Dataset Information

Maximum predictive power of the microarray-based models for clinical outcomes is limited by correlation between endpoint and gene expression profile.

ABSTRACT:

Background

Microarray data have been used for gene signature selection to predict clinical outcomes. Many studies have attempted to identify factors that affect models' performance with only little success. Fine-tuning of model parameters and optimizing each step of the modeling process often results in over-fitting problems without improving performance.

Results

We propose a quantitative measurement, termed consistency degree, to detect the correlation between disease endpoint and gene expression profile. Different endpoints were shown to have different consistency degrees to gene expression profiles. The validity of this measurement to estimate the consistency was tested with significance at a p-value less than 2.2e-16 for all of the studied endpoints. According to the consistency degree score, overall survival milestone outcome of multiple myeloma was proposed to extend from 730 days to 1561 days, which is more consistent with gene expression profile.

Conclusion

For various clinical endpoints, the maximum predictive powers of different microarray-based models are limited by the correlation between endpoint and gene expression profile of disease samples as indicated by the consistency degree score. In addition, previous defined clinical outcomes can also be reassessed and refined more coherent according to related disease gene expression profile. Our findings point to an entirely new direction for assessing the microarray-based predictive models and provide important information to gene signature based clinical applications.

SUBMITTER: Zhao C

PROVIDER: S-EPMC3287499 | biostudies-literature | 2011 Dec

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Maximum predictive power of the microarray-based models for clinical outcomes is limited by correlation between endpoint and gene expression profile.

Zhao Chen C Shi Leming L Tong Weida W Shaughnessy John D JD Oberthuer André A Pusztai Lajos L Deng Youping Y Symmans W Fraser WF Shi Tieliu T

BMC genomics 20111223

<h4>Background</h4>Microarray data have been used for gene signature selection to predict clinical outcomes. Many studies have attempted to identify factors that affect models' performance with only little success. Fine-tuning of model parameters and optimizing each step of the modeling process often results in over-fitting problems without improving performance.<h4>Results</h4>We propose a quantitative measurement, termed consistency degree, to detect the correlation between disease endpoint an ...[more]

PMID: 22369035

Similar Datasets

Project description:Background: A pressing challenge in treating non-small cell lung cancer (NSCLC) lies in the significant variability of patient responses to immunotherapy. This issue underscores the critical need for innovative predictive models that can navigate beyond conventional biomarkers to enhance patient outcomes, particularly by addressing immunotherapy resistance or responsiveness. Hypothesis: We hypothesize that an integrated multi-omics approach will uncover interactions within the NSCLC tumor immune microenvironment (TIME) and identify novel biomarkers that are predictive of individual immunotherapy responses, thus aiding in the development of a robust personalized treatment planning model. Objective: To develop a predictive model for NSCLC immunotherapy response/resistance by identifying new biomarkers using co-detection by indexing (CODEX) and Digital Spatial Profiling of whole transcriptome atlas (DSP-WTA). Methods: We utilized a multi-omics approach, combining CODEX for spatial mapping of protein expression at the single-cell level and DSP-WTA for comprehensive transcriptomic insights from the cell types identified by CODEX. This methodology facilitated a detailed examination of the TIME in NSCLC samples from patients undergoing first-line immunotherapy. Results: Our analysis identified three cell types, proliferating tumor cells, granulocytes, and vessels, that are associated with resistance to immunotherapy. The high proportion of these cell types demonstrated a hazard ratio (HR) of 3.8 (p = 0.004) in the training cohort (N = 33) and 1.8 (p = 0.05) in the validation cohort (N = 35). In the response cell type model, higher levels of M1 macrophages, M2 macrophages, and CD4 T cells returned a HR of 0.4 (p = 0.019) in the training set and 0.49 (p = 0.036) in the validation set. Gene signatures related to these cell types also predicted outcomes with high accuracy. The resistant gene model, which included 8 genes associated with epithelial-mesenchymal transition (EMT) and cell migration showed a HR of 5.3 (p < 0.001) in the training set and 2.2 (p = 0.036) in the validation set. The response gene model, consisting of 8 genes associated with immunomodulation, had an HR of 0.22 (p = 0.005) in the training set and 0.38 (p = 0.034) in the validation set. Conclusion: This research highlights the potential of a multi-omics strategy in advancing NSCLC treatment toward precision oncology. By offering insights into the TIME and unveiling novel biomarkers, our model seeks to define resistance and to improve prediction of response to treatment.

Dataset Information

Maximum predictive power of the microarray-based models for clinical outcomes is limited by correlation between endpoint and gene expression profile.

Background

Results

Conclusion

Publications

Maximum predictive power of the microarray-based models for clinical outcomes is limited by correlation between endpoint and gene expression profile.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure