncogenes or disabled tumour suppressor genes in an immunologically competent environment. Thus, aspects of human carcinogenesis may be better understood highlighting the clinically translational aspects of the research.
In the animals examined in this study tumour growth seemed to occur at a later point of time in male animals as compared to female animals, furthermore females showed clinical signs of tumour necessitating to sacrifice the animals earlier compared to male animals (Table1). This has not been reported for the SPC-raf transgenic animal model, yet. However, due to the genetic mechanism of tumour induction in this model it might represent a relevant finding to understand lung tumour carcinogenesis. Further experiments have to be performed in order to validate the potential finding and present a hypothesis for the origin. The primary focus of this study was to demonstrate the use of micro-CT for assessment of tumour load and growth. The issue demonstrates the potential additional benefit of the method for assessment of cofactors affecting carcinogenesis applying intraindividual time-course assessment.
Micro-CT imaging applying the setup described above did not result in adverse events, even though animals had advanced tumour stages at the later time points. In synopsis with other studies performed we attribute this to the use of isoflurane inhalation anaesthesia, respiratory monitoring and the use of a heating blanket. We performed prospective respiratory gating as it has been reported to significantly increase image quality. However, more sophisticated gating techniques such as retrospective or intrinsic gating or high-speed single-breath hold techniques could further optimize imaging [19–22]. MRI has been reported to allow high spatial resolution imaging of the lung. Martiniova et al. even reported better detection of small lesions with MRI as compared to micro-CT . Optical imaging techniques or micro-PET enable assessment of functional parameters but have limitations in high resolution imaging of morphology [8, 23]. Various post-processing strategies have been reported allowing precise evaluation of specific aspects of the image data.
Dose measurements for the applied micro-CT protocol were performed in a phantom and ex-vivo in previous studies. The effective dose calculated from these measurements was 154 mGy for the selected protocol. An additional radiation dose has to be added for planning the scan-area under fluoroscopy. For this a dose of 19 mGy/min was measured, resulting in 202 mGy/scan . Animals received between 4 and 15 repetitive exams with 4 weeks interscan interval (MV = 13.0, SD = 3.05). The calculated accumulative dose ranged from 808 mGy within 91 days (4 exams) to 3030 mGy within 475 d (15 exams). The mean calculated accumulative dose was 2626 mGy within approximately 450 d. These dose values in synopsis with a reported LD50/30 (dose that is lethal in 50% of the animals within 30 days) of 7.52 Gy demonstrate the relevance of the issue . However, we consider direct adverse effects (structural changes to the lungs or unintended radiation effects on the tumour growth) to be unlikely. Although gene expression changes have been seen in cell cultures with doses as low as 20-500 mGy  structural changes like fibrosis were not even seen following doses as high as 7-9 Gy  and the reported values for therapeutic radiation also amounted to values as high as 15.5 Gy .
In conclusion the presented region-growing segmentation algorithm allows longitudinal in-vivo quantification of multifocal lung adenocarcinoma in SPC-raf transgenic mice. This enables the assessment of tumor load and growth kinetics for the study of carcinogenesis and the evaluation of novel treatment strategies.