Differences in predictions of ODE models of tumor growth: a cautionary example

Murphy, Hope; Jaafari, Hana; Dobrovolny, Hana M.

doi:10.1186/s12885-016-2164-x

Cited by 144 publications

(176 citation statements)

References 54 publications

Supporting

Mentioning

114

Contrasting

Unclassified

Order By: Relevance

“…Animal models can help in that respect (Mehrara and Forssellaronsson, 2014;Baratchart et al, 2015). But even with rather long time series, distinguishing between different growth models by individual curve fitting is very delicate (Murphy et al, 2016). Here we rely on microscopic biological models and translate them into clinically relevant observables: local lesion volumes and the tumor load.…”

Section: Methodsmentioning

confidence: 99%

“…2.2, we introduce our multi-scale mathematical modeling of disease progression. We first propose several biophysical models for the local lesion growth [V(t|r, v 0 ), blue boxes in Fig 2], corresponding to different scenarii at the microscopic scale (Simeoni et al, 2004;Ayati et al, 2010;Gerlee, 2013;Benzekry et al, 2014;Murphy et al, 2016). We also propose an effective model for the crossover from the single-lesion regime to the dissemination regime [V tot (t|R, V 0 ), green box in Fig 2], which builds on the local lesion growth model, as well as on the IKS model, a model for the dissemination process (Iwata et al, 2000).…”

Section: Overview Of the Modeling Approachmentioning

confidence: 99%

“…This raises the question of how to properly analyze these data, both focusing on static and dynamical properties, a question of both scientific and clinical interest, which has recently started to be addressed (Hartung et al, 2017;Claret et al, 2018) Numerous theoretical models of cancer evolution have been developed, grasping some of the complexity of the biological processes at stake, both for general aspects or more specific to a particular pathology. Descriptive models have in particular been derived for single lesions (Simeoni et al, 2004;Ayati et al, 2010;Herman et al, 2011;Gerlee, 2013;Benzekry et al, 2014;Murphy et al, 2016) as well as for the distribution of disseminated tumors (Iwata et al, 2000;Baratchart et al, 2015). But those are in general not well tested, due to the lack of observations at the right scale, or to the rarity of fully quantitative population datasets.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Towards Quantitative Imaging Biomarkers of Tumor Dissemination: a Multi-scale Parametric Modeling of Multiple Myeloma

Piraud

Wennmann

Kintzelé

et al. 2019

Preprint

View full text Add to dashboard Cite

The advent of medical imaging and automatic image analysis is bringing the full quantitative assessment of lesions and tumor burden at every clinical examination within reach. This opens avenues for the development and testing of functional disease models, as well as their use in the clinical practice for personalized medicine. In this paper, we introduce a Bayesian statistical framework, based on mixed-effects models, to quantitatively test and learn functional disease models at different scales, on population longitudinal data. We also derive an effective mathematical model for the crossover between initially detected lesions and tumor dissemination, based on the Iwata-Kawasaki-Shigesada model. We finally propose to leverage this descriptive disease progression model into model-aware biomarkers for personalized risk-assessment, taking all available examinations and relevant covariates into account. As a use case, we study Multiple Myeloma, a disseminated plasma cell cancer, in which proper diagnostics is essential, to differentiate frequent precursor state without end-organ damage from the rapidly developing disease requiring therapy. After learning the best biological models for local lesion growth and global tumor burden evolution on clinical data, and computing corresponding population priors, we use individual model parameters as biomarkers, and can study them systematically for correlation with external covariates, such as sex or location of the lesion. On our cohort of 63 patients with smoldering Multiple Myeloma, we show that they perform substantially better than other radiological criteria, to predict progression into symptomatic Multiple Myeloma. Our study paves the way for modeling disease progression patterns for Multiple Myeloma, but also for other metastatic and disseminated tumor growth processes, and for analyzing large longitudinal image data sets acquired in oncological imaging. It shows the unprecedented potential of model-based biomarkers for better and more personalized treatment decisions and deserves being validated on larger cohorts to establish its role in clinical decision making.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Overview Of the Modeling Approachmentioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation

Towards Quantitative Imaging Biomarkers of Tumor Dissemination: a Multi-scale Parametric Modeling of Multiple Myeloma

Piraud

Wennmann

Kintzelé

et al. 2019

Preprint

View full text Add to dashboard Cite

show abstract

“…A5) In the absence of immunosurveillance, the tumor growth is limited only by the available resources. Current literature presents many mathematical models to describe the growth of tumors; see, e.g., [39]. Here we use the classic model defined by a logistic law that states that the tumor initially grows rapidly but growth slows as the tumor increases [16].…”

