Digital Therapeutics as a New Therapeutic Modality: A Review from the Perspective of Clinical Pharmacology.

The promise of transforming digital technologies into treatments is what drives the development of digital therapeutics (DTx), generally known as software applications embedded within accessible technologies-such as smartphones-to treat, manage, or prevent a pathological condition. Whereas DTx solutions that successfully demonstrate effectiveness and safety could drastically improve the life of patients in multiple therapeutic areas, there is a general consensus that generating therapeutic evidence for DTx presents challenges and open questions. We believe there are three main areas where the application of clinical pharmacology principles from the drug development field could benefit DTx development: the characterization of the mechanism of action, the optimization of the intervention, and, finally, its dosing. We reviewed DTx studies to explore how the field is approaching these topics and to better characterize the challenges associated with them. This leads us to emphasize the role that the application of clinical pharmacology principles could play in the development of DTx and to advocate for a development approach that merges such principles from development of traditional therapeutics with important considerations from the highly attractive and fast-paced world of digital solutions.

Model enhanced reinforcement learning to enable precision dosing: A theoretical case study with dosing of propofol.

Extending the potential of precision dosing requires evaluating methodologies offering more flexibility and higher degree of personalization. Reinforcement learning (RL) holds promise in its ability to integrate multidimensional data in an adaptive process built toward efficient decision making centered on sustainable value creation. For general anesthesia in intensive care units, RL is applied and automatically adjusts dosing through monitoring of patient's consciousness. We further explore the problem of optimal control of anesthesia with propofol by combining RL with state-of-the-art tools used to inform dosing in drug development. In particular, we used pharmacokinetic-pharmacodynamic (PK-PD) modeling as a simulation engine to generate experience from dosing scenarios, which cannot be tested experimentally. Through simulations, we show that, when learning from retrospective trial data, more than 100 patients are needed to reach an accuracy within the range of what is achieved with a standard dosing solution. However, embedding a model of drug effect within the RL algorithm improves accuracy by reducing errors to target by 90% through learning to take dosing actions maximizing long-term benefit. Data residual variability impacts accuracy while the algorithm efficiently coped with up to 50% interindividual variability in the PK and 25% in the PD model's parameters. We illustrate how extending the state definition of the RL agent with meaningful variables is key to achieve high accuracy of optimal dosing policy. These results suggest that RL constitutes an attractive approach for precision dosing when rich data are available or when complemented with synthetic data from model-based tools used in model-informed drug development.

Circulating tumor DNA: Opportunities and challenges for pharmacometric approaches.

To support further development of model-informed drug development approaches leveraging circulating tumor DNA (ctDNA), we performed an exploratory analysis of the relationships between treatment-induced changes to ctDNA levels, clinical response and tumor size dynamics in patients with cancer treated with checkpoint inhibitors and targeted therapies. This analysis highlights opportunities for pharmacometrics approaches such as for optimizing sampling design strategies. It also highlights challenges related to the nature of the data and associated variability overall emphasizing the importance of mechanistic modeling studies of the underlying biology of ctDNA processes such as shedding, release and clearance and their relationships with tumor size dynamic and treatment effects.

Reinforcement learning as an innovative model-based approach: Examples from precision dosing, digital health and computational psychiatry.

Model-based approaches are instrumental for successful drug development and use. Anchored within pharmacological principles, through mathematical modeling they contribute to the quantification of drug response variability and enables precision dosing. Reinforcement learning (RL)-a set of computational methods addressing optimization problems as a continuous learning process-shows relevance for precision dosing with high flexibility for dosing rule adaptation and for coping with high dimensional efficacy and/or safety markers, constituting a relevant approach to take advantage of data from digital health technologies. RL can also support contributions to the successful development of digital health applications, recognized as key players of the future healthcare systems, in particular for reducing the burden of non-communicable diseases to society. RL is also pivotal in computational psychiatry-a way to characterize mental dysfunctions in terms of aberrant brain computations-and represents an innovative modeling approach forpsychiatric indications such as depression or substance abuse disorders for which digital therapeutics are foreseen as promising modalities.

Model-Informed Artificial Intelligence: Reinforcement Learning for Precision Dosing.

