Dimitris Bertsimas

BertsimasDimitris Bertsimas is currently the Boeing Professor of Operations Research and the codirector of the Operations Research Center at the Massachusetts Institute of Technology. He has  received a BS in Electrical Engineering and Computer Science at the National  Technical University of Athens, Greece in 1985, a MS  in Operations Research at MIT  in 1987, and a Ph.D in Applied Mathematics and Operations Research at MIT in 1988. Since 1988, he has been in the MIT faculty.

His research interests include  optimization,  stochastic systems, data mining,  and their application. In recent years he has worked in robust optimization, health care and finance. He has co-authored more than 100 scientific papers and  he has co-authored the following  books: “Introduction to Linear Optimization” (with J. Tsitsiklis,  Athena Scientific and Dynamic Ideas, 2008), “Data, models and decisions” (with R. Freund, Dynamic Ideas,  2004) and   “Optimization over Integers” (with R. Weismantel, Dynamic Ideas, 2005). He is currently department editor in Optimization for Management Science and former area editor  in Operations Research in Financial Engineering. He has supervised 42 doctoral students and he is currently supervising 10 others.

He is a member of the National Academy of Engineering, and he has received numerous research awards including the Farkas prize (2008), the Erlang prize (1996), the SIAM prize in optimization (1996), the Bodossaki prize (1998) and the Presidential Young Investigator award (1991-1996).

“Designing Clinical Trials for Cancer: An Analytics Approach”

Cancer continues to be one of the leading causes of death in the world.

Multi-drug chemotherapy continues to be the pre-dominant method of treatment in advanced cancers. The design of clinical trials for advanced cancers today  relies pre-dominantly  on the experience/intuition  of the medical doctors involved in the  trial.  The data  from the scientific papers written during 1970-2010   is only available in the form of the original papers published and not in a database.

Correspondingly,  it is not surprising that analytics techniques  have not been applied to designing chemotherapy clinical trials.

In this multi-year effort, we are developing a data based approach for designing chemotherapy clinical trials that has the following ingredients:

a) Using text-mining methods, we   develop a database from  previous clinical trials for the following cancers: gastric, lung, breast, colon, prostate and brain.

This database includes  the control and treatment groups (arms) of the trial, the demographics of the patients in the trial, the drugs used and their dosages, survival statistics and toxicity measurements.

b) Using the database from part (a), we develop statistical models from earlier trials that are capable of predicting the  survival and toxicity of the combination of the drugs used, when the drugs used have been seen in earlier trials, but in different combinations.

c) Using the statistical models from part (b), we develop optimization models that select novel treatment regimens that could be tested in clinical trials, based on the totality of data available on existing combinations. These models  make tradeoffs between exploration and exploitation, that is, strike a balance between learning new things about treatments that may be useful in the future  and also extending the duration of life expectancy by designing more efficient strategies.

We report encouraging progress to date with this research plan.

Joint work with my students Allison O’ Hair,  Steve Relyea and John Silberholz.