Mathematical Biosciences最新文献

Background:

The increased application of immune checkpoint inhibitors (ICIs) targeting PD-1/PD-L1 in lung cancer treatment generates clinical need to reliably predict individual patients’ treatment outcomes.

Methods:

To bridge the prediction gap, we examine four different mathematical models in the form of ordinary differential equations, including a novel delayed response model. We rigorously evaluate their individual and combined predictive capabilities with regard to the patients’ progressive disease (PD) status through equal weighting of model-derived outcome probabilities.

Results:

Fitting the complete treatment course, the novel delayed response model ($R^2=0.938$) outperformed the simplest model ($R^2=0.865$). The model combination was able to reliably predict patient PD outcome with an overall accuracy of 77% (sensitivity = 70%, specificity = 81%), solely through calibration with primary tumor longest diameter measurements. It autonomously identified a subset of 51% of patients where predictions with an overall accuracy of 81% (sensitivity = 81%, specificity = 81%) can be achieved. All models significantly outperformed a fully data-driven machine learning-based approach.

Implications

: These modeling approaches provide a dynamic baseline framework to support clinicians in treatment decisions by identifying different treatment outcome trajectories with already clinically available measurement data.

Limitations and future directions:

Conjoint application of the presented approach with other predictive tools and biomarkers, as well as further disease information (e.g. metastatic stage), could further enhance treatment outcome prediction. We believe the simple model formulations allow widespread adoption of the developed models to other cancer types. Similar models can easily be formulated for other treatment modalities.

Vaccination against severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) results in transient antibody response against the spike protein. The individual immune status at the time of vaccination influences the response. Using mathematical models of antibody decay, we determined the dynamics of serum immunoglobulin G (IgG) and serum immunoglobulin A (IgA) over time. Data fitting to longitudinal IgG and IgA titers was used to quantify differences in antibody magnitude and antibody duration among infection-naïve and infection-positive vaccinees. We found that prior infections result in more durable serum IgG and serum IgA responses, with prior symptomatic infections resulting in the most durable serum IgG response and prior asymptomatic infections resulting in the most durable serum IgA response. These findings can guide vaccine boosting schedules.

The paper studies multi-competitive continuous-time epidemic processes. We consider the setting where two viruses are simultaneously prevalent, and the spread occurs due to individual-to-individual interaction. In such a setting, an individual is either not affected by any of the viruses, or infected by one and exactly one of the two viruses. One of the equilibrium points is the coexistence equilibrium, i.e., multiple viruses simultaneously infect separate fractions of the population. We provide a sufficient condition for the existence of a coexistence equilibrium. We identify a condition such that for certain pairs of spread matrices either every coexistence equilibrium lies on a line that is locally exponentially attractive, or there is no coexistence equilibrium. We then provide a condition that, for certain pairs of spread matrices, rules out the possibility of the existence of a coexistence equilibrium, and, as a consequence, establishes global asymptotic convergence to the endemic equilibrium of the dominant virus. Finally, we provide a mitigation strategy that employs one virus to ensure that the other virus is eradicated. The theoretical results are illustrated using simulations.

Antimicrobial heteroresistance refers to the presence of different subpopulations with heterogeneous antimicrobial responses within the same bacterial isolate, so they show reduced susceptibility compared with the main population. Though it is widely accepted that heteroresistance can play a crucial role in the outcome of antimicrobial treatments, predictive Antimicrobial Resistance (AMR) models accounting for bacterial heteroresistance are still scarce and need to be refined as the techniques to measure heteroresistance become standardised and consistent conclusions are drawn from data. In this work, we propose a multivariate Birth-Death (BD) model of bacterial heteroresistance and analyse its properties in detail. Stochasticity in the population dynamics is considered since heteroresistance is often characterised by low initial frequencies of the less susceptible subpopulations, those mediating AMR transmission and potentially leading to treatment failure. We also discuss the utility of the heteroresistance model for practical applications and calibration under realistic conditions, demonstrating that it is possible to infer the model parameters and heteroresistance distribution from time-kill data, i.e., by measuring total cell counts alone and without performing any heteroresistance test.

