Toward a science of tumor forecasting for clinical oncology

Abstract

We propose that the quantitative cancer biology community makes a concerted effort to apply lessons from weather forecasting to develop an analogous methodology for predicting and evaluating tumor growth and treatment response. Currently, the time course of tumor response is not predicted; instead, response is only assessed post hoc by physical examination or imaging methods. This fundamental practice within clinical oncology limits optimization of a treatment regimen for an individual patient, as well as to determine in real time whether the choice was in fact appropriate. This is especially frustrating at a time when a panoply of molecularly targeted therapies is available, and precision genetic or proteomic analyses of tumors are an established reality. By learning from the methods of weather and climate modeling, we submit that the forecasting power of biophysical and biomathematical modeling can be harnessed to hasten the arrival of a field of predictive oncology. With a successful methodology toward tumor forecasting, it should be possible to integrate large tumor-specific datasets of varied types and effectively defeat one cancer patient at a time.

Figure 1

The figure presents an overview of how numerical tumor prediction would work in practice. The diagnostic phase consists of assembling the appropriate patient specific data (panel a) and building the initial state vector, T(r,t₀) (panel b), which provides an initial snapshot of the key biological characteristics of the tumor at position r and initial time t₀. In this example, T(r,t₀) consists of a series of genomic, G_i(r,t₀), and imaging, I_i(r,t₀), measurements. The initial state vector then initializes the predictive model (panel c) which is projected forward in time to yield a final state vector (panel d) which is then interpreted (panel e) to give a snapshot of the key biological characteristics of the tumor at some future time point, t_f.

Figure 2

The scheme in figure 1 is easily extended to allow for patient specific clinical trials. Namely, after collecting the data to build the initial state vector, T(r,t₀) (panel a), the data is used to initialize a predictive model (panel b) that includes the effects of therapy. Here, “therapy” can be any of n different regimens that systematically vary (for example) dose, timing, and order of multiple treatments (panel c) to yield multiple final state vectors, T_i(r,t₀) (panel d). The i^th therapy yields the i^th final state vector which is then compared to select the optimal therapeutic regimen for the particular patient under investigation.