Progenics presents Phase 2 prostate cancer imaging analysis data

Progenics Pharmaceuticals reported data demonstrating the utility of its imaging analysis technology, which uses artificial intelligence and machine learning to quantify and automate the reading of PSMA targeted imaging. The data were presented in an oral presentation at the 2018 Society of Nuclear Medicine and Molecular Imaging Annual Meeting on June 23rd in Philadelphia, Pennsylvania. In the presentation, titled “Automated Detection and Quantification of Prostatic PSMA Uptake in SPECT/CT using a Deep Learning Model for Segmentation of Pelvic Anatomy,” researchers described the validation of a deep learning algorithm for the automatic detection and quantification of 1404 uptake from SPECT/CT images. 1404 is Progenics’ PSMA-targeted SPECT/CT imaging agent, currently in Phase 3 development. The algorithm developed by Progenics’ imaging analysis technology was validated using the data from the Company’s Phase 2 study of 1404, which included 102 high-risk prostate cancer patients who all underwent PSMA imaging prior to radical prostatectomy. The validation scans were manually quantified by measuring the maximum uptake of 1404 in a circular region of interest of the prostate where the highest uptake values were determined visually. The algorithm used volumetric segmentation to measure uptake at every voxel in the prostate and determined the maximum uptake of 1404 automatically. The Pearson correlation coefficient was used to assess the concordance between manual and automated quantification of uptake. The automated maximum uptake value was significantly correlated to the manually obtained uptake value. The algorithm was fully automated and deterministic, resulting in 100% repeatability.

https://bit.ly/2Kg1hip