The objective of this research was to develop a physiologically based pharmacokinetic (PBPK) model for AAV-based gene therapy, which can capture the nonlinearity observed in both viral vector and transgene product pharmacokinetics (PK) across a wide range of doses, while accounting for the effect of immunogenicity. To develop the PBPK model, previously published PK data generated in mice using AAV8 vector containing the transgene for a non-binding monoclonal antibody was used. Immunocompetent mice were administered with AAV at a wide range of doses (1E8, 1E9, 5E9, 1E10, 2E10, 1E11, 2E11, 1E12, and 1E13vg per mouse), and the PK of transgene and transgene product (i.e., antibody) in plasma and/or tissue was collected. The nonlinearity in transgene product concentrations was characterized using a saturable production process and a concentration-dependent antibody elimination rate was used to characterize the effect of anti-drug antibody (ADA) on transgene product. The model successfully described the PK of both the vector and the transgene product across all dose levels and accurately captured the sigmoidal dose-exposure–response relationship for AAV. Notably, the model described a dose-dependent ADA response, with the high dose group exhibiting an earlier onset and faster rate of transgene product elimination. Lower dose group showed delayed onset and minimal ADA-mediated elimination of transgene product. Overall, the PBPK model presented here effectively characterizes vector and transgene product kinetics in mice and demonstrates utility in preclinical-to-clinical translation and dose optimization of AAV-based gene therapies.
Immunogenicity assessment for therapeutic bispecific antibodies often requires domain-specific characterization, including evaluation of neutralizing antibody (NAb) responses to individual domains. For our beta Nerve Growth Factor (β-NGF)–Tumor Necrosis Factor Receptor 2 (TNFR2) bispecific antibody, anti-drug antibody (ADA) mapping revealed a predominant immune response to the β-NGF binding domain, leading us to explore the correlation of β-NGF domain–specific NAbs with clinical impact. Because drug or target in the sample can interfere with NAb detection, the development of a NAb assay format with sufficient drug and target tolerance is required. To that end, we developed a domain-specific NAb assay, integrating a modified biotin-drug extraction and acid dissociation (BEAD) procedure that includes target blocking to eliminate false-positives. We termed this assay format the “BEAD with target blocking” (BEAD-TB) method. The workflow features sequential acid dissociation to liberate drug-bound NAbs, followed by selective enrichment of β-NGF domain–specific Nabs. We conducted a comprehensive assessment to ensure adequate sensitivity, selectivity, drug tolerance (up to 10 μg/mL), target tolerance (up to 10 ng/mL) and robustness for the assay to accurately detect β-NGF binding domian–specific NAbs in ADA-positive patients. Our domain specific NAb assay demonstrated a robust, innovative solution for detecting neutralizing responses in the context of complex bispecific biologics, leading to reliable immunogenicity evaluation in drug development.
The prediction of bioavailability (BA) and bioequivalence (BE) for immediate-release (IR) solid oral dosage forms is critical during development and post-approval phases of drug products. ICH M13A provides the option of using in vitro tests, such as disintegration and dissolution in biorelevant media, pilot studies, and modeling to justify not conducting in vivo fed BE studies. The gastrointestinal (GI) tract environment changes significantly after food intake, including variations in pH, enzyme activity, and fluid viscosity that influence drug dissolution and absorption. Existing biorelevant dissolution media (FeSSIF, FeSSGF etc.) simulate human GI fluid composition but do not account for food-induced viscosity changes. This study explored impacts of food-induced fluid viscosity (pH 1.2 buffer containing hydroxypropyl methylcellulose (HPMC)) on disintegration and dissolution profiles of midodrine HCl tablets. Using USP Apparatus II and disintegration testing, we evaluated dissolution and disintegration of five approved generic midodrine HCl tablets with demonstrated BE. High-viscosity media significantly delayed tablet disintegration and drug dissolution compared to low-viscosity fasting conditions. We modified dissolution methods to closely simulate fed state GI conditions by varying agitation speed, prolonging dissolution time, adding medium stagewise to simulate dilution of gastric fluid, and adding water to mimic co-administered drinks. Results showed increased dissolution with higher agitation speeds and continued dissolution in viscous media over extended time. Stagewise dissolution in medium with reduced viscosity to mimic gastric secretion did not accelerate drug release compared to viscous media. In contrast, co-administered water enabled tablet swelling and disintegration, leading to complete drug release within 15 min, which is more consistent with the in vivo PK data under fed conditions. These findings may help refine in vitro dissolution conditions to better mimic fed-state.
Cannabidiol (CBD) use continues to increase, particularly among older adults. However, knowledge of CBD disposition in this ever-growing and understudied population, along with the relative contributions of cytochrome P450 (CYP) and uridine 5′-diphospho-glucuronosyltransferase (UGT) enzymes to CBD metabolism, remain limited. Inclusion of UGT activators (e.g., alamethicin, MgCl2) may alter in vitro estimates of enzyme contributions. CBD has also been reported to produce time-dependent inhibition (TDI) of certain CYPs, potentially affecting its own disposition. We combined in vitro data with physiologically based pharmacokinetic (PBPK) modeling to further characterize TDI and reversible CYP inhibition by CBD, quantify the contributions of CYPs and UGTs to CBD metabolism, and determine in vitro assay conditions that best reflect in vivo CBD metabolism. PBPK modeling was next used to extrapolate CBD exposure from young (18–64 years) to older (65–98 years) adults. CBD exhibited TDI only toward CYP1A2, which had a negligible effect on CBD metabolism. Inclusion of alamethicin + MgCl2 increased the apparent contribution by UGTs to CBD metabolism, leading to an overestimate of UGT-mediated metabolism. PBPK model validation using drug-drug interaction studies supported contributions of 86
Colorectal cancer (CRC) is one of the most common malignant neoplasms worldwide, being the third most frequently diagnosed cancer and the second leading cause of death. Due to the high incidence of CRC cases, its biological complexity, and high morbidity and mortality, the development of more effective and less aggressive therapeutic approaches is an urgent need. Extracellular vesicles (EVs), especially EXs, are a very viable alternative as a nanocarrier for anticancer molecules for the treatment of CRC, as EXs are biologically relevant EVs with considerable therapeutic potential. EXs are nanoscale vesicles approximately 30–200 nm in diameter, identical in composition to the membrane of their parent cells, and carry a bioactive cargo of proteins, lipids, nucleic acids, and glycoconjugates. Beyond their physiological functions, EXs are gaining prominence as next-generation drug delivery platforms. Their ability to improve the efficacy of antitumor therapies, potentially reducing adverse effects associated with conventional chemotherapy and radiotherapy, makes them very promising. Furthermore, functionalizing the surface with targeting ligands, such as MUC1 aptamers, AS1411, or iRGD peptides, can further increase its specificity for CRC cells. Advances in drug delivery techniques, including passive incubation, electroporation, and chemical modification, have enabled the incorporation of conventional chemotherapeutic agents such as doxorubicin, 5-fluorouracil, and SN-38, as well as nucleic acid-based therapies, including microRNA and siRNA. Due to the limitations and invasive nature of current treatments for CRC, new studies to obtain innovative and targeted therapeutic strategies are essential to reduce harmful effects. Therefore, exosome-based drug delivery systems represent a promising and clinically relevant avenue for future cancer therapy.
To address the challenges associated with keloid pharmacotherapy, we developed a melittin (MEL)-modified paclitaxel (PTX) liposomal delivery system as a reconstitutable lyophilized formulation intended for subcutaneous injection. MEL, an amphipathic membrane-active peptide, was incorporated as a functional component to enhance liposome-cell interactions and facilitate intracellular delivery of PTX. This design aimed to improve the in vitro anti-keloid activity of PTX liposomes while providing a lyophilized injectable formulation suitable for reconstitution before use. After MEL modification, the zeta potential shifted upward with increasing MEL content, supporting MEL surface modification of PTX liposomes through electrostatic interactions. Moreover, freeze-dried MEL-modified PTX liposomes showed comparable particle size, zeta potential, and release behavior to the original liposomes. Although the encapsulation efficiency (EE
The bioavailability of Rifampicin (RIF) in fixed-dose combination (FDC) with Isoniazid (INH) poses a significant challenge in tuberculosis therapy due to RIF's rapid degradation under gastric conditions when co-administered with INH, resulting in subtherapeutic plasma levels and compromised therapeutic efficacy. To overcome this issue, we developed a novel Selective Laser Sintering (SLS)-mediated 3D printing approach using SmartEx QD 100 and Eudragit L 100–55 polymers. This advanced technique enabled the fabrication of compartmentalized FDC 3D printed bilayer tablets, designed to release INH in the stomach and RIF in the intestine, thereby preserving RIF stability while maintaining synergistic drug action. The bilayer tablets met pharmacopeial standards for dimensional accuracy, weight variation, friability, and hardness, with adequate tensile and shear strength for ease of handling. Micro-CT and SEM–EDX analysis confirmed uniform pore formation, drug distribution, and successful interlayer fusion. In vitro dissolution demonstrated complete release of INH within 45 min in the gastric region, while RIF exhibited minimal release in the gastric environment and sustained release in the intestinal region. Pharmacokinetic studies in New Zealand white rabbits indicated INH and RIF achieved respective Cmax values of 432.00 ± 109.16 ng/mL and 84.57 ± 43.79 ng/mL, with distinct Tmax profiles supporting compartmentalized release. This study concludes with the transformative potential of SLS-mediated 3D printing in developing personalized, stable, and effective anti-tuberculosis therapies. By ensuring precise compartmentalized drug delivery, this innovation enhances treatment outcomes, improves patient compliance, and offers a promising platform for future advancements in complex drug delivery systems.
Currently, the manufacture of injectable drug products relies on a quality-by-test approach, where product quality is evaluated at fixed manufacturing steps. The introduction of the quality-by-design approach, supported on process analytical technologies (PAT) is changing this paradigm. The US Food and Drug Administration established the foundations for the application of PAT in real-time control of the drug manufacturing process. One of the most versatile PAT tools is the Raman spectroscopy. Nevertheless, the application of this technique to injectables is still underexplored. This work intends to expand the use of Raman spectroscopy as a PAT tool for injectable drug products. A Raman spectroscopy method was developed for the assay of paracetamol in paracetamol infusion solutions, using a PLS (partial least squares) algorithm. The method is specific, with a root mean square error of calibration (RMSEC), cross-validation (RMSECV) and prediction (RMSEP) of 1.067
Uncertainty is an integral facet to the biotechnology and life sciences industry. From early target drug discovery to regulatory approval and commercialization, investment decisions are made under conditions of incomplete knowledge and asymmetric information. Indices of probability, particularly probability of technical and regulatory success, probability of clinical success, and risk-adjusted net present value, have become foundational tools for structuring these uncertainties within portfolio decision-making frameworks. However, their interpretation and application often lack conceptual precision and contextual nuance. This commentary examines the use of probability indices in describing and managing portfolio risk in biotechnology and life sciences. It reflects on their role in assigning risk to new drug targets, novel therapeutic modalities, and stage-specific development programs.
Generic and biosimilar medicines are foundational to sustainable healthcare, enabling broader patient access globally through high-quality, cost-effective alternatives. However, their market access is hindered by regulatory fragmentation across major authorities, where regulatory expectations are not always clearly aligned across jurisdictions or transparently justified, introducing additional complexity and potential delays in development and approval. This study integrates structured interviews with representatives from generic and biosimilar companies, alongside comprehensive literature and regulatory database reviews, to assess the impact of these regulatory inconsistencies. Analysis of 52 pharmaceutical products across four categories identified several key challenges including duplicated clinical studies across jurisdictions, inconsistent reference product (RP) requirements, and divergent manufacturing standards, potentially contributing to increased development costs and delayed market entry. These challenges disproportionately affect complex generics and orphan drugs, affecting global patient access to critical medicines. Despite promising international harmonization initiatives, substantial obstacles remain. The findings highlight key areas for greater regulatory alignment, including acceptance of foreign RPs, harmonization of technical standards, streamlined development pathways, and enhanced collaboration among regulatory agencies. Such measures are critical to facilitate global development of generic and biosimilar medicines, ensuring timely patient access without compromising safety or efficacy.