The availability of multidimensional data together with the development of modern techniques for data analysis represent an exceptional opportunity for clinical pharmacology. Data science-defined in this special issue as the novel approaches to the collection, aggregation, and analysis of data-can significantly contribute to characterize drug-response variability at the individual level, thus enabling clinical pharmacology to become a critical contributor to personalized healthcare through precision dosing. We propose a minireview of methodologies for achieving precision dosing with a focus on an artificial intelligence technique called reinforcement learning, which is currently used for individualizing dosing regimen in patients with life-threatening diseases. We highlight the interplay of such techniques with conventional pharmacokinetic/pharmacodynamic approaches and discuss applicability in drug research and early development.

From waterfall plots to spaghetti plots in early oncology clinical development.

Waterfall plots are used to describe changes in tumor size observed in clinical studies. They are frequently used to illustrate the overall drug response in oncology clinical trials because of its simple representation of results. Unfortunately, this visual display suffers a number of limitations including (1) potential misguidance by masking the time dynamics of tumor size, (2) ambiguous labelling of the y-axis, and (3) low data-to-ink ratio. We offer some alternatives to address these shortcomings and recommend moving away from waterfall plots to the benefit of plots showing the individual time profiles of sum of lesion diameters (according to RECIST). The spider plot presents the individual changes in tumor measurements over time relative to baseline tumor burden. Baseline tumor size is a well-known confounding factor of drug effect which has to be accounted for when analyzing data in early clinical trials. While spider plots are conveniently correct for baseline tumor size, they cannot be presented in isolation. Indeed, percentage change from baseline has suboptimal statistical properties (including skewed distribution) and can be overly optimistic in favor of drug efficacy. We argued that plots of raw data (referred to as spaghetti plots) should always accompany spider plots to provide an equipoised illustration of the drug effect on lesion diameters.

Models and Machines: How Deep Learning Will Take Clinical Pharmacology to the Next Level.

Recent advances in machine learning (ML) have led to enthusiasm about its use throughout the biopharmaceutical industry. The ML methods can be applied to a wide range of problems and have the potential to revolutionize aspects of drug development. The incorporation of ML in modeling and simulation (M&S) has been eagerly anticipated, and in this perspective, we highlight examples in which ML and M&S approaches can be integrated as complementary parts of a clinical pharmacology workflow.

Prediction of the Optimal Dosing Regimen Using a Mathematical Model of Tumor Uptake for Immunocytokine-Based Cancer Immunotherapy.

Purpose: Optimal dosing is critical for immunocytokine-based cancer immunotherapy to maximize efficacy and minimize toxicity. Cergutuzumab amunaleukin (CEA-IL2v) is a novel CEA-targeted immunocytokine. We set out to develop a mathematical model to predict intratumoral CEA-IL2v concentrations following various systemic dosing intensities.Experimental Design: Sequential measurements of CEA-IL2v plasma concentrations in 74 patients with solid tumors were applied in a series of differential equations to devise a model that also incorporates the peripheral concentrations of IL2 receptor-positive cell populations (i.e., CD8+, CD4+, NK, and B cells), which affect tumor bioavailability of CEA-IL2v. Imaging data from a subset of 14 patients were subsequently utilized to additionally predict antibody uptake in tumor tissues.Results: We created a pharmacokinetic/pharmacodynamic mathematical model that incorporates the expansion of IL2R-positive target cells at multiple dose levels and different schedules of CEA-IL2v. Model-based prediction of drug levels correlated with the concentration of IL2R-positive cells in the peripheral blood of patients. The pharmacokinetic model was further refined and extended by adding a model of antibody uptake, which is based on drug dose and the biological properties of the tumor. In silico predictions of our model correlated with imaging data and demonstrated that a dose-dense schedule comprising escalating doses and shortened intervals of drug administration can improve intratumoral drug uptake and overcome consumption of CEA-IL2v by the expanding population of IL2R-positive cells.Conclusions: The model presented here allows simulation of individualized treatment plans for optimal dosing and scheduling of immunocytokines for anticancer immunotherapy. Clin Cancer Res; 24(14); 3325-33. ©2018 AACRSee related commentary by Ruiz-Cerdá et al., p. 3236.

Enhanced Method for Diagnosing Pharmacometric Models: Random Sampling from Conditional Distributions.

