Skip to Content
U.S. National Institutes of Health
Cancer Diagnosis Program Cancer Imaging Program Cancer Therapy Evaluation Program Developmental Therapeutics Program Radiation Research Program Translational Research Program Biometric Research Branch Office of Cancer Complementary and Alternative Medicine
Last Updated: 04/25/2012


Contributed by Dr. Richard Simon

Dr. Richard Simon has continued his research on the statistical genomics of cancer and the development and validation of molecular diagnostics for predictive cancer medicine. During 2010, Dr. Simon published or had accepted 19 research papers and gave 13 invited scientific presentations. His interactions with diverse groups from industry, government, and academic research have enabled him to communicate this roadmap to key stakeholders and to attempt to eliminate unnecessary roadblocks to the adoption of useful decision tools that provide patient utility. He has also used his opportunities to interact with diverse groups to address misinterpretations, to increase his understanding of their constraints, to develop innovative clinical trial designs and analysis methodologies that address their needs.

Dr. Simon, in collaboration with Drs. Dan Hayes and Soonmyung Paik developed the “prospective-retrospective design” for evaluating the medical utility of a prognostic or predictive biomarker using a focused prospective analysis plan with archived tissues from an appropriate clinical trial. This approach was used in the development of the OncotypeDx test for patients with breast cancer and for the restriction of labeling of cetuximab and pannitumumab to patients with K-ras wild type colorectal cancer. The manuscript, with Dr. Simon as first author, was published in the Journal of the National Cancer Institute in December 2009 and has been enthusiastically received by many stakeholders. The manuscript provides conditions under which prospective-retrospective studies can provide level I evidence for the medical utility of a prognostic or predictive biomarker. The manuscript provides a modification of the Levels of Evidence Scale previously published by Dr. Hayes and used by the American Society of Clinical Oncology (ASCO) in making their recommendations about which markers should be measured by oncologists.

Dr. Simon and his postdoctoral fellow Dr. Jyothi Subramanian reviewed the literature of prognostic signatures for early lung cancer. They found substantial problems with the planning, analysis, and reporting of these studies and found that none warranted prospective evaluation, much less being ready for clinical use. These results were published in the April 2010 issue of the Journal of the National Cancer Institute. Drs. Subramanian and Simon developed guidelines for the planning, analysis, and reporting of prognostic signatures that appeared in the June 2010 issue of Nature Reviews-Clinical Oncology.

Dr. Simon is pioneering innovations to clinical trial design for a new era in which therapeutics are being targeted to the genomic characterization of individual tumors. His recent publications on the use of genomics in clinical trial design has provided innovative designs for randomized phase 3 clinical trials that prospectively use predictive biomarkers, with prospectively defined analysis plans for evaluating how treatment effect varies among biomarker defined subsets in a manner that reduces the number of patients required compared to previous methods. With his postdoctoral research fellow Dr. Wenyu Jiang and his collaborator Dr. Boris Freidlin, Dr. Simon developed the “Adaptive Signature Design” that permits both development and proper validation of a gene expression signature predictive of benefit from a new treatment in a randomized clinical trial of the new treatment versus control. This approach, based on complete cross-validation, represents an important step forward for using phase 3 clinical trial data to personalize treatment recommendations in a more reliable manner than the traditional post-hoc exploratory analyses. Dr. Simon has generalized this approach and proposed it as a new statistical framework for the design and analysis of predictive clinical trials. Dr. Simon has described this new approach at several conferences and short courses and in publications. At the 2010 Brookings-Friends of Cancer Research Conference, Dr. Simon worked closely with Dr. Eric Rubin of Merck and Dr. Howard Scher of Memorial Sloan Kettering Cancer Center to develop a protocol for use of the adaptive signature design for development of drugs for advanced prostate cancer and was able to obtain Food and Drug Administration (FDA) buy-in to this design. This should provide broad encouragement and guidance for sponsors to develop new cancer drugs using predictive biomarkers.

Dr. Simon has continued to extend the capabilities of his BRB-ArrayTools software for microarray gene expression and copy number variation analysis. BRB-ArrayTools has over 13,000 registered users in 68 countries and has been cited in over 1500 publications. There are more than 300 users in China, one of whom translated the extensive BRB-ArrayTools users guide into Chinese. This software provides biomedical scientists worldwide access to the most sophisticated methods of statistical genomics. The software is the most widely used informatics system provided by NCI.

The BRB website ( receives over 20,000 accesses monthly. Dr. Simon has actively promoted the use of the Biometric Research Branch website to extend the influence of the BRB staff through web-based tools, message boards, Power Point presentations with sound tracks, reprints and technical reports. The website includes web-based programs for the design of clinical trials involving new drugs and predictive biomarkers. These sample size planning tools receive over 100 hits per month. The reprints page receives over 400 hits per month, and over 2000 hits per month for the BRB-ArrayTools message board in which uses from around the world have questions about microarray expression profiling answered.

