Apixaban, a factor Xa inhibitor, is a direct oral anticoagulant with a well-balanced elimination; it is eliminated evenly via feces, urine (with no active secretion), and as metabolites after oral administration. The common understanding is that biliary secretion and enterohepatic circulation (EHC) of apixaban are limited in humans, and that fecal excretion may be attributable to intestinal secretion. However, a decrease in apixaban blood concentration with activated charcoal coadministration in humans suggests possible involvement of EHC. This study aimed to evaluate the contribution of biliary excretion, EHC, and intestinal secretion to apixaban pharmacokinetics (PK) using a physiologically-based pharmacokinetic (PBPK) model. A top-down analysis was performed using blood concentration and mass balance data from healthy volunteers. Model parameters were optimized using the Cluster-Gauss Newton method (CGNM), followed by the bootstrap method. The model accurately described observed data and indicated moderate to high biliary secretion relative to metabolic clearance. Simulated biliary secretion into the duodenum well predicted the biliary secretion data in humans (< 1% of dose at 8 h post-dose). Virtual knockout of EHC resulted in a shortened half-life from 8.7 to 2.9 h, and 17% and 55% decrease in area under the concentration curve (AUC) and fecal excretion after intravenous dosing, respectively, confirming the significant contribution of biliary excretion and EHC. The model also accurately described apixaban PK with activated charcoal coadministration at 2 or 6 h post-dose. Although further experimental validation (e.g., sandwich-cultured hepatocytes) would strengthen these findings, our study demonstrates that biliary secretion and EHC play a substantial role in apixaban elimination and disposition in humans.

Model-based meta-analysis (MBMA) informs key drug development decisions by integrating data, published or unpublished, from multiple studies. Due to these various sources of information and the use of summary-level data (e.g., mean responses over treatment arms or percent responders), MBMA models require careful implementation. This tutorial provides a comprehensive guide for conducting an MBMA with MonolixSuite, focusing on longitudinal continuous and categorical data. Two case studies are presented: the first examining naproxen in osteoarthritis and the second evaluating canakinumab compared to existing treatments in rheumatoid arthritis. The tutorial explains the process of model building and handling study heterogeneity in Monolix, including how to include between-study variability and between-treatment-arm variability. It also shows how to apply appropriate weighting due to the use of summary-level data. For model evaluation, the tutorial demonstrates the use of automatically generated diagnostic plots, statistical tests, and convergence assessment tools. Furthermore, it illustrates how to use the model in Simulx to support the decision-making process, such as by simulating clinical trials. This step-by-step guidance offers practical insights for leveraging MBMA in model-informed drug development.

We present a framework for maximum likelihood analysis on count observations that begin high and quickly drop to zero, for example, from hollow fiber drug comparison studies. This simulation study focuses on treating observed counts as Poisson or normally distributed for the purpose of estimating infection rebound after effective treatment. CFU profiles were simulated from inoculation to 96 h post-treatment. The PK-PD link was an Emax inhibitory model. Random parameters were pathogen growth and natural decay rates, drug concentration for half-maximal effect, and drug pathogen kill rate. Other parameters, including PK, were fixed. Parameters were adjusted to attain 67% efficacy at 24 h. Random parameter values were optimized for profiles observed at 24, 48, 72, and 96 h assuming each of four probability assumptions: (1) all CFU measurements were Poisson distributed (truth); (2) CFU < 128 were Poisson, higher values were normally distributed; (3) all observations were normally distributed; and (4) observations were normally distributed but CFU < 10 were censored. CFU-time profiles were re-simulated using the optimized parameter densities. Rebound percentage (CFU ≥ 10 at 24 h post-treatment) was best predicted using strategy 2, above. For limited periodically collected time series count data that quickly fall to 0, the true proportion reaching 0 (lack of rebound) was best modeled by assuming Poisson distribution at low counts. At higher counts (≥ 128), assuming normality is reasonable. Censoring observations leads to biased models.

