DNA Microarrays Are Predictive of Cancer Prognosis: A Re-evaluation

@article{Fan2010DNAMA,
  title={DNA Microarrays Are Predictive of Cancer Prognosis: A Re-evaluation},
  author={Xiaohui Fan and Leming Shi and Hong Fang and Yiyu Cheng and Roger Perkins and Weida Tong},
  journal={Clinical Cancer Research},
  year={2010},
  volume={16},
  pages={629 - 636}
}
Purpose: The reliability of microarray-based cancer prognosis is questioned by Michiels et al. They reanalyzed seven studies published in the prominent journals as successful stories of microarray-based cancer prognosis and concluded that the originally reported assessments are overoptimistic. We set to investigate the reality of microarrays for predicting cancer prognosis by using the same data sets with commonly accepted data analysis approaches. Experiment Design: Michiels et al.'s analysis… 

Figures and Tables from this paper

A Simple but Highly Effective Approach to Evaluate the Prognostic Performance of Gene Expression Signatures
TLDR
The use of an arbitrary cut-off value for evaluation of signature significance is not suitable for this type of research, but should be defined for each dataset separately.
Determination of Minimum Training Sample Size for Microarray-Based Cancer Outcome Prediction–An Empirical Assessment
TLDR
External validation results based on another 3 cancer datasets confirmed that the SSNR-based approach could not only determine the minimum number of training samples efficiently, but also provide a valuable strategy for estimating the underlying performance of classifiers in advance.
Shifting from Population-wide to Personalized Cancer Prognosis with Microarrays
TLDR
Clinical confidence could serve as a beneficial metric for personalized cancer prognosis prediction utilizing microarrays using microarray-based predictive models, and separates patients with different prediction accuracies and survival times.
A Novel Computational Approach for Biomarker Detection for Gene Expression-Based Computer-Aided Diagnostic Systems for Breast Cancer
TLDR
A novel feature selection method for microarrays involving a two-step filtering process to select a minimum set of genes with greater consistency and relevance is introduced, and it is demonstrated that the selected gene set considerably enhances the diagnostic accuracy of cancer.
Gene expression based computer aided diagnostic system for breast cancer: a novel biological filter for biomarker detection
TLDR
The aim of this study was to enhance the diagnostic accuracy of Breast Cancer Computer Aided Diagnostic Systems (CADs) that use gene expression profiling of peripheral blood cells, by introducing a novel feature selection method called Bi-biological filter that was further refined by Best First Search with Support Vector Machines SVM (BFS-SVM) to select a small set of the most effective genes predictive of breast cancer.
Consistent metagenes from cancer expression profiles yield agent specific predictors of chemotherapy response
TLDR
The use of multiple data sets to derive potential biomarkers can filter out data set-specific noise and can increase the efficiency in identifying clinically accurate biomarkers.
Do DNA Microarrays Tell the Story of Gene Expression?
TLDR
In this paper, several aspects of poor reproducibility of DNA microarray technology are analyzed and strong interactions within intracellular biomolecular networks make it highly problematic to build a bridge between transcription rates of individual genes and structural fidelity of their genetic codes.
Does Applicability Domain Exist in Microarray-Based Genomic Research?
TLDR
It is found that the issue of applicability domain may not exist in microarray-based genomic research for clinical applications and it is not practicable to improve model validity based on applicabilitydomain.
Cross-Platform Comparison of Microarray-Based Multiple-Class Prediction
TLDR
It is established that it is possible to successfully apply multiple-class prediction models across different commercial microarray platforms, offering a number of important benefits such as accelerating the possible translation of biomarkers identified with microarrays to clinically-validated assays.
Independent Validation of a Prognostic Genomic Signature (ColoPrint) for Patients With Stage II Colon Cancer
TLDR
ColoPrint is able to predict the development of distant metastasis of patients with stage II colon cancer and facilitates the identification of patients who may be safely managed without chemotherapy.
...
...

References

SHOWING 1-10 OF 32 REFERENCES
Predictive ability of DNA microarrays for cancer outcomes and correlates: an empirical assessment
Converting a breast cancer microarray signature into a high-throughput diagnostic test
TLDR
This report demonstrates for the first time that microarray technology can be used as a reliable diagnostic tool and is therefore an excellent tool to predict outcome of disease in breast cancer patients.
Pitfalls in the use of DNA microarray data for diagnostic and prognostic classification.
TLDR
Statistical issues that arise from the use of DNA microarrays for an important group of objectives that has been called “class prediction” are addressed, which includes derivation of predictors of prognosis, response to therapy, or any phenotype or genotype defined independently of the gene expression profile.
Selection bias in gene extraction on the basis of microarray gene-expression data
  • C. Ambroise, G. McLachlan
  • Psychology
    Proceedings of the National Academy of Sciences of the United States of America
  • 2002
TLDR
It is demonstrated that when correction is made for the selection bias, the cross-validated error is no longer zero for a subset of only a few genes.
Thousands of samples are needed to generate a robust gene list for predicting outcome in cancer.
TLDR
A previously undescribed mathematical method is introduced, probably approximately correct (PAC) sorting, for evaluating the robustness of reported predictive gene lists, and for several published data sets the number of samples that are needed to achieve any desired level of reproducibility.
Development and evaluation of therapeutically relevant predictive classifiers using gene expression profiling.
  • R. Simon
  • Computer Science
    Journal of the National Cancer Institute
  • 2006
TLDR
Clinical studies of gene expression profi les offer the possibility of improving risk predication and optimizing treatment selection for individual patients and it is shown that various cross-validation approaches can provide better estimates of predictive accuracy.
Concordance among gene-expression-based predictors for breast cancer.
TLDR
Four of the five gene sets used for prognostication in patients with breast cancer showed significant agreement in the outcome predictions for individual patients and are probably tracking a common set of biologic phenotypes.
A Paradigm for Class Prediction Using Gene Expression Profiles
TLDR
The prediction paradigm will serve as a good framework for comparing different prediction methods and may accelerate the development of molecular classifiers that are clinically useful.
The balance of reproducibility, sensitivity, and specificity of lists of differentially expressed genes in microarray studies
TLDR
The results provide practical guidance to choose the appropriate FC and P-value cutoffs when selecting a given number of DEGs and recommend the use of FC-ranking plus a non-stringent P cutoff as a straightforward and baseline practice in order to generate more reproducible DEG lists.
...
...