Clinicopathological correlation of immune response in human cancers
Metrics: PDF 446 views | Full Text 624 views | ?
1 Department of Bioinformatics and Computational Biology, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA
Keywords: clinicopathological characteristics; immune response; immunogenicity; cancer
Received: August 01, 2019 Accepted: September 24, 2019 Published: October 08, 2019
The clinicopathologic association of tumor immune response is largely unknown. We systematically investigated this matter in human cancers.
Different cancer types exhibited distinct immune gene profiling. Four cancer types exhibited a significant and positive correlation of immune response with patient age. Significant but inconsistent correlation of immune response was observed with gender, surgical stage, and TNM stage in a small number of cancer types. In contrast, the histological grade appears to have much stronger and more consistent association with immune response as compared to the other clinicopathologic factors. Specifically, patients with high grade had significantly higher immune responses than those with low grade in 5 out of 12 analyzed cancer types. In addition, both histological and molecular classifications had a significant and strong association with tumor immune response.
t-distributed stochastic neighbor embedding was used to assess similarity of immune gene profiling in human cancers. The Mann-Whitney or Kruskal-Wallis test was, respectively, used to compare the tumor immune response in two or more groups that were stratified by patient clinicopathological characteristics, such as gender, grade, stage (including surgical and TNM stage), histology, and molecular subtypes. Spearman correlation with student’s t-test was used to examine the association of patient age with immune response. Multiple tests with the Benjamini-Hochberg correction also were performed.
Tumor grade should be taken into account in selection of patient candidates for immunotherapy. Prospective verification is needed before use of the findings for clinical practice.
All site content, except where otherwise noted, is licensed under a Creative Commons Attribution 3.0 License.