Abstract
An unmet clinical need in solid tumor cancers is the ability to harness the intrinsic spatial information in primary tumors that can be exploited to optimize prognostics, diagnostics and therapeutic strategies for precision medicine. We have developed a transformational spatial analytics (SpAn) computational and systems biology platform that predicts clinical outcomes and captures emergent spatial biology that can potentially inform therapeutic strategies. Here we apply SpAn to primary tumor tissue samples from a cohort of 432 chemo-naïve colorectal cancer (CRC) patients iteratively labeled with a highly multiplexed (hyperplexed) panel of fifty-five fluorescently tagged antibodies. SpAn predicted the 5-year risk of CRC recurrence with a mean area under the ROC curve of 88.5% (SE of 0.1%), significantly better than current state-of-the-art methods. SpAn also inferred the emergent network biology of the tumor spatial domains revealing a synergistic role of known features from CRC consensus molecular subtypes that will enhance precision medicine.
Footnotes
Revised manuscript includes new figures and tables (Figs. 1, 5d, and S7; Table S3). The figure numbers have been accordingly modified. Certain existing figures and tables (Figs. 3 and 4; Tables S1 and S2) have been revised for clarity. Text in the main manuscript and the Methods section has been updated to better describe the figures and the tables. New subsections have been added to the Methods section to improve method description and clarify associated details.