Supplementary materials for article titled: “Tumor Volume Doubling Time of Less Than One Year is Associated with a Higher Risk of Death from Medullary Thyroid Cancer.”
Description
This is supplementary figures and tables for my research article titled: “Tumor Volume Doubling Time of Less Than One Year is Associated with a Higher Risk of Death from Medullary Thyroid Cancer.” Abstract Context Tumor volume doubling time (TVDT) is emerging as a useful tool in predicting oncologic outcomes. There is limited data on the prognostic role of TVDT in metastatic medullary thyroid cancer (MTC). Purpose The goal of this study was to assess the value of TVDT in predicting disease-specific survival (DSS) in patients with hereditary and sporadic MTC. Methods This was an Institutional Review Board-approved cohort study including patients with metastatic MTC having at least 3 consecutive imaging studies. TVDT of up to the five largest lesions per organ was calculated using a standardized formula. The association between TVDT and DSS was analyzed using Kaplan-Meier survival curves. Cox proportional regression model was used to account for confounding factors. Results The study sample consisted of 51 patients presenting with 286 metastatic lesions measured with 457 scans during the follow-up of 51 (IQR 25-102) months. Median age was 19 years (IQR 15-41), 53% female patients. Cumulative volumes of all metastatic lesions and proportion of patients with TVDT of <1 year were higher in patients with sporadic as compared with hereditary MTC (p<0.01). Factors independently associated with shorter DSS were TVDT of <1 year based on 3 initial and 3 last scans as well as lung, brain and prostate as the organs with the fastest growing tumor. TVDT based on 2-dimentional and 3-dimentional measurements showed strong correlation (r=0.94, p<0.05). Conclusions Three baseline and three most recent scans preceding follow-up visit enable calculation of TVDT and can be used as predictors of mortality from MTC.
Files
Steps to reproduce
DT measurements A total of 457 CT/MRI scans were studied. MRI of 8mm slice thickness was used for liver, prostatic and brain lesions, with contrast study in axial ± coronal images using different pulse sequences to ensure accuracy. CT of 5mm thickness with IV contrast was used to measure the neck lesions, pulmonary nodules, and lymph nodes. Two independent radiologists provided the reading of the disease burden (one - on clinical grounds, and the second one for the study purposes). A single reader (NB), a radiologist with more than 25-year experience conducted the measurements enabling the calculations of tumor burden change over time. The lesions (up to 5 largest target lesions per organ) were measured manually in 2 dimensions; length and width and in 3 dimensions; length, width, and thickness for each lesion on the same section over time (at least 3 consecutive scans performed at least three months apart) archiving and tracking the data in the Picture Archiving and Communication System (PACS). Tumor volume based on 2D measurements was calculated using ellipsoid formula of π/6 x largest x smallest x smallest dimensions, while based on 3D measurements was established using π/6 x largest x smallest x depth equation. Survival analyses were performed taking into account: (1) TVDT of all analyzed lesions in all organs; (2) TVDT of all lesions in the organ with the fastest growing tumor; (3) individual lesions with the shortest doubling time (fastest growing). Biochemical studies Calcitonin (Ct-DT) and CEA doubling times (CEA-DT) were assessed during the same study period as the CT/MRI scans per the standard American Thyroid Association recommendations.