A favorable taste is highly desirable for oral formulations, as an unpleasant bitter taste can severely compromise patient compliance. Acemetacin (ACM) is a bitter and poorly soluble drug, posing a significant challenge for oral administration. In this study, Purolite A430MR (A430MR) was employed to mask the bitterness of ACM via a precipitation method, forming ACM-A430MR resin composites (ARCs). A 1:1 weight ratio of ACM to A430MR was identified as the optimized formulation, achieving favorable drug loading and process yield. The ARCs1:1 was characterized by SEM–EDS, PXRD, DSC, FTIR, and molecular docking to investigate their structural and physicochemical properties and were further evaluated by in vitro release and human sensory testing. The formulation exhibited a drug load of 0.95 mg/mg and a process yield of 65.8
Quality control is fundamental to drug development, with critical quality attributes (CQAs) serving as key indicators of product performance. However, missing or incomplete CQA data remains a persistent challenge throughout pharmaceutical research and manufacturing. In this study, we evaluated the utility of machine learning techniques (MLT) for CQA imputation across five pharmaceutical datasets comprising 14 CQAs. Five MLT were benchmarked against conventional imputation methods to assess their imputation performance. In several datasets and CQAs, MLT achieved imputation accuracies exceeding R2 = 0.80, including under conditions of up to 30
Population pharmacokinetic (popPK) models are commonly developed using ordinary differential equations (ODEs) to describe deterministic concentration–time profiles, with unexplained variability typically attributed to interindividual variability or residual error. When model misspecification is present, system-level deviations may be absorbed into these conventional variability terms, making the source and magnitude of model inadequacy difficult to assess quantitatively. Stochastic differential equations (SDEs) provide an alternative framework by introducing an explicit system-noise component into the structural model, allowing model–data mismatch to be evaluated more directly. However, historical implementation of SDE-based models in NONMEM has been technically challenging. The availability of the Fortran plug-in subroutine SDE.f90 substantially lowers this barrier and enables more practical implementation. In this work, SDE-based nonlinear mixed-effects models were evaluated as a quantitative diagnostic framework for probing popPK model misspecification. The SDE.f90 implementation was first verified using simulated one-compartment intravenous bolus datasets with stochastic process noise. Additional simulation–estimation scenarios were then conducted under intentionally misspecified structural or stochastic assumptions, including time-varying elimination, compartmental misspecification, and residual error misspecification. Across these scenarios, the estimated system-noise parameter was generally sensitive to misspecification, with larger values usually associated with greater structural or stochastic mismatch. SDE-based modeling also helped partially separate system-level variability from residual variability and, in selected settings, supported localization of misspecification to specific model components, thereby helping guide model refinement. Overall, SDE-based popPK modeling is a useful addition to the pharmacometric diagnostic toolbox, with system-noise estimates best interpreted alongside structural model evaluation, residual diagnostics, parameter behavior, and pharmacologic plausibility.
Target-mediated drug disposition (TMDD) models are often used to describe drug-receptor interactions and help derive receptor occupancy (RO), i.e., the ratio of drug-receptor complex to total receptors, which is important for understanding drug responses especially in early clinical development. A full TMDD model can be complex, and data may not always be sufficient for fitting the model. In such cases, approximations of the full TMDD model (e.g., Michaelis-Menten, rapid-binding or quasi-steady-state models) are used, and RO is calculated often using the Michaelis-Menten constant ( K_ M ) derived from these models. In this simulation study, we conduct analyses using models with parameter values sourced from literature to show that RO derived from K_ M may not reflect the true RO. When data available for model-fitting consists only of drug concentrations, the accuracy and precision of the model parameters such as K_ M depend on the selected approximate models and their model assumptions, e.g., where binding takes place. We hope these analyses provide insights in performing model-derived RO predictions and dose optimization—which often depends on RO—in clinical drug development.
The purpose of the present study was to evaluate whether dissolution testing in biorelevant bicarbonate buffer (BCB) can predict the solubility-limited oral absorption of high-dose free acid drugs. Valsartan and naproxen were selected as model compounds exhibiting solubility-limited absorption at high doses. A biorelevant BCB solution (concentration: 10 mM, initial pH (pHini) = 6.5, buffer capacity (β) = 4.4 mM/pH) (BCB), a phosphate buffer solution with the same pH and β (PPB), and a USP phosphate buffer solution (β = 27 mM/pH, modified to pHini 6.5) were employed for dissolution tests. The dose-to-fluid volume ratio (Dose/FV) was set to reflect the clinical or compendial conditions. The fraction of a dose absorbed in humans (Fa) predicted using a simple theoretical framework was compared with the clinical Fa calculated from pharmacokinetic data in the literature. When using BCB or PPB with the same β and the clinical Dose/FV, the bulk-phase pH after 4 h (pHbulk, 4 h) was markedly lower than pHini (< pH 6.0 for naproxen, < pH 5.0 for valsartan). Consequently, the dissolved drug concentration at 4 h (Cdissolv, 4 h) was markedly lower than the solubility at pH 6.5. Using Cdissolv, 4 h, the dose dependency of Fa was appropriately predicted. In contrast, when using USP or the compendial Dose/FV, the pHbulk, 4 h and Cdissolv, 4 h were less reduced, leading to an overestimation of Fa. Overall, the use of BCB or PPB and a clinical Dose/FV improved the in vivo predictability for high-dose free acid drugs.
