Optimal screening schedules for prevention of metastatic cancer

Abstract

We develop methodological, mathematical, statistical, and computational approaches to constructing schedules of cancer screening that maximize the probability that by the time of primary tumor detection it has not yet metastasized. Solving this problem is based on a comprehensive mechanistic model of cancer progression. We apply the model with realistic parameters and the screening optimization methodology to mammographic screening for breast cancer within the American female population. We uncover some general patterns of optimal screening schedules. We show that optimization of screening regimens leads to a significant reduction in the probability of detecting breast cancer that has already disseminated.

Evolution of Premalignant Disease

“…In current public health policy, a more or less "onesize-fits-all" approach using population average risk estimates has been used in natural history Markov models to assess optimal screening ages and intervals to help inform screening guidelines (Winawer et al 2006;de Koning et al 2014). Additionally, other mathematical models for optimizing screening schedules take a more evolutionary approach by incorporating stochastic premalignant and malignant clonal expansions explicitly (Jeon et al 2008;Hanin and Pavlova 2013;Curtius et al 2015;Kroep et al 2017). Although case-control studies aim to identify "who" benefits most from treatment during a set surveillance screen schedule, the goal of this modeling is to predict "when" and at what intervals should premalignant patients return to the clinic during follow-up.…”

Optimal timing for cancer screening and adaptive surveillance using mathematical modeling

“…In this study, we applied such models to optimize screening and surveillance. Three examples of modeling techniques that could be formally assessed for screening timing include Markov models for natural history of disease [40, 41, 42], biologically-based models that incorporate dynamic processes like clonal expansions and biomarker shedding [18, 19, 22, 25, 43, 44], and biological event timing models that infer ordered genetic events [45]. While utilizing different data types, these all include distinct stages of latency periods separated by rate-limiting events on a patient’s forecasted trajectory.…”

“…Several previous studies address optimal screening design via biomathematical modeling [18, 19, 23, 46], but this study targeted cancer precursors specifically and allowed for this flexible weighting of events to determine the specific age to screen (and whether to screen) in an analytical framework that does not require simulation. Microsimulation studies using models have also been used to inform policy-making decisions on clinical recommendations/modalities, for example in colorectal cancer [47] and lung cancer [40] screening, by the US Preventive Services Task Force (USPSTF).…”

“…Biological processes not revealed in population data alone give rise to considerable inter-patient heterogeneity during the progression from normal tissue to incident cancer, and thus optimal timing for follow-up is expected to be highly heterogeneous within the population based on individual screening outcome. To account for this heterogeneity along with the temporal effects of screening (e.g., different prognoses resulting from screening earlier versus later), mathematical models can employ a stochastic approach to incorporate premalignant and malignant clonal expansions during clonal evolution explicitly (Figure 1B) [18, 19, 20, 21, 22]. This offers the advantage that the onset of (detectable) precursor clones can be defined as random variables and clonal population sizes can be tracked over time so that dwell times, risk of future cancer, and adaptive surveillance intervals may be formally derived (Figure 1C).…”

Classification in two-stage screening

“…We have not addressed the issue of scheduling (setting the frequency of) screening . We point out though that scheduling indirectly affects classification rules in screening.…”