Digital twins for predictive oncology will be a paradigm shift for precision cancer care

Hernandez‐Boussard, Tina; Macklin, Paul; Greenspan, Emily J.; Gryshuk, Amy L; Stahlberg, Eric; Syeda-Mahmood, Tanveer; Shmulevich, Ilya

doi:10.1038/s41591-021-01558-5

Cited by 97 publications

(71 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We argue that the MED derived from our approach could lead to increased patient survival and less toxicity. We believe this approach partially supports the notion of a digital twin in cancer care 60, 61 .…”

Section: Discussionsupporting

confidence: 59%

Containing Cancer with Personalized Minimum Effective Dose

Masud

Kim

2022

Preprint

View full text Add to dashboard Cite

Resistance to treatment is a challenge in many cancer therapies. This is partly due to the heterogeneous nature of tumors, where drug-sensitive and drug-resistant cells compete for the same resources. This competition is largely shaped by cancer treatment. The rapid reduction of drug-sensitive cell population during therapy with a maximum-tolerated dose relaxes competitive stress on the drug-resistant cell population, promoting relapse. Therefore, maintaining a high level of drug-sensitive cell population with a treatment break or lower dose can impose effective competitive stress on drug-resistant cell populations. Adaptive therapy (AT) exploits the competition between cancer cells. However, given the heterogeneous treatment response of individual patients, determining a personalized optimal treatment that can fine-tune competitive stress remains challenging. Using a deterministic model of cancer cell population competition, this study defines an effective dose window (EDW) as a range of doses that conserve sufficient sensitive cells, while maintaining the tumor volume below a threshold (e.g., initial tumor volume), to maintain a sustained competition against resistant cells. As a proof of concept, we sought to determine the EDW for a small cohort of patients with melanoma (n=8). We first fitted the model to longitudinal tumor response data from each patient. We performed structural and practical identifiability analyses to confirm the reproducibility and uniqueness of the estimated parameters. Then, we considered a subset of the cohort with uniquely identifiable parameters and estimated patient-specific EDW. We demonstrated that if the dose belongs to the EDW, the tumor volume for each patient could be indefinitely contained either using continuous or AT strategy. Using the optimal control theory, we concluded that the lower bound of the EDW approximates the minimum effective dose (MED) for containing cancer. Taken together, using tumor biomarker data, this study provides a proof of concept that there may exist a patient-specific EDW that keeps the tumor below a threshold (e.g., initial volume) by maintaining sustained competition on resistant cells.

show abstract

Section: Discussionsupporting

confidence: 59%

Containing Cancer with Personalized Minimum Effective Dose

Masud

Kim

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…For example, mathematical models of metabolism and cell proliferation indicated new molecular targets (48) and such multicellular models can further predict outcomes such as necrosis and growth arrest (49), cancer cell migration (50), and immune cell invasion (51). It can be envisioned that the computational efforts described herein can contribute to proposed Digital Twins for personalized medicine in cancer (52).…”

Section: Discussionmentioning

confidence: 99%

Capturing Biomarkers and Molecular Targets in Cellular Landscapes From Dynamic Reaction Network Models and Machine Learning

Mertins

2022

Front. Oncol.

View full text Add to dashboard Cite

Computational dynamic ODE models of cell function describing biochemical reactions have been created for decades, but on a small scale. Still, they have been highly effective in describing and predicting behaviors. For example, oscillatory phospho-ERK levels were predicted and confirmed in MAPK signaling encompassing both positive and negative feedback loops. These models typically were limited and not adapted to large datasets so commonly found today. But importantly, ODE models describe reaction networks in well-mixed systems representing the cell and can be simulated with ordinary differential equations that are solved deterministically. Stochastic solutions, which can account for noisy reaction networks, in some cases, also improve predictions. Today, dynamic ODE models rarely encompass an entire cell even though it might be expected that an upload of the large genomic, transcriptomic, and proteomic datasets may allow whole cell models. It is proposed here to combine output from simulated dynamic ODE models, completed with omics data, to discover both biomarkers in cancer a priori and molecular targets in the Machine Learning setting.

show abstract

“…In a sequential article we will describe the developing FAIR orchestration architecture to deal effectively with distributed analytics over FAIR data and in particular FDTs. Although the concept of a digital twin is getting some traction in the life sciences field (Hernandez-Boussard et al, 2021 ) we would like to emphasize that digital twins need to be FAIR in order to be “AI ready” (i.e., machine actionable and interpretable) to be optimally repurposable. To support this argument, we give here a more current example that allows an insight in how cardinal assertions and Knowlets can already be used with existing infrastructure ( www.euretos.com ) for rapid rationalization of real-world observations using curated established knowledge.…”

Section: Real World Observations and Fdts Of People And Objectsmentioning

confidence: 99%

FAIR Digital Twins for Data-Intensive Research

Schultes¹,

Roos

Santos

et al. 2022

Front. Big Data

View full text Add to dashboard Cite

Although all the technical components supporting fully orchestrated Digital Twins (DT) currently exist, what remains missing is a conceptual clarification and analysis of a more generalized concept of a DT that is made FAIR, that is, universally machine actionable. This methodological overview is a first step toward this clarification. We present a review of previously developed semantic artifacts and how they may be used to compose a higher-order data model referred to here as a FAIR Digital Twin (FDT). We propose an architectural design to compose, store and reuse FDTs supporting data intensive research, with emphasis on privacy by design and their use in GDPR compliant open science.

show abstract

Digital twins for predictive oncology will be a paradigm shift for precision cancer care

Cited by 97 publications

References 8 publications

Containing Cancer with Personalized Minimum Effective Dose

Containing Cancer with Personalized Minimum Effective Dose

Capturing Biomarkers and Molecular Targets in Cellular Landscapes From Dynamic Reaction Network Models and Machine Learning

FAIR Digital Twins for Data-Intensive Research

Contact Info

Product

Resources

About