A Machine-Learning Approach to Identify a Prognostic Cytokine Signature That Is Associated With Nivolumab Clearance in Patients With Advanced Melanoma

Artificial Intelligence: From Buzzword to Useful Tool in Clinical Pharmacology

The advent of artificial intelligence (AI) in clinical pharmacology and drug development is akin to the dawning of a new era. Previously dismissed as merely technological hype, these approaches have emerged as promising tools in different domains, including health care, demonstrating their potential to empower clinical pharmacology decision making, revolutionize the drug development landscape, and advance patient care. Although challenges remain, the remarkable progress already made signals that the leap from hype to reality is well underway, and AI promises to offer clinical pharmacology new tools and possibilities for optimizing patient care is gradually coming to fruition. This review dives into the burgeoning world of AI and machine learning (ML), showcasing different applications of AI in clinical pharmacology and the impact of successful AI/ML implementation on drug development and/or regulatory decisions. This review also highlights recommendations for areas of opportunity in clinical pharmacology, including data analysis (e.g., handling large data sets, screening to identify important covariates, and optimizing patient population) and efficiencies (e.g., automation, translation, literature curation, and training). Realizing the benefits of AI in drug development and understanding its value will lead to the successful integration of AI tools in our clinical pharmacology and pharmacometrics armamentarium.

Antibody Drug Clearance: An Underexplored Marker of Outcomes with Checkpoint Inhibitors

Immune-checkpoint inhibitor (ICI) therapy has dramatically changed the clinical landscape for several cancers, and ICI use continues to expand across many cancer types. Low baseline clearance (CL) and/or a large reduction of CL during treatment correlates with better clinical response and longer survival. Similar phenomena have also been reported with other monoclonal antibodies (mAb) in cancer and other diseases, highlighting a characteristic of mAb clinical pharmacology that is potentially shared among various mAbs and diseases. Though tempting to attribute poor outcomes to low drug exposure and arguably low target engagement due to high CL, such speculation is not supported by the relatively flat exposure-response relationship of most ICIs, where a higher dose or exposure is not likely to provide additional benefit. Instead, an elevated and/or increasing CL could be a surrogate marker of the inherent resistant phenotype that cannot be reversed by maximizing drug exposure. The mechanisms connecting ICI clearance, therapeutic efficacy, and resistance are unclear and likely to be multifactorial. Therefore, to explore the potential of ICI CL as an early marker for efficacy, this review highlights the similarities and differences of CL characteristics and CL-response relationships for all FDA-approved ICIs, and we compare and contrast these to selected non-ICI mAbs. We also discuss underlying mechanisms that potentially link mAb CL with efficacy and highlight existing knowledge gaps and future directions where more clinical and preclinical investigations are warranted to clearly understand the value of baseline and/or time-varying CL in predicting response to ICI-based therapeutics.

Quantitative Systems Pharmacology and Machine Learning: A Match Made in Heaven or Hell?

As pharmaceutical development moves from early-stage in vitro experimentation to later in vivo and subsequent clinical trials, data and knowledge are acquired across multiple time and length scales, from the subcellular to whole patient cohort scale. Realizing the potential of this data for informing decision making in pharmaceutical development requires the individual and combined application of machine learning (ML) and mechanistic multiscale mathematical modeling approaches. Here we outline how these two approaches, both individually and in tandem, can be applied at different stages of the drug discovery and development pipeline to inform decision making compound development. The importance of discerning between knowledge and data are highlighted in informing the initial use of ML or mechanistic quantitative systems pharmacology (QSP) models. We discuss the application of sensitivity and structural identifiability analyses of QSP models in informing future experimental studies to which ML may be applied, as well as how ML approaches can be used to inform mechanistic model development. Relevant literature studies are highlighted and we close by discussing caveats regarding the application of each approach in an age of constant data acquisition. SIGNIFICANCE STATEMENT: We consider when best to apply machine learning (ML) and mechanistic quantitative systems pharmacology (QSP) approaches in the context of the drug discovery and development pipeline. We discuss the importance of prior knowledge and data available for the system of interest and how this informs the individual and combined application of ML and QSP approaches at each stage of the pipeline.