Non-Hodgkin lymphoma (NHL) is the fifth most common malignancy and accounts for 5% of all cancers in the US, with the largest proportion being B-cell CD20 positive NHL. Odronextamab is a CD20xCD3 IgG4 bispecific T-cell-engaging monoclonal antibody under development for the treatment of relapsed or refractory (R/R) B-NHL. The objectives of this analysis were to characterize the pharmacokinetics (PK) of odronextamab in adult patients, and elucidate sources and correlates of variability. PK data of 507 patients with R/R B-NHL from ELM-1 (NCT02290951, Phase I; n = 167) and ELM-2 (NCT03888105, Phase II; n = 340) were analyzed. Odronextamab concentration-time profiles following intravenous administration of 0.03 mg to 320 mg doses were described by a bi-exponential decline with parallel linear (first-order) and non-linear (Michaelis-Menten) elimination processes. The modified Michaelis-Menten or target-mediated elimination was not only concentration-dependent but also time-dependent. A reduction in target-mediated clearance over time suggests a reduction in target abundance to a larger extent than associated with concentration alone, which is consistent with the treatment-induced depletion of the B cells observed in patients who underwent assessment. Linear clearance (CL) and steady-state volume of distribution were 0.189 L/day and 9.41 L, respectively. Target-mediated clearance was ~5 L/day at baseline, with an asymptote of ~0.03 L/day at steady state. With the largest covariate effect on odronextamab exposure, baseline body weight was directly correlated with CL and volume of distribution, albumin was inversely correlated with CL and volume of distribution, and baseline interleukin-10 was inversely correlated with CL.

We developed a comprehensive, mechanistic model of human copper metabolism to support biomarker qualification for VTX-801, an adeno-associated vector-based gene therapy which is being developed to restore the mutated ATP7B copper transporter gene in Wilson disease (WD). The model integrates physiological copper kinetics with pathophysiological features of WD by distinguishing between ceruloplasmin-bound and non-ceruloplasmin-bound copper (NCC), and by explicitly incorporating ATP7B-dependent processes: biliary excretion and ceruloplasmin loading of copper. Literature-derived time-activity data from healthy subjects, heterozygous carriers, and WD patients, as well as clinical radiocopper data in plasma and feces from a pilot study in non-WD subjects, were used for model development and validation. VTX-801's dose-response was quantified in WD mouse models using ceruloplasmin oxidase activity measurement and Cu fecal excretion. This enabled derivation of activity factors (AFs) corresponding to restored ATP7B function, with 15% and 40% selected as minimal and optimal efficacy targets. Simulations linked AFs to clinical biomarkers, demonstrating that the 48/2-h plasma radioactivity ratio can effectively differentiate VTX-801 responders from non-responders, providing a decision criterion to safely withdraw standard treatment in participants of a phase 1/2 trial. To broaden applicability beyond radiotracer studies, we simulated "cold" copper kinetics under steady-state conditions, deriving expected values for plasma copper, NCC, urinary copper excretion, and relative exchangeable copper (REC). These simulations suggest that REC may also serve as a suitable and simpler to implement, non-radioactive biomarker for ATP7B gene therapy. This model provides a robust quantitative framework to assess copper-related biomarkers in WD and their response to treatment in silico. Trial Registration: EudraCT number: 2019-001157-13.

Pediatric extrapolation strategies issued by health authorities have streamlined pediatric drug development and reduced the unnecessary burden of conducting pediatric clinical studies. In line with these strategies, physiologically based pharmacokinetic (PBPK) models have been utilized extensively for initial dosing regimen and sampling timepoint selection for pediatric studies, as well as dose validation throughout pediatric drug development. Here, the status and challenges of PBPK modeling in pediatric drug development have been summarized by the IQ Pediatric PBPK Working Group. Our work reviews current practices for pediatric PBPK modeling across various therapeutic areas. To enable best practice, we propose an optimized workflow for pediatric PBPK modeling recommendations. Two selected key pediatric PBPK case examples are also described, where modeling impacted the drug label extension to pediatric patients. Moreover, we analyze the current gaps and challenges in our understanding of drug absorption, distribution, metabolism, and elimination in pediatric PBPK model development. Since neonates are the least studied and the most medically fragile, the depth of our understanding of their rapidly evolving physiological processes is limited and so there exist significant modeling gaps which we summarize here. Finally, we provide recommendations, including building a public data repository, leveraging real-world data, and implementing microdose studies for addressing pediatric PBPK modeling challenges.

Analyzing exposure-response (E-R) relationships for time-to-event (TTE) endpoints presents challenges due to the inherent time-dependent nature of the data. Some authors address these difficulties by using a fixed timepoint approach, where exposure is assessed at a predetermined time rather than dynamically over time. (e.g., initial exposure or last exposure). The aim of the current work is to compare the use of time-static and time-varying metrics to assess the E-R relationship through simulations. PK exposures were simulated from a one-compartment model and TTE data from a parametric proportional hazard model, involving the weekly average PK concentration as a time-varying covariate. Several scenarios were considered to handle the type of dosing (fixed or adaptive), the accumulation of the drug (low or strong), the type of event (efficacy, safety or independent), and the timing of the event onset (early or late). Wald tests on the exposure effect parameter were performed to assess the significance of the E-R relationship. For each simulation scenario, the type-I error and the power of the Wald tests were reported, revealing that no time-static metric consistently produced reliable results across all conditions. In order to ensure adequate statistical properties, we recommend using time-varying exposure, which shows good performance across all scenarios.