Global climate change projections indicate that the atmospheric concentration of carbon dioxide will increase twofold by the end of this century. However, how the elevated carbon dioxide affects aquatic carbon sequestration and species composition within aquatic microbial communities remains inconclusive. To address this knowledge gap, we formulate a bacteria-algae interaction model to characterize the effects of elevated carbon dioxide on aquatic ecosystems and rigorously derive the thresholds determining the persistence and extinction of algae or bacteria. We explore the impacts of abiotic factors, such as light intensity, nutrient concentration, inorganic carbon concentration and water depth, on algae and bacteria dynamics. The main findings indicate that the elevated atmospheric carbon dioxide will increase algae biomass and thus facilitate carbon sequestration. On the other hand, the elevated atmospheric carbon dioxide will reduce bacterial biomass, and excessive carbon dioxide concentrations can even destroy bacterial communities. Numerical simulations indicate that eutrophication and intensified light intensity can reduce aquatic carbon sequestration, while elevated atmospheric carbon dioxide levels can mitigate eutrophication. Furthermore, higher algae respiration and death rates are detrimental to carbon sequestration, whereas the increased bacterial respiration rates promote carbon sequestration.

Investigating disease progression, transmission of infection and impacts of Multidrug Therapy (MDT) to inhibit demyelination in leprosy involves a certain amount of difficulty in terms of the in-built uncertain complicated and complex intracellular cell dynamical interactions. To tackle this scenario and to elucidate a more realistic, rationalistic approach of examining the infection mechanism and associated drug therapeutic interventions, we propose a four-dimensional ordinary differential equation-based model. Stochastic processes has been employed on this deterministic system by formulating the Kolmogorov forward equation introducing a transition state and the quasi-stationary distribution, exact distribution analysis have been investigated which allow us to estimate an expected time to extinction of the infected Schwann cells into the human body more prominently. Additionally, to explore the impact of uncertainty in the key intracellular factors, the stochastic system is investigated incorporating random perturbations and environmental noises in the disease dissemination, proliferation and reinfection rates. Rigorous numerical simulations validating the analytical outcomes provide us significant novel insights on the progression of leprosy and unravelling the existing major treatment complexities. Analytical experiments along with the simulations utilizing Monte-Carlo method and Euler–Maruyama scheme involving stochasticity predicts that the bacterial density is underestimated due to the recurrence of infection and suggests that maintaining a drug-efficacy rate in the range $0.6-0.8$ would be substantially efficacious in eradicating leprosy.

The invasion of hematophagous arthropod species in human settlements represents a threat, not only to the economy but also to the health system in general. Recent examples of this phenomenon were seen in Paris and Mexico City, evidencing the importance of understanding these dynamics. In this work, we present a reaction–diffusion model to describe the invasion dynamics of hematophagous arthropod species. The proposed model considers a denso-dependent growth rate and parameters related to the control of the invasive species. Our results illustrate the existence of two invasion levels (presence and infestation) within a region, depending on control parameter values. We also prove analytically the existence of the presence and infestation waves and show different theoretical types of invasion waves that result from varying control parameters. In addition, we present a condition threshold that determines whether or not an infestation occurs. Finally, we illustrate some results when considering the case of bedbugs and brown dog ticks as invasion species.

We introduce a class of discrete single species models with distributed delay in the reproductive process and a cohort dependent survival function that accounts for survival pressure during that delay period. These delay recurrences track the mature population for species in which individuals reach maturity after at least $\tau$ and at most $\tau +{\tau }_M$ breeding cycles. Under realistic model assumptions, we prove the existence of a critical delay threshold, ${\tilde{\tau }}_c$. For given delay kernel length ${\tau }_M$, if each individual takes at least ${\tilde{\tau }}_c$ time units to reach maturity, then the population is predicted to go extinct. We show that the positive equilibrium is decreasing in both $\tau$ and ${\tau }_M$. In the case of a constant reproductive rate, we provide an equation to determine ${\tilde{\tau }}_c$ for fixed ${\tau }_M$, and similarly, provide a lower bound on the kernel length, ${\tilde{\tau }}_M$ for fixed $\tau$ such that the population goes extinct if ${\tau }_M\ge {\tilde{\tau }}_M$. We compare these critical thresholds for different maturation distributions and show that if all else is the same, to avoid extinction it is best if all individuals in the population have the shortest delay possible. We apply the model derivation to a Beverton–Holt model and discuss its global dynamics. For this model with kernels that share the same mean delay, we show that populations with the largest variance in the time required to reach maturity have higher population levels and lower chances of extinction.