PURPOSE: For nonlinear mixed-effects pharmacometric models, diagnostic approaches often rely on individual parameters, also called empirical Bayes estimates (EBEs), estimated through maximizing conditional distributions. When individual data are sparse, the distribution of EBEs can "shrink" towards the same population value, and as a direct consequence, resulting diagnostics can be misleading. METHODS: Instead of maximizing each individual conditional distribution of individual parameters, we propose to randomly sample them in order to obtain values better spread out over the marginal distribution of individual parameters. RESULTS: We evaluated, through diagnostic plots and statistical tests, hypothesis related to the distribution of the individual parameters and show that the proposed method leads to more reliable results than using the EBEs. In particular, diagnostic plots are more meaningful, the rate of type I error is correctly controlled and its power increases when the degree of misspecification increases. An application to the warfarin pharmacokinetic data confirms the interest of the approach for practical applications. CONCLUSIONS: The proposed method should be implemented to complement EBEs-based approach for increasing the performance of model diagnosis.

Early change in tumour size predicts overall survival in patients with first-line metastatic breast cancer.

PURPOSE: Clinical trials using change in tumour size (CTS) as a primary end-point benefit from earlier evaluation of treatment effect and increased study power over progression-free survival, ultimately resulting in more timely regulatory approvals for cancer patients. In this work, a modelling framework was established to further characterise the relationship between CTS and overall survival (OS) in first-line metastatic breast cancer (mBC). METHODS: Data from three randomised phase III trials designed to evaluate the clinical benefit of gemcitabine combination therapy in mBC patients were collated. Two drug-dependent models were developed to describe tumour growth dynamics: the first for paclitaxel/gemcitabine treatment and the second for docetaxel/gemcitabine treatment. A parametric survival model was used to characterise survival as a function of CTS and baseline patient demographics. RESULTS: While the paclitaxel/gemcitabine model incorporated tumour shrinkage by both paclitaxel and gemcitabine with resistance to paclitaxel, the docetaxel/gemcitabine model incorporated shrinkage and resistance to docetaxel alone. Predictors for OS were CTS at week 8, baseline tumour size and ECOG performance status. Model predictions reveal that for an asymptomatic mBC patient with a 6-cm tumour burden, first-line paclitaxel/gemcitabine treatment offers a median OS of 28.6 months, compared to 26.0 months for paclitaxel alone. CONCLUSION: A modelling framework was established, quantitatively describing the tumour growth inhibitory effects of various gemcitabine combotherapies and the effect of the resulting CTS on survival in first-line mBC. This work further supports the use of early CTS as a go/no-go decision point during phase II clinical evaluation of treatments for mBC.

Multi-scale Modeling in Clinical Oncology: Opportunities and Barriers to Success.

Hierarchical processes spanning several orders of magnitude of both space and time underlie nearly all cancers. Multi-scale statistical, mathematical, and computational modeling methods are central to designing, implementing and assessing treatment strategies that account for these hierarchies. The basic science underlying these modeling efforts is maturing into a new discipline that is close to influencing and facilitating clinical successes. The purpose of this review is to capture the state-of-the-art as well as the key barriers to success for multi-scale modeling in clinical oncology. We begin with a summary of the long-envisioned promise of multi-scale modeling in clinical oncology, including the synthesis of disparate data types into models that reveal underlying mechanisms and allow for experimental testing of hypotheses. We then evaluate the mathematical techniques employed most widely and present several examples illustrating their application as well as the current gap between pre-clinical and clinical applications. We conclude with a discussion of what we view to be the key challenges and opportunities for multi-scale modeling in clinical oncology.

Mechanism-based modeling of the clinical effects of bevacizumab and everolimus on vestibular schwannomas of patients with neurofibromatosis type 2.

PURPOSE: To describe the natural growth of vestibular schwannoma in patients with neurofibromatosis type 2 and to predict tumor volume evolution in patients treated with bevacizumab and everolimus. METHODS: Clinical data, including longitudinal tumor volumes in patients treated by bevacizumab (n = 13), everolimus (n = 7) or both (n = 2), were analyzed by means of mathematical modeling techniques. Together with clinical data, data from the literature were also integrated to account for drugs mechanisms of action. RESULTS: We developed a model of vestibular schwannoma growth that takes into account the effect of vascular endothelial growth factors and mammalian target of rapamycin complex 1 on tumor growth. Behaviors, such as tumor growth rebound following everolimus treatment stops, was correctly described with the model. Preliminary results indicate that the model can be used to predict, based on early tumor volume dynamic, tumor response to variation in treatment dose and regimen. CONCLUSION: The developed model successfully describes tumor volume growth before and during bevacizumab and/or everolimus treatment. It might constitute a rational tool to predict patients' response to these drugs, thus potentially improving management of this disease.

Early Prediction of Disease Progression in Small Cell Lung Cancer: Toward Model-Based Personalized Medicine in Oncology.

Predictive biomarkers can play a key role in individualized disease monitoring. Unfortunately, the use of biomarkers in clinical settings has thus far been limited. We have previously shown that mechanism-based pharmacokinetic/pharmacodynamic modeling enables integration of nonvalidated biomarker data to provide predictive model-based biomarkers for response classification. The biomarker model we developed incorporates an underlying latent variable (disease) representing (unobserved) tumor size dynamics, which is assumed to drive biomarker production and to be influenced by exposure to treatment. Here, we show that by integrating CT scan data, the population model can be expanded to include patient outcome. Moreover, we show that in conjunction with routine medical monitoring data, the population model can support accurate individual predictions of outcome. Our combined model predicts that a change in disease of 29.2% (relative standard error 20%) between two consecutive CT scans (i.e., 6-8 weeks) gives a probability of disease progression of 50%. We apply this framework to an external dataset containing biomarker data from 22 small cell lung cancer patients (four patients progressing during follow-up). Using only data up until the end of treatment (a total of 137 lactate dehydrogenase and 77 neuron-specific enolase observations), the statistical framework prospectively identified 75% of the individuals as having a predictable outcome in follow-up visits. This included two of the four patients who eventually progressed. In all identified individuals, the model-predicted outcomes matched the observed outcomes. This framework allows at risk patients to be identified early and therapeutic intervention/monitoring to be adjusted individually, which may improve overall patient survival.

Increasing the Time Interval between PCV Chemotherapy Cycles as a Strategy to Improve Duration of Response in Low-Grade Gliomas: Results from a Model-Based Clinical Trial Simulation.

BACKGROUND: We previously developed a mathematical model capturing tumor size dynamics of adult low-grade gliomas (LGGs) before and after treatment either with PCV (Procarbazine, CCNU, and Vincristine) chemotherapy alone or with radiotherapy (RT) alone. OBJECTIVE: The aim of the present study was to present how the model could be used as a simulation tool to suggest more effective therapeutic strategies in LGGs. Simulations were performed to identify schedule modifications that might improve PCV chemotherapy efficacy. METHODS: Virtual populations of LGG patients were generated on the basis of previously evaluated parameter distributions. Monte Carlo simulations were performed to compare treatment efficacy across in silico clinical trials. RESULTS: Simulations predicted that RT plus PCV would be more effective in terms of duration of response than RT alone. Additional simulations suggested that, in patients treated with PCV chemotherapy, increasing the interval between treatment cycles up to 6 months from the standard 6 weeks can increase treatment efficacy. The predicted median duration of response was 4.3 years in LGGs treated with PCV cycles given every 6 months versus 3.1 years in patients treated with the classical regimen. CONCLUSION: The present study suggests that, in LGGs, mathematical modeling could facilitate clinical research by helping to identify, in silico, potentially more effective therapeutic strategies.

A population pharmacodynamic model for lactate dehydrogenase and neuron specific enolase to predict tumor progression in small cell lung cancer patients.

The development of individualized therapies poses a major challenge in oncology. Significant hurdles to overcome include better disease monitoring and early prediction of clinical outcome. Current clinical practice consists of using Response Evaluation Criteria in Solid Tumors (RECIST) to categorize response to treatment. However, the utility of RECIST is restricted due to limitations on the frequency of measurement and its categorical rather than continuous nature. We propose a population modeling framework that relates circulating biomarkers in plasma, easily obtained from patients, to tumor progression levels assessed by imaging scans (i.e., RECIST categories). We successfully applied this framework to data regarding lactate dehydrogenase (LDH) and neuron specific enolase (NSE) concentrations in patients diagnosed with small cell lung cancer (SCLC). LDH and NSE have been proposed as independent prognostic factors for SCLC. However, their prognostic and predictive value has not been demonstrated in the context of standard clinical practice. Our model incorporates an underlying latent variable ("disease level") representing (unobserved) tumor size dynamics, which is assumed to drive biomarker production and to be influenced by exposure to treatment; these assumptions are in agreement with the known physiology of SCLC and these biomarkers. Our model predictions of unobserved disease level are strongly correlated with disease progression measured by RECIST criteria. In conclusion, the proposed framework enables prediction of treatment outcome based on circulating biomarkers and therefore can be a powerful tool to help clinicians monitor disease in SCLC.