Statins are frequently prescribed for hyperlipidemia, a common comorbidity in patients with obesity and/or metabolic dysfunction-associated steatohepatitis (MASH). However, limited knowledge exists on how MASH may alter statin disposition within hepatocytes where the statin target, 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase, is located. This study used a physiologically based pharmacokinetic (PBPK)/permeability-limited multicompartment liver (PerMCL) framework, incorporating zonal transporter and drug-metabolizing enzyme data. Systemic and hepatocellular concentrations of pravastatin, rosuvastatin, and atorvastatin were simulated in Healthy Volunteers (HV), Obese, Morbidly Obese, and MASH virtual populations with the Simcyp Simulator. A pharmacodynamic model in Simcyp Designer was then used to simulate alterations in rosuvastatin cholesterol-lowering efficacy between these populations. Hepatic transport and metabolism pathways were verified against clinical data. Organic anion transporting polypeptide (OATP)1B model uptake pathways were verified using genotype and drug-drug interaction data. Atorvastatin metabolism pathways were verified using metabolite data. Steady-state plasma and zonal hepatocellular concentration-time profiles for each statin were simulated across virtual populations of 100 individuals aged 40-65 years. Simulations predicted > 70% increases in maximal total plasma concentrations and area under the curve for pravastatin and rosuvastatin in MASH compared to HV, with changes in these parameters for atorvastatin simulated to increase > 250%. In MASH, unbound hepatocellular exposure increased by up to 127% in the periportal region for atorvastatin and decreased by up to 55% in the pericentral region for rosuvastatin. The pharmacodynamic model simulated decreased rosuvastatin cholesterol-lowering efficacy in MASH compared with Obese, which could be compensated for with a 50% increase in dose according to exploratory simulations.

A common challenge in conducting phase 1 studies that assess the impact of organ impairment on the pharmacokinetics of a drug is the recruitment of a demographically matched control group. The work presented here evaluated alternative approaches for generating control groups in these studies. Available phase 1 data from the upadacitinib and elagolix clinical programs were leveraged as case studies. A statistical matching approach and a population pharmacokinetic model-based approach were evaluated retrospectively for these programs' hepatic and renal impairment clinical studies. Geometric mean ratios of logarithmically transformed C and AUC were used to compare exposure in organ impairment groups to respective matched or virtual control groups. In the statistical matching approach, the genetic matching algorithm using Mahalanobis distance showed that external control groups were adequately demographically balanced across all impairment groups of the study except for age. A 3:1 k-match approach minimized the prediction error between matched and reference in-study results for both case studies, resulting in differences in geometric mean ratios ranging from -19% to 3% and -27% to 40% for upadacitinib and elagolix, respectively, compared to in-study controls. Similarly, the population pharmacokinetic approach used models developed from phase 1 data in healthy participants and found that the results were generally comparable to the in-study results, with differences in geometric mean ratios ranging from -30% to 17% and -24% to 41% for upadacitinib and elagolix, respectively. These analyses demonstrate that both approaches may be viable alternatives to assess the impact of organ impairment on pharmacokinetics.

Rituximab (RTX), an anti-CD20 monoclonal antibody, has been used to treat autoimmune diseases such as rheumatoid arthritis (RA). However, variability in therapeutic response to RTX remains a challenge. Here, a systems model is developed to mimic B cell differentiation leading to antibody-secreting cells (ASCs), including plasmablasts (PBs) and plasma cells (PCs). The model features the localization of B cell subsets in the bone marrow and secondary lymphoid organs and incorporates the internalization process of the CD20-RTX complex. To reproduce clinical data from patients with RA receiving RTX and glucocorticoids, pharmacokinetic models for the drugs were built and respective pharmacodynamic profiles of CD19 and CD20 cells and PBs were well captured by optimizing model parameters, which were estimated with good precision. As ASCs are the primary source of pathogenic autoantibodies in RA, the extent and duration of ASC depletion were hypothesized as drivers of therapeutic response to RTX. Global sensitivity analyses identified the CD20-RTX binding affinity and elimination rate constant (i.e., Fcγ-mediated degradation, internalization) as major determinants of both CD19 cells and ASCs. The influence of baseline PBs and PCs on ASCs was also suggested, providing potential mechanisms underlying responder and non-responder variability. The model accurately reproduced the temporal changes in CD19 cells after combination treatment with RTX and glucocorticoids suggesting successful model validation. This study provides a mechanistic framework and insights into key drivers of responses to CD20-depletion treatment using B cell dynamics as an indirect biomarker of clinical endpoints, which might ultimately improve therapeutic outcomes.