Section: A3) Memory T Cells Are Activated By Tumor Cells This Is a Smentioning

confidence: 99%

Three-Compartment Model of CAR T-cell Immunotherapy

Rodrigues

Barros

Almeida

2019

Preprint

View full text Add to dashboard Cite

Immunotherapy has gained great momentum with CAR T (chimeric antigen receptor) cellular therapy, in which the patient's T lymphocytes are genetically manipulated to recognize tumor-specific antigens to increase elimination efficiency. Although recently approved by FDA to treat B cell malignancies, issues such as dose, administration protocol, toxicity, resistance to immunotherapy, among others, remain open and are the subject of intense research nowadays. Improved CAR T cell immunotherapy requires a better understanding of the interplay between CAR T cell doses and tumor burden. We developed a threecompartment mathematical model to describe tumor response to CAR T cell immunotherapy in immunodeficient mouse models (NSG and SCID/beige) based on two published articles from literature. We modeled different receptors as CART 19BBz or CART 123, and different tumor targets as HDML-2 and RAJI. We considered interactions between tumor cells, effector T cells, and T cell differentiation into memory T cells; tumor-induced immunosuppressive effects, conversion of memory T cells into effector T cells in the presence of tumor cells, and individual specificities considered as uncertainties in the parameters of the model. The model was able to represent the two considered immunotherapy scenarios. For the HDML-2 scenario, the tumor is eliminated after the immunotherapy with the CAR T cells even in case of a challenge due to the memory T cells long-term immune protection. For the Raji tumor, expressing indoleamine 2,3-dioxygenase (IDO) activity, the model represented tumor dynamics even when IDO inhibitors are introduced. Using in silico studies considering different dosing quantities and tumor burden, we showed that the proposed model can represent the three possible outcomes: tumor elimination, equilibrium, and escape. We found that therapy effectiveness may also depend on small variations in the parameters' values, regarded as intrinsic individual specificities, as T cell proliferation capacity, as well as immunosuppressive tumor microenvironment factors. These issues may significantly reduce the chance of tumor elimination. In this way, the developed model provides potential use for assessing different CAR T cell protocols and associated efficacy without further in vivo experiments.

show abstract

“…Some standard and simple mathematical models are commonly used in tumor growth modeling and prediction studies ( [9,13,14,19,23,21,25,26]). A rather striking commonality in most of these studies is the small longitudinal nature (i.e., in terms of the duration and number N of observations or time points) of the data sets employed for model validation.…”

Section: Introductionmentioning

confidence: 99%

The Effect of Statistical Error Model Formulation on the Fit and Selection of Mathematical Models of Tumor Growth for Small Sample Sizes

Bekele-Maxwell¹,

Noorman²,

Menda³

et al. 2018

Int. J. of Pure and Appl. Math.

View full text Add to dashboard Cite

When fitting a mathematical model to a given data set using inverse problems, the correctness of both the mathematical model and the statistical error models are important since an incorrect statistical or observational model directly affects both the estimates and their corresponding standard errors. The effects of these models, among many other factors, are dependent on the sample size and the information content of the data set.In this article, we investigate how the choice of the statistical error model affects the mathematical model fit and accuracy of parameter estimates in small sample size tumor growth data sets. We specifically seek to determine the appropriate statistical error model for small sample size breast, lung and HPV tumor growth data sets obtained from studies on mice. T. Banks et alWe find that for small sample sizes the selection of the best statistical error model is not straightforward and requires the examination of multiple criteria for model fit and uncertainty.Therefore, selection of the best mathematical model is not an easy process for small sample size tumor data and selection of a model based on few data points may not prove accurate.We encourage further research on the optimal design of experiments (duration and number of observations) in order to best fit mathematical models to tumor growth data.

show abstract

Differences in predictions of ODE models of tumor growth: a cautionary example

Cited by 144 publications

References 54 publications

Towards Quantitative Imaging Biomarkers of Tumor Dissemination: a Multi-scale Parametric Modeling of Multiple Myeloma

Towards Quantitative Imaging Biomarkers of Tumor Dissemination: a Multi-scale Parametric Modeling of Multiple Myeloma

Three-Compartment Model of CAR T-cell Immunotherapy

The Effect of Statistical Error Model Formulation on the Fit and Selection of Mathematical Models of Tumor Growth for Small Sample Sizes

Contact Info

Product

Resources

About