Integrative Analysis of Histopathological Images and Genomic Data Predicts Clear Cell Renal Cell Carcinoma Prognosis

Abstract

In cancer, both histopathologic images and genomic signatures are used for diagnosis, prognosis, and subtyping. However, combining histopathologic images with genomic data for predicting prognosis, as well as the relationships between them, has rarely been explored. In this study, we present an integrative genomics framework for constructing a prognostic model for clear cell renal cell carcinoma. We used patient data from The Cancer Genome Atlas (n = 410), extracting hundreds of cellular morphologic features from digitized whole-slide images and eigengenes from functional genomics data to predict patient outcome. The risk index generated by our model correlated strongly with survival, outperforming predictions based on considering morphologic features or eigengenes separately. The predicted risk index also effectively stratified patients in early-stage (stage I and stage II) tumors, whereas no significant survival difference was observed using staging alone. The prognostic value of our model was independent of other known clinical and molecular prognostic factors for patients with clear cell renal cell carcinoma. Overall, this workflow and the shared software code provide building blocks for applying similar approaches in other cancers. Cancer Res; 77(21); e91-100. ©2017 AACR.

Figure 1.

Data analysis and integration workflow. (A) Cellular morphological feature extraction pipeline. (B) Schematic diagram for gene co-expression analysis and summarization. (C) Integrative analysis of image features with eigengenes. Univariate survival analysis is used for an initial selection of survival-associated variables, and then these variables are used to train a lasso-Cox prognostic model. Correlation between image features and eigengenes is also explored.

Figure 2.

Image features and eigengenes predict the survival outcomes of ccRCC patients. Both image features (A and B) and eigengenes (C) identify poor-prognosis subtypes with high percentage of stroma. Gene module 2 is enriched with extracellular matrix genes. RMean_bin10 (D) and eigengene3 (E) are the most significant variables for image features and eigengenes, respectively. Integrative analysis of histopathological images and genomic data using lasso-Cox can significantly improve the prognosis prediction power (F).

Figure 3.

Pairwise correlation heat map between 33 survival-associated image features and all 15 eigengenes, using Spearman rank correlation.

Figure 4.

Image features and eigengenes predict the survival outcomes in early-stage (stage I and II) ccRCC patients. Stage is strongly associated with survival (A) but cannot stratify early-stage patients (B). However, image features (C, D), eigengenes (E), and lasso-Cox model (F) are significantly related to survival in early-stage patients.