Physiologically based pharmacokinetic (PBPK) modeling of placental drug transfer is an evolving tool for predicting fetal drug exposure. In this study, a pregnancy-specific metformin PBPK model was developed, and the following four approaches were evaluated to predict metformin placental transfer: (1) perfusion-limited model, and permeability-limited models using (2) ex vivo cotyledon open system apparent clearance, (3) ex vivo cotyledon closed system data fit to a three-compartment model to estimate clearance, and (4) active transport kinetics and passive clearance. Simulated metformin maternal plasma concentrations (MPCs) and umbilical cord venous plasma concentrations (UCCs) were compared to observed in vivo data from subjects with gestational diabetes mellitus taking metformin 500 mg twice daily. Model selection criteria were determined by the percentage of observed clinical data falling within the 5th to 95th percentiles of the simulated population. Among the approaches, the model that included passive permeability and in vitro intrinsic transporter clearances (Approach 4) best described placental metformin transfer, with 92% of UCCs falling within the 5th to 95th percentiles of the simulated population. Furthermore, maternal uptake transport had the largest influence on predicted UCCs. A two-fold increase in maternal uptake transport increased the predicted population mean UCC C by 97%, whereas a 0.5-fold decrease resulted in a 49% decrease in UCC C. This refined PBPK model offers a valuable framework for predicting placental transfer and fetal exposure of metformin when placental transporters are altered throughout pregnancy and/or with pathological conditions.

Patients with metabolic dysfunction-associated steatotic liver disease (MASLD) may exhibit altered pharmacokinetics (PK) and pharmacodynamics (PD) of drugs compared with healthy populations. However, no physiologically based pharmacokinetic/pharmacodynamic (PBPK/PD) model has been specifically developed for MASLD. Acetaminophen (APAP), a widely used analgesic, was selected to develop a PBPK/PD model predicting PK/PD changes of APAP and its metabolites in MASLD-related populations. Based on a comprehensive review of published APAP PK studies and examination of existing PBPK models, a simultaneous parent-metabolite PBPK model for APAP was developed and optimized in healthy people. The model simulated the dynamics of APAP and its five major metabolites: APAP-glucuronide (APAP-glu), APAP-sulfate (APAP-sul), N-acetyl-p-benzoquinone imine (NAPQI), APAP-cysteine (APAP-cys), and APAP-mercapturate (APAP-merc). The validated model was expanded to MASLD-related populations, including overweight, obese, nonalcoholic fatty liver disease (NAFLD), nonalcoholic steatohepatitis (NASH), and cirrhosis with different severities. Finally, a PD model was integrated to correlate APAP's PK with pain relief scores. The PBPK model reproduced published clinical PK data for APAP and its metabolites in healthy and MASLD-related populations. At therapeutic doses, the toxic NAPQI remained at very low levels. APAP's pain relief efficacy was retained, but onset time may change in MASLD-related populations. This PBPK/PD approach provides a strategy for projecting drug exposure in MASLD-related populations, even without specific PK or PD data. It highlights modeling's utility for personalized medicine in MASLD patients and MASLD treatment drug development.

