Deep Representation Learning of Tissue Metabolome and Computed Tomography Images Annotates Non-invasive Classification and Prognosis Prediction of NSCLC

TL;DR

This study introduces a deep learning framework that integrates tissue metabolomics and CT imaging to non-invasively classify NSCLC histology and predict patient prognosis, outperforming traditional radiomics and clinical models.

Contribution

The paper presents a novel deep learning approach that embeds metabolomic information into CT images, enabling non-invasive histology classification and prognosis prediction in NSCLC.

Findings

Achieved F1-score of 0.78 for histology classification.

Attained a c-index of 0.72 for prognosis prediction.

Surpassed radiomics and clinical models in performance.

Abstract

The rich chemical information from tissue metabolomics provides a powerful means to elaborate tissue physiology or tumor characteristics at cellular and tumor microenvironment levels. However, the process of obtaining such information requires invasive biopsies, is costly, and can delay clinical patient management. Conversely, computed tomography (CT) is a clinical standard of care but does not intuitively harbor histological or prognostic information. Furthermore, the ability to embed metabolome information into CT to subsequently use the learned representation for classification or prognosis has yet to be described. This study develops a deep learning-based framework -- tissue-metabolomic-radiomic-CT (TMR-CT) by combining 48 paired CT images and tumor/normal tissue metabolite intensities to generate ten image embeddings to infer metabolite-derived representation from CT alone. In clinical NSCLC settings, we ascertain whether TMR-CT achieves state-of-the-art results in solving histology classification/prognosis tasks in an unseen international CT dataset of 742 patients. TMR-CT non-invasively determines histological classes - adenocarcinoma/ squamous cell carcinoma with an F1-score=0.78 and further asserts patients' prognosis with a c-index=0.72, surpassing the performance of radiomics models and clinical features. Additionally, our work shows the potential to generate informative biology-inspired CT-led features to explore connections between hard-to-obtain tissue metabolic profiles and routine lesion-derived image data.