| Affiliation: | 1. Maxillofacial Surgery Unit, S. Orsola-Malpighi Hospital, Department of Biomedical and Neuromotor Sciences, (Head: Prof. Claudio Marchetti), Alma Mater Studiorum University of Bologna, Bologna, Italy;2. Department of Experimental, (Head: Prof. Emanuela Marcelli), Diagnostic and Specialty Medicine, Bioengineering Laboratory, S. Orsola-Malpighi Hospital, University of Bologna, Italy |
| Abstract: | PurposeThis study was designed as a retrospective observational study, focusing on the correlation between the preoperative CT-scan tumor volume, tumor sphericity, and the disease-related prognosis.MethodsA total of 30 consecutive patients, affected by primary oral cancer, were retrospectively identified from our oral cancer database. The preoperative images (DICOM data) for the study population were uploaded into a modular software package designed to convert patients' medical images into 3D digital models. Multislice interpolation and threshold segmentation tools were used to segment the tumor mass. This was then converted into a 3D mesh and exported in STL format, in order to calculate the corresponding volume. We applied the concept of sphericity — a measurement of how closely the shape of an object approaches that of a mathematically perfect sphere — to the segmented tumor mass.ResultsMean tumor volume was larger in patients with tumor recurrence and/or who had died than in patients who were disease free/alive. Tumor sphericity was influential on clinical outcomes. It appeared to be lower in patients who had tumor recurrence and/or who had died (0.54 ± 0.09 and 0.53 ± 0.05) than in patients who were disease free/alive (0.65 ± 0.07). This difference was statistically significant (p < 0.05). Cumulative recurrence-free survival was 86.2% for patients with a tumor volume lower than the cut-off value. Otherwise, it was 0% for those with a tumor volume higher than the cut-off value (p < 0.01; log rank test). Cumulative recurrence-free survival was 86.3% for patients with a higher sphericity index, compared with 13.6% for those with a lower sphericity index.ConclusionThe prognostic model, based on a tridimensional, CT-based characterization of the tumor size, which includes both tumor volume and tumor sphericity, uses readily available information and could be considered when formulating prognoses for patients with oral cancer. |