A twice-daily administration of oral selumetinib (SLT) in the fasted state is the only approved pharmaceutical option for treating inoperable neurofibromatosis type I (NF-1) and plexiform neurofibromas (PN). In children, exposure to SLT is highly variable, and fasting presents a substantial burden. Therapeutic drug monitoring and pharmacokinetic modeling can support individualized therapy accompanied by a more rational alimentary routine. Twenty-eight children diagnosed with inoperable NF-1 or PN were recruited at a major pediatric oncological center. Twenty-two patients donated 156 blood samples in steady state for nonparametric population pharmacokinetic modeling. An equation was developed experimentally for estimating model error. Eleven three-compartment models were compared in terms of statistical performance. Monte Carlo simulations were performed to validate a limited external model using six additional patients and to compare the trough-to-peak SLT concentration ratios simulated for various dosing regimens to develop better control over exposure. A pharmacokinetic model that included total body weight as a covariate provided the best fit between predicted and observed concentrations (r = 0.994) and the best performance statistics. In the first Monte Carlo simulation, measured concentrations fell within the 0.01%-95% (median: 19.7%) quantiles of the simulated ranges. The second simulation revealed that 6-h (q6h), 8-h (q8h), and 12-h (q12h) dosing intervals would yield comparable trough-to-peak concentration ratios, with medians of 0.126 (range: 0.001-0.335), 0.104 (0.000-0.306), and 0.065 (0.000-0.279), respectively. The nonparametric population model provides efficient priors for making individual predictions of SLT concentrations. The simulation did not reveal any disadvantages of q6h or q8h dosing.

Atopic dermatitis (AD) clinical trials exhibit substantial placebo response variability, confounding efficacy assessments of novel therapies. Traditional meta-analyses have identified potential contributors to this variability but rely on single time-point estimates, which fail to account for dynamic, longitudinal response patterns across trials. To overcome this limitation, we developed a model-based meta-analysis (MBMA) framework that characterizes time-course projections of EASI-75 placebo responses while accounting for key covariates. A systematic literature review identified 40 moderate-to-severe AD trials (18 Phase 2, 22 Phase 3), encompassing 4827 patients, suitable for longitudinal modeling. Modeling results highlighted concomitant therapy as a significant driver of placebo response, with trials permitting topical corticosteroids (TCS) demonstrating a 1.8-fold increase in EASI-75 placebo rates compared to trials without concomitant therapy. Additionally, baseline disease severity of the study population, as reflected by the mean baseline EASI score, was inversely associated with placebo response; each 1-point increase in baseline EASI reduced EASI-75 placebo rates at Weeks 12 and 16 by 0.96-fold. Time-course modeling suggested that placebo responses plateaued by Week 12, with EASI-75 outcomes at Week 12 capturing 94% of the projected response at Week 16. Overall, this MBMA framework provides quantitative guidance to optimize clinical trial design, refine power calculations, and improve the differentiation between therapeutic and placebo effects in AD drug development.

With the recent advances in machine learning (ML) and artificial intelligence (AI), data-driven modeling approaches for pharmacokinetics (PK) and pharmacodynamics (PD) have gained popularity due to their versatility in diverse settings and reduced reliance on prior assumptions. However, most of the ML methods ignore the hidden dynamics behind the data, lacking interpretability. This study investigated the applicability of neural controlled differential equation (NCDE), a novel ML method that is suitable for data-driven modeling of PK and PD profiles, especially in the setting of multiple dosing. We demonstrated that NCDE was capable of combining differential-equation-based dynamics with data-driven characteristics, flexibly incorporating various types of inputs, and embedding discontinuous dynamics. Moreover, a direct correspondence was identified between the learned dynamics of NCDE and the dynamics behind the data, which highlights the intrinsic interpretability of NCDE. Additionally, the influence of important hyperparameters was systematically investigated, and it was found that L1 regularization and the AdaMax optimizer were useful for stabilizing the training process and leading to a generalizable NCDE model. Together, these findings demonstrate the accuracy, generalizability, and interpretability of NCDE, indicating that NCDE is a reliable method for further application. In the future, NCDE may further facilitate PK and PD prediction in general.

This work aimed to develop an appropriate model to evaluate the exposure-response relationship (ERR) for monthly migraine days (MMD) in atogepant's key migraine prevention clinical trials to support dose selection. The ERR between atogepant concentration and MMD over time was analyzed utilizing data from one phase 2b/3 and three phase 3 studies in patients with episodic or chronic migraine (EM/CM). Several distribution models were evaluated for placebo data, whereas two modified normal distributions were introduced enabling bounded MMD modeling. Exposure metrics and shapes for ERR were tested for the most suitable distribution. Stepwise covariate search, visual predictive checks, and plots of model-predicted MMD over the range of exposure metrics were utilized in model development, evaluation, and selection. The final MMD exposure-response model was able to model patients with EM/CM simultaneously and was based on a modified normal distribution with E ERR on C. The model adequately described the observed data over time. Due to the E relationship, MMD at Week 9-12 plateaued around their model-based atogepant C-EC of 3.71 nM, which is similar to most C exposures seen at the 10 mg once-daily regimen. All approved atogepant dosages for EM/CM achieve effective concentrations to inhibit the calcitonin gene-peptide receptor by 90%. Patients who have been failed by conventional oral migraine preventive treatments or patients with a higher baseline MMD may require a longer treatment period to reach atogepant's maximal effect. No significant difference in efficacy was evident in patients exposed to prior oral migraine preventives compared to treatment-naïve patients.