Antibody-drug conjugates (ADCs) represent a rapidly evolving therapeutic modality, combining the selective targeting of monoclonal antibodies with highly potent small molecule payloads. Their inherent structural complexity demands a sophisticated and multi-faceted bioanalytical approach spanning preclinical discovery through late-stage clinical development. This white paper, developed by the ADC Working Group of the AAPS bioanalytical community, comprising over 100 members from industry, contract research organizations (CROs), and regulatory agencies, provides updated recommendations for ADC bioanalysis. Building upon the foundational 2013 AAPS position paper and recent publications governing regulatory frameworks on PK considerations, this work addresses advances in bioanalytical quantitation strategies for total antibody (tAb), conjugated ADC, free (unconjugated) payload, and drug-to antibody ratio (DAR); novel immunogenicity assessment considerations; soluble target interference; critical reagent lifecycle management; payload-specific stability requirements; cross-validation strategies; and regulatory considerations. The recommendations presented herein reflect over a decade of scientific progress and are designed to serve as a comprehensive, empirically validated, and industry aligned bioanalytical framework for contemporary ADC drug development.
Therapeutic peptides and proteins form higher-order structures (HOS) including oligomeric forms in formulation, critical for drug efficacy, safety, and stability. To quickly assess protein oligomerization, dynamic light scattering (DLS) was applied to measure a series of protein standards with molecular weight (MW) range of 1.3–660 kDa, yielding translational diffusion coefficients (Ddls), which were corrected to Dcorr using water diffusion data. A correlation between Dcorr and protein monomeric MW, representing their hydrodynamic MW (MWhd), was established as 𝐥𝐨𝐠(D_𝐜𝐨𝐫𝐫)=-0.428𝐥𝐨𝐠(𝐌𝐖_𝐡𝐝)-5.412 . This DLS/MWhd model was subsequently applied to therapeutic protein formulations with monomeric MW ranging from 3.8 to 149 kDa. The resulting MWhd values were several-fold greater than the corresponding monomeric MW of the glucagon-like peptide-1/2 (GLP-1/2) analogs, insulin analogs, and the monoclonal antibodies (mAbs) infliximab and bevacizumab, indicating varying degrees of oligomerization. The observed oligomerization states were largely consistent with those reported in literature. For insulin and mAb oligomers, pseudo-spherical diffusion coefficients (Ds) back calculated from the oligomer MW agreed within 6
Lipid nanoparticles (LNPs) containing mRNA that encodes therapeutic proteins have emerged as a promising therapeutic modality for a myriad of diseases. However, the pharmacokinetic (PK) relationships between the nanoparticle carrier, its mRNA payload, and the expressed protein remain poorly understood. Here we have investigated whole-body PK of these components in mice following intravenous administration of an mRNA-LNP that expresses a non-cross-reactive monoclonal antibody. LNPs encapsulating mRNA were prepared via microfluidic mixing and characterized for physicochemical properties and encapsulation efficiency. Following a single intravenous administration of mRNA-LNP, blood, plasma, and tissues were collected for PK measurement over 2 weeks. Ionizable lipid (i.e., ALC-0315) concentrations were determined using LC–MS, mRNA concentrations were measured using qPCR, and the expressed antibody concentrations were determined using ELISA. It was found that each component exhibited a distinct PK profile. For example, lipid exposure was highest in liver and spleen, while mRNA accumulation peaked in spleen and heart. Strikingly, the expressed antibody demonstrated a distribution pattern that did not mirror mRNA exposure, with the lung showing the greatest antibody levels despite only modest mRNA delivery. Tissue-to-plasma ratios for the expressed antibody exceeded values reported for intravenously administered antibodies, suggesting localized production or enhanced retention following in situ translation. These findings highlight that functional protein exposure is governed by tissue-specific translational efficiency and protein properties, rather than just nanoparticle delivery. The comprehensive PK data presented here also provides a foundation for the development of systems-based PK models for antibody expressing mRNA-LNPs.
Physiologically Based Biopharmaceutics Models (PBBMs) are evolving tools which can be used to aid in drug product quality specifications. PBBMs have gained increased regulatory interest and acceptance over the past 5 years. For molnupiravir, PBBM has been used for: a) calculation of a bioequivalence safe space for capsule formulation and tablets, b) informing of the definitive bioequivalence study (P011) with a target tablet formulation selection designed to be bioequivalent to the capsule formulation, c) selection of a deliberately slower release tablet formulation (beyond f2 similarity criteria) designed to widen the dissolution knowledge space, d) help select the clinically relevant specification and the dissolution bioequivalence safe space. PBBM data inputs, model development, validation, and application using data from several independent clinical studies are described. Predictive errors for Cmax and AUC were < 25
The virtual bioequivalence (VBE) approach, which employs physiologically based pharmacokinetic (PBPK) modeling in lieu of clinical bioequivalence (BE) studies, provides a valuable alternative for the pharmaceutical industry. The VBE has the potential to substantially reduce both