Improving prediction of prostate cancer recurrence using chemical imaging

Jin Tae Kwak, André Kajdacsy-Balla, Virgilia Macias, Michael Walsh, Saurabh Sinha, Rohit Bhargava

Research output: Contribution to journalArticlepeer-review

Abstract

Precise Outcome prediction is crucial to providing optimal cancer care across the spectrum of solid cancers. Clinically-useful tools to predict risk of adverse events (metastases, recurrence), however, remain deficient. Here, we report an approach to predict the risk of prostate cancer recurrence, at the time of initial diagnosis, using a combination of emerging chemical imaging, a diagnostic protocol that focuses simultaneously on the tumor and its microenvironment, and data analysis of frequent patterns in molecular expression. Fourier transform infrared (FT-IR) spectroscopic imaging was employed to record the structure and molecular content from tumors prostatectomy. We analyzed data from a patient cohort that is mid-grade dominant - which is the largest cohort of patients in the modern era and in whom prognostic methods are largely ineffective. Our approach outperforms the two widely used tools, Kattan nomogram and CAPRA-S score in a head-to-head comparison for predicting risk of recurrence. Importantly, the approach provides a histologic basis to the prediction that identifies chemical and morphologic features in the tumor microenvironment that is independent of conventional clinical information, opening the door to similar advances in other solid tumors.

Original languageEnglish (US)
Article number8758
JournalScientific reports
Volume5
DOIs
StatePublished - 2015

ASJC Scopus subject areas

  • General

Fingerprint Dive into the research topics of 'Improving prediction of prostate cancer recurrence using chemical imaging'. Together they form a unique fingerprint.

Cite this