Infliximab, a monoclonal antibody used for immune-mediated diseases, shows high interpatient pharmacokinetic variability. Prolonged exposure increases the risk of adverse effects and costs, making dose personalization essential to balance safety, efficacy, and cost-effectiveness. Population pharmacokinetic models support individualized dosing, but different models may predict varying drug exposure for the same patient. This study aims to identify compatible models for each patient and assess the impact of model selection on dosing. This retrospective study included adult Crohn's disease patients receiving infliximab. Published pharmacokinetic models were screened. Model-patient compatibility was evaluated using Multivariate Exact Discrepancy through 100,000 Monte Carlo simulations. The Metropolis-Hastings algorithm generated individual parameter distributions. For each model-patient pair, the median and 90% confidence interval of the dose required to achieve a target exposure of 2079 mg*day/L were computed. Sixteen models were tested. No model was compatible with all patients. Dosing was calculated only for compatible pairs. The average median dose was 9.25 mg/kg, with an average imprecision of 6.63 mg/kg. The highest median dose reached 23.21 mg/kg, reflecting inter-model differences, while the greatest imprecision (25.69 mg/kg) stemmed from patient variability. This concentration-based method personalizes dosing via pharmacokinetic profiling. Patients can be classified into three groups: (1) those for whom all models provide similar recommendations, indicating high reliability across models; (2) those incompatible with all models, for whom the posology recommended by the manufacturer should be prioritized; and (3) those for whom some models are compatible but intensified therapeutic drug monitoring is required.

Asundexian is a potent, selective, and reversible inhibitor of activated clotting Factor XI currently under development for secondary prevention of recurrent ischemic stroke in the ongoing Phase III OCEANIC-STROKE study (NCT05686070). Here, we report the development of a population pharmacokinetic (popPK) model for asundexian. Plasma concentration data were available from 2914 participants enrolled in nine Phase I and II studies of asundexian. The pharmacokinetics (PK) of asundexian were well described by the popPK model. Within the investigated dose range of asundexian 10-100 mg once daily, the PK of asundexian was dose-proportional. The systemic apparent clearance (CL/F) of asundexian was estimated to be 2.25 L/h and the central volume of distribution (V/F) was 35.3 L. Body weight, age, sex, concomitant administration of cytochrome P450 3A4 (CYP3A4) inhibitors, and renal function were identified as statistically significant covariates influencing the PK of asundexian. After accounting for differences in the distribution of these covariates, the PK of asundexian was comparable in healthy participants and participants at risk for thromboembolic/cardiovascular events. Similarly, no significant differences in PK were noted among participants with atrial fibrillation, ischemic stroke, or acute myocardial infarction. No clinically relevant covariates were identified that would warrant dose adjustments in various special populations of interest, including those defined by body weight, age, sex, and renal function, for the prevention of secondary ischemic strokes.

The tyrosine kinase inhibitor, nintedanib, reduces the rate of decline in forced vital capacity (FVC) in a comparable manner in patients with idiopathic pulmonary fibrosis (IPF), other forms of progressive pulmonary fibrosis (PPF), and systemic sclerosis-associated ILD (SSc-ILD). The recommended dose of nintedanib in all indications is 150 mg twice daily (BID). Data from Phase II and III trials in IPF, PPF, and SSc-ILD were incorporated into a meta-model to holistically investigate the relationship between nintedanib exposure and efficacy. Using data from 2642 patients with IPF, PPF, or SSc-ILD treated with nintedanib doses ranging from 50 to 150 mg BID, disease progression models with a maximum drug effect on the annual rate of change in absolute FVC (i.e., mL), FVC %predicted, and FVC Z-score were developed. The estimated plasma concentration producing 50% of the maximum drug effect (EC) ranged from 6.21 to 10.4 nM (with respect to nintedanib trough concentration) across the explored FVC-based endpoints. While the disease progression for absolute FVC (mL), FVC %predicted, and FVC Z-score was different between IPF and PPF patients compared to SSc-ILD patients, the relative treatment effect of nintedanib, described by a disease-modifying E effect, was comparable across indications. The majority of patients achieve exposure levels at or exceeding the EC with the approved starting dose of 150 mg BID.