Dr. Simon has continued to be very active in statistical bioinformatics research. He and postdoctoral research fellow Dr. Ahrim Youn published a new method for the identification of cancer driver genes in tumor sequencing studies. The methods provide improvements in sensitivity and specificity for identifying driver mutations and these methods may be widely adopted in international tumor sequencing projects. Postdoctoral research fellow Dr. Kyung-In Kim and Dr. Simon published a new method for the evaluation of probabilistic classifiers with high dimensional data. This focus on probabilistic classifiers is of importance for medical decision making and is likely to move the direction of research in the area of high-dimensional classification. Drs. Simon and Subramanian evaluated resampling methods for evaluation of the accuracy of survival risk prediction models with high dimensional data.

Dr. Simon provides leadership for the Biometric Research Branch and mentorship of postdoctoral fellows and guest researchers. The branch offers a postdoctoral research training program in statistical cancer genomics and statistical bioinformatics and hosts guest researchers for training in these areas. The branch continues to be very effective in collaborative activities with all DCTD programs and in responsibilities with the CCR. The branch continues to offer a supportive atmosphere for outstanding biostatistians and provides them an environment that encourages creativity and contribution to cancer research.

Freidlin B, Jiang W, Simon R. The cross-validated adaptive signature design. Clinl Cancer Res 2010:16:691-8.

Hunsberger S, Zhao Y, Simon R. A comparison of phase II study strategies. Clinl Cancer Res 2009:15:5950-5.

Kim K, Simon R. Probabilistic classifiers with high dimensional data. Biostat 2011:12;399-412 (Published online Nov. 17, 2010: doi:10.1093/biostatistics/kxq069).

Simon R, Paik S, Hayes DH. Use of archived specimens in evaluation of prognostic and predictive biomarkers. J Natl Cancer Inst 2009:101;446-52.

Simon R. Analysis of DNA microarray expression data. Best Pract Res Clin Haematol 2009:22;271-282.

Simon R. Clinical trial designs for evaluating the medical utility of prognostic and predictive biomarkers in oncology. Per Med 2010:7;33-47.

Simon R. Clinical trials for predictive medicine: New challenges and paradigms. Clin Trials 2010:7;516-24.

Simon R. Moving from correlative science to predictive oncology. EPMA J 2010:1;377-87.

Simon R. Translational clinical trials in oncology: Key bottlenecks and new paradigms. Expert Rev Mol Med 2010:12; e32.

Simon, RM, Subramanin J, Li MC, Menezes S. Using cross validation to evaluate prediction accuracy of survival risk classifiers based on high dimensional data. Brief Bioinform 2011:12;203-14.

Simon R, Simon NR. Using randomization tests to preserve type I error with response-adaptive and covariate-adaptive randomization. Statistics and Probability Letters (Published online

Simon RM, Freidlin B. Re: Designing a randomized clinical trial to evaluate personalized medicine: A new approach based on risk prediction. J Natl Cancer Inst 2011:103;445.

Simon R. Advances in clinical trial design for predictive biomarker discovery and validation, Current Breast Cancer Reports 2009:1;216-21.

Simon R. Drug and pharmacodiagnostic co-development: Statistical Considerations. In: Molecular Diagnostics. Jorgensen JT, Winther H, eds. Singapore: Pan Stanford Publishing;2010.

Simon R. Use of genomics in therapeutic clinical trials. In: Fundamentals of Oncology Clinical Trials, Kelly WK, Halabi S, eds. New York, NY:Demos Medical Publishing;2010.

Simon R. Statistical considerations in the development and validation of cancer biomarkers. In: Molecular Biology of Cancer, Garcia-Foncillas J, ed. (In press).

Subramanian J, Simon R. Gene expression-based prognostic signatures in lung cancer: Ready for clinical use? J Natl Cancer Inst 2010:102;464-74.

Subramanian J, Simon R. What physicians should look for in evaluating reports of gene expression signatures. Nat Rev Clin Oncol 2010:7;466-75.

Subramanian J, Simon R. An evaluation of resampling methods for assessment of survival risk prediction in high dimensional settings. Stat Med (In Press)

Youn A, Simon R. Identifying cancer driver genes in tumor genome sequencing studies. Bioinformatics 27(2):175-81, 2011.

Zhao Y, Simon R. Gene expression deconvolution in clinical samples. Genome Med 2010:2;93.

Zhao Y, Simon R. Development and validation of predictive indices for a continuous outcome using gene expression profiles. Cancer Inform 2010:9;105-14.