Breast Cancer Survival Strongly Depends On Where First Metastases Develops

Paul K Newton PhD Professor of Aerospace & Mechanical Engineering, Mathematics, and Norris Comprehensive Cancer Center USC Viterbi University of Southern California University Park Campus Los Angeles, CA 90089-4012

Dr. Newton Interview with:
Paul K Newton PhD
Professor of Aerospace & Mechanical Engineering, Mathematics, and Norris Comprehensive Cancer Center
USC Viterbi
University of Southern California
University Park Campus
Los Angeles, CA  90089-4012 

Medical Research: What is the background for this study? What are the main findings?

Dr. Newton: We obtained a longitudinal data set of 446 breast cancer patients from Memorial Sloan Kettering Cancer Center, tracked from 1975 to 2009.

All of the patients had primary breast cancer at the time they entered, with no metastatic tumors. All subsequently developed metastatic breast cancer.

From this time-resolved data set, we first developed what we called tree-ring diagrams showing the full spatiotemporal patterns of progression. We then used this information
to develop a Markov chain dynamical model of metastatic breast cancer. This is a model based on the concept that where the disease currently is located strongly influences where it will spread next.

The systemic nature of metastatic breast cancer is clearly shown in these kinds of network based models.

The main findings are that survival depends very strongly on where the first metastatic tumor develops. For example, if the first metastatic tumor appears in the bone, as happens in roughly 35% of the patients, survival is much better than if it appears in the brain (less than 5% of the patients). Furthermore, for those patients with a first met to the bone, survival is far better for those who develop their next met in the lung area, as compared with those that develop it in the liver.

Metastatic sites are categorized as `spreader’ sites, or `sponge’ sites. Bone and chest wall are generally the primary spreader sites of metastatic breast cancer, dynamically involved in spreading the disease throughout the metastatic process. On the other hand, liver seems to be a key sponge site, where circulating tumor cells most likely accumulate. If one were to focus on an active therapeutic program targeting metastatic sites, most likely the spreader sites would give the most bang-for-buck in terms of survival.

Medical Research: What should clinicians and patients take away from your report?

Dr. Newton: Metastatic breast cancer is a systemic disease with fairly predictable spatiotemporal patterns. It could be beneficial to target treatment not only at the primary site, but also the key spreader/sponge sites of breast cancer.

Medical Research: What recommendations do you have for future research as a result of this study?

Dr. Newton: :More research needs to focus on therapeutic response to targeted metastatic sites in breast cancer. We would like to access similar longitudinal data associated with metastatic lung, prostate, colorectal and other cancers so we can develop similar models for each. Ideally, we would like high quality data sets, as in the MSKCC data set, but with even larger cohorts. Our model is similar to the model Google uses to predict spatiotemporal patterns of web surfers, but their models are based on millions of web surfers, with updates nearly daily. So by comparison, even the very valuable MSKCC data set is small. We are currently looking into the VA (Veterans Administration) databases to see if they have more information on different cancer types. We believe these sorts of models will lead to more accurate and useful statistical forecasting tools to allow patients and doctors to make more informed and quantitative decisions about their future progression and what therapy options are optimal.


Spatiotemporal progression of metastatic breast cancer: a Markov chain model highlighting the role of early metastatic sites

Paul K Newton, Jeremy Mason, Neethi Venkatappa, Maxine S Jochelson Brian Hurt, Jorge Nieva,Elizabeth Comen Larry Norton & Peter Kuhn

npj Breast Cancer 1, Article number: 15018  (2015)



Paul K Newton PhD (2015). Breast Cancer Survival Strongly Depends On Where First Metastases Develops 

Last Updated on November 23, 2015 by Marie Benz MD FAAD