Therapeutic drug monitoring of immune checkpoint inhibitors: based on their pharmacokinetic properties and biomarkers

As a new means of oncology treatment, immune checkpoint inhibitors (ICIs) can improve survival rates in patients with resistant or refractory tumors. However, there are obvious inter-individual differences in the unsatisfactory response rate, drug resistance rate and the occurrence of immune-related adverse events (irAE). These questions have sparked interest in researchers looking for a way to screen sensitive populations and predict efficacy and safety. Therapeutic drug monitoring (TDM) is a way to ensure the safety and effectiveness of medication by measuring the concentration of drugs in body fluids and adjusting the medication regimen. It has the potential to be an adjunctive means of predicting the safety and efficacy of ICIs treatment. In this review, the author outlined the pharmacokinetic (PK) characteristics of ICIs in patients. The feasibility and limitations of TDM of ICIs were discussed by summarizing the relationships between the pharmacokinetic parameters and the efficacy, toxicity and biomarkers.

A comprehensive regulatory and industry review of modeling and simulation practices in oncology clinical drug development

Exposure-response (E-R) analyses are an integral component in the development of oncology products. Characterizing the relationship between drug exposure metrics and response allows the sponsor to use modeling and simulation to address both internal and external drug development questions (e.g., optimal dose, frequency of administration, dose adjustments for special populations). This white paper is the output of an industry-government collaboration among scientists with broad experience in E-R modeling as part of regulatory submissions. The goal of this white paper is to provide guidance on what the preferred methods for E-R analysis in oncology clinical drug development are and what metrics of exposure should be considered.

Population Pharmacokinetic Modeling Combined With Machine Learning Approach Improved Tacrolimus Trough Concentration Prediction in Chinese Adult Liver Transplant Recipients

This study aimed to develop and evaluate a population pharmacokinetic (PPK) combined machine learning approach to predict tacrolimus trough concentrations for Chinese adult liver transplant recipients in the early posttransplant period. Tacrolimus trough concentrations were retrospectively collected from routine monitoring records of liver transplant recipients and divided into the training data set (1287 concentrations in 145 recipients) and the test data set (296 concentrations in 36 recipients). A PPK model was first established using NONMEM. Then a machine learning model of Xgboost was adapted to fit the estimated individual pharmacokinetic parameters obtained from the PPK model with Bayesian forecasting. The performance of the final PPK model and Xgboost model was compared in the test data set. In the final PPK model, tacrolimus daily dose, postoperative days, hematocrit, aspartate aminotransferase, and concomitant voriconazole, were identified to significantly influence the clearance. The postoperative days along with hematocrit significantly influence the volume of distribution. In the Xgboost model, the first 5 predictors for predicting the clearance were concomitant with voriconazole, sex, single nucleotide polymorphisms of CYP3A4*1G and CYP3A5*3 in recipients, and tacrolimus daily dose, for the volume of distribution were postoperative days, age, weight, total bilirubin and graft : recipient weight ratio. In the test data set, the Xgboost model showed the minimum median prediction error of tacrolimus concentrations, less than the PPK model with or without Bayesian forecasting. In conclusion, a PPK combined machine learning approach could improve the prediction of tacrolimus concentrations for Chinese adult liver transplant recipients in the early posttransplant period.

Predictive models and early postoperative recurrence evaluation for hepatocellular carcinoma based on gadoxetic acid-enhanced MR imaging

BACKGROUND

The prognosis of hepatocellular carcinoma (HCC) is still poor largely due to the high incidence of recurrence. We aimed to develop and validate predictive models of early postoperative recurrence for HCC using clinical and gadoxetic acid-enhanced magnetic resonance (MR) imaging-based findings.

METHODS

In this retrospective case-control study, 209 HCC patients, who underwent gadoxetic acid-enhanced MR imaging before curative-intent resection, were enrolled. Boruta algorithm and backward stepwise selection with Akaike information criterion (AIC) were used for variables selection Random forest, Gradient-Boosted decision tree and logistic regression model analysis were used for model development. The area under the receiver operating characteristic curve (AUC), calibration plots, and decision curve analysis were used to evaluate model's performance.

RESULTS

One random forest model with Boruta algorithm (RF-Boruta) was developed consisting of preoperative serum ALT and AFP levels and six MRI findings, while preoperative serum AST and AFP levels and four MRI findings were included in one logistic regression model with backward stepwise selection method (Logistic-AIC).The two predictive models demonstrated good discrimination performance in both the training set (RF-Boruta: AUC, 0.820; Logistic-AIC: AUC, 0.853), internal validation set (RF-Boruta: AUC, 0.857, Logistic-AIC: AUC, 0.812) and external validation set(RF-Boruta: AUC, 0.805, Logistic-AIC: AUC, 0.789). Besides, in both the internal validation and external validation sets, the RF-Boruta model outperformed Barcelona Clinic Liver Cancer (BCLC) stage (p < 0.05).

CONCLUSIONS

The RF-Boruta and Logistic-AIC models with good prediction performance for early postoperative recurrence may lead to optimal and comprehensive treatment approaches, and further improve the prognosis of HCC after resection.

Artificial Intelligence and Advanced Melanoma: Treatment Management Implications

Artificial intelligence (AI), a field of research in which computers are applied to mimic humans, is continuously expanding and influencing many aspects of our lives. From electric cars to search motors, AI helps us manage our daily lives by simplifying functions and activities that would be more complex otherwise. Even in the medical field, and specifically in oncology, many studies in recent years have highlighted the possible helping role that AI could play in clinical and therapeutic patient management. In specific contexts, clinical decisions are supported by "intelligent" machines and the development of specific softwares that assist the specialist in the management of the oncology patient. Melanoma, a highly heterogeneous disease influenced by several genetic and environmental factors, to date is still difficult to manage clinically in its advanced stages. Therapies often fail, due to the establishment of intrinsic or secondary resistance, making clinical decisions complex. In this sense, although much work still needs to be conducted, numerous evidence shows that AI (through the processing of large available data) could positively influence the management of the patient with advanced melanoma, helping the clinician in the most favorable therapeutic choice and avoiding unnecessary treatments that are sure to fail. In this review, the most recent applications of AI in melanoma will be described, focusing especially on the possible finding of this field in the management of drug treatments.

Applying interpretable machine learning workflow to evaluate exposure-response relationships for large-molecule oncology drugs

The application of logistic regression (LR) and Cox Proportional Hazard (CoxPH) models are well-established for evaluating exposure-response (E-R) relationship in large molecule oncology drugs. However, applying machine learning (ML) models on evaluating E-R relationships has not been widely explored. We developed a workflow to train regularized LR/CoxPH and tree-based XGboost (XGB) models, and derive the odds ratios for best overall response and hazard ratios for overall survival, across exposure quantiles to evaluate the E-R relationship using clinical trial datasets. The E-R conclusions between LR/CoxPH and XGB models are overall consistent, and largely aligned with historical pharmacometric analyses findings. Overall, applying this interpretable ML workflow provides a promising alternative method to assess E-R relationships for impacting key dosing decisions in drug development.

Performance of Cox proportional hazard models on recovering the ground truth of confounded exposure-response relationships for large-molecule oncology drugs

A Cox proportional hazard (CoxPH) model is conventionally used to assess exposure-response (E-R), but its performance to uncover the ground truth when only one dose level of data is available has not been systematically evaluated. We established a simulation workflow to generate realistic E-R datasets to assess the performance of the CoxPH model in recovering the E-R ground truth in various scenarios, considering two potential reasons for the confounded E-R relationship. We found that at high doses, when the pharmacological effects are largely saturated, missing important confounders is the major reason for inferring false-positive E-R relationships. At low doses, when a positive E-R slope is the ground truth, either missing important confounders or mis-specifying the interactions can lead to inaccurate estimates of the E-R slope. This work constructed a simulation workflow generally applicable to clinical datasets to generate clinically relevant simulations and provide an in-depth interpretation on the E-R relationships with confounders inferred by the conventional CoxPH model.

Adoption of Machine Learning in Pharmacometrics: An Overview of Recent Implementations and Their Considerations

Pharmacometrics is a multidisciplinary field utilizing mathematical models of physiology, pharmacology, and disease to describe and quantify the interactions between medication and patient. As these models become more and more advanced, the need for advanced data analysis tools grows. Recently, there has been much interest in the adoption of machine learning (ML) algorithms. These algorithms offer strong function approximation capabilities and might reduce the time spent on model development. However, ML tools are not yet an integral part of the pharmacometrics workflow. The goal of this work is to discuss how ML algorithms have been applied in four stages of the pharmacometrics pipeline: data preparation, hypothesis generation, predictive modelling, and model validation. We will also discuss considerations before the use of ML algorithms with respect to each topic. We conclude by summarizing applications that hold potential for adoption by pharmacometricians.

Application of SHAP values for inferring the optimal functional form of covariates in pharmacokinetic modeling

In population pharmacokinetic (PK) models, interindividual variability is explained by implementation of covariates in the model. The widely used forward stepwise selection method is sensitive to bias, which may lead to an incorrect inclusion of covariates. Alternatives, such as the full fixed effects model, reduce this bias but are dependent on the chosen implementation of each covariate. As the correct functional forms are unknown, this may still lead to an inaccurate selection of covariates. Machine learning (ML) techniques can potentially be used to learn the optimal functional forms for implementing covariates directly from data. A recent study suggested that using ML resulted in an improved selection of influential covariates. However, how do we select the appropriate functional form for including these covariates? In this work, we use SHapley Additive exPlanations (SHAP) to infer the relationship between covariates and PK parameters from ML models. As a case-study, we use data from 119 patients with hemophilia A receiving clotting factor VIII concentrate peri-operatively. We fit both a random forest and a XGBoost model to predict empirical Bayes estimated clearance and central volume from a base nonlinear mixed effects model. Next, we show that SHAP reveals covariate relationships which match previous findings. In addition, we can reveal subtle effects arising from combinations of covariates difficult to obtain using other methods of covariate analysis. We conclude that the proposed method can be used to extend ML-based covariate selection, and holds potential as a complete full model alternative to classical covariate analyses.

Bioinformatic and Machine Learning Applications in Melanoma Risk Assessment and Prognosis: A Literature Review

Over 100,000 people are diagnosed with cutaneous melanoma each year in the United States. Despite recent advancements in metastatic melanoma treatment, such as immunotherapy, there are still over 7000 melanoma-related deaths each year. Melanoma is a highly heterogenous disease, and many underlying genetic drivers have been identified since the introduction of next-generation sequencing. Despite clinical staging guidelines, the prognosis of metastatic melanoma is variable and difficult to predict. Bioinformatic and machine learning analyses relying on genetic, clinical, and histopathologic inputs have been increasingly used to risk stratify melanoma patients with high accuracy. This literature review summarizes the key genetic drivers of melanoma and recent applications of bioinformatic and machine learning models in the risk stratification of melanoma patients. A robustly validated risk stratification tool can potentially guide the physician management of melanoma patients and ultimately improve patient outcomes.

Augmented clearance of nivolumab is associated with renal functions in chronic renal disease model rats

The clinically approved dose of nivolumab is 240 mg Q2W. However, previous studies have shown that baseline nivolumab clearance (CL) is associated with treatment outcomes in patients with solid cancers, thus motivating researchers to identify prognostic factors and indices influencing nivolumab CL. This study used chronic kidney disease model rats to investigate whether chronic renal impairment affected nivolumab CL and explored the surrogate markers associated with nivolumab CL. We observed that the total CL for nivolumab (CL_tot) was approximately 1.42-times higher in chronic kidney disease model rats than that in sham rats with an increased urinary excretion. Additionally, CL_tot showed positive correlation with renal CL for nivolumab (CL_R), but not with extrarenal CL. Furthermore, the baseline levels of creatinine, blood urea nitrogen, creatinine CL, and urinary albumin/creatine ratio based on laboratory data were also significantly correlated with CL_R Our findings suggest that nivolumab CL increases as renal function deteriorates due to an increased excretion of nivolumab in the urine; additionally, laboratory data reflecting renal function may be a feasible index to qualitatively estimate nivolumab CL prior to nivolumab treatment under conditions of renal impairment. Significance Statement We demonstrated that nivolumab was rapidly eliminated from the circulation in chronic kidney disease model rats compared to sham rats with an increased urinary nivolumab excretion. Moreover, nivolumab clearance was significantly correlated with the baseline levels of certain laboratory parameters reflecting renal functions. These results indicate the potential applicability of baseline renal function as a prognostic index to qualitatively estimate nivolumab clearance prior to nivolumab treatment under conditions with renal impairment.

Application of Machine Learning in Translational Medicine: Current Status and Future Opportunities

The exponential increase in our ability to harness multi-dimensional biological and clinical data from experimental to real-world settings has transformed pharmaceutical research and development in recent years, with increasing applications of artificial intelligence (AI) and machine learning (ML). Patient-centered iterative forward and reverse translation is at the heart of precision medicine discovery and development across the continuum from target validation to optimization of pharmacotherapy. Integration of advanced analytics into the practice of Translational Medicine is now a fundamental enabler to fully exploit information contained in diverse sources of big data sets such as “omics” data, as illustrated by deep characterizations of the genome, transcriptome, proteome, metabolome, microbiome, and exposome. In this commentary, we provide an overview of ML applications in drug discovery and development, aligned with the three strategic pillars of Translational Medicine (target, patient, dose) and offer perspectives on their potential to transform the science and practice of the discipline. Opportunities for integrating ML approaches into the discipline of Pharmacometrics are discussed and will revolutionize the practice of model-informed drug discovery and development. Finally, we posit that joint efforts of Clinical Pharmacology, Bioinformatics, and Biomarker Technology experts are vital in cross-functional team settings to realize the promise of AI/ML-enabled Translational and Precision Medicine.

Biomarkers of drug-induced liver injury: a mechanistic perspective through acetaminophen hepatotoxicity

Introduction: Liver injury induced by drugs is a serious clinical problem. Many circulating biomarkers for identifying and predicting drug-induced liver injury (DILI) have been proposed.Areas covered: Biomarkers are mainly predicated on the mechanistic understanding of the underlying DILI, often in the context of acetaminophen overdose. New panels of biomarkers have emerged that are related to recovery/regeneration rather than injury following DILI. We explore the clinical relevance and limitations of these new biomarkers including recent controversies. Extracellular vesicles have also emerged as a promising vector of biomarkers, although the biological role for EVs may limit their clinical usefulness. New technological approaches for biomarker discovery are also explored.Expert opinion: Recent clinical studies have validated the efficacy of some of these new biomarkers, cytokeratin-18, macrophage colony-stimulating factor receptor, and osteopontin for DILI prognosis. Low prevalence of DILI is an inherent limitation to DILI biomarker development. Furthering mechanistic understanding of DILI and leveraging technological advances (e.g. machine learning/omics) is necessary to improve upon the newest generation of biomarkers. The integration of omics approaches with machine learning has led to novel insights in cancer research and DILI research is poised to leverage these technologies for biomarker discovery and development.

Application of Machine Learning for Tumor Growth Inhibition - Overall Survival Modeling Platform

Machine learning (ML) was used to leverage tumor growth inhibition (TGI) metrics to characterize the relationship with overall survival (OS) as a novel approach and to compare with traditional TGI‐OS modeling methods. Historical dataset from a phase III non‐small cell lung cancer study (OAK, atezolizumab vs. docetaxel, N = 668) was used. ML methods support the validity of TGI metrics in predicting OS. With lasso, the best model with TGI metrics outperforms the best model without TGI metrics. Boosting was the best linear ML method for this dataset with reduced estimation bias and lowest Brier score, suggesting better prediction accuracy. Random forest did not outperform linear ML methods despite hyperparameter optimization. Kernel machine was marginally the best nonlinear ML method for this dataset and uncovered nonlinear and interaction effects. Nonlinear ML may improve prediction by capturing nonlinear effects and covariate interactions, but its predictive performance and value need further evaluation with larger datasets.

Confounding factors in exposure-response analyses and mitigation strategies for monoclonal antibodies in oncology

Dose selection and optimization is an important topic in drug development to maximize treatment benefits for all patients. While exposure-response (E-R) analysis is a useful method to inform dose-selection strategy, in oncology, special considerations for prognostic factors are needed due to their potential to confound the E-R analysis for monoclonal antibodies. The current review focuses on 3 different approaches to mitigate the confounding effects for monoclonal antibodies in oncology: (i) Cox-proportional hazards modelling and case-matching; (ii) tumour growth inhibition-overall survival modelling; and (iii) multiple dose level study design. In the presence of confounding effects, studying multiple dose levels may be required to reveal the true E-R relationship. However, it is impractical for pivotal trials in oncology drug development programmes. Therefore, the strengths and weaknesses of the other 2 approaches are considered, and the favourable utility of tumour growth inhibition-overall survival modelling to address confounding in E-R analyses is described. In the broader scope of oncology drug development, this review discusses the downfall of the current emphasis on E-R analyses using data from single dose level trials and proposes that development programmes be designed to study more dose levels in earlier trials.

Cancer Immunotherapy Dosing: A Pharmacokinetic/Pharmacodynamic Perspective

Immune check-point inhibitors are drugs that are markedly different from other anticancer drugs because of their indirect mechanisms of antitumoral action and their apparently random effect in terms of efficacy and toxicity. This marked pharmacodynamics variability in patients calls for reconsidering to what extent approved dosing used in clinical practice are optimal or whether they should require efforts for customization in outlier patients. To better understand whether or not dosing could be an actionable item in oncology, in this review, preclinical and clinical development of immune checkpoint inhibitors are described, particularly from the angle of dose finding studies. Other issues in connection with dosing issues are developed, such as the flat dosing alternative, the putative role therapeutic drug monitoring could play, the rise of combinatorial strategies, and pharmaco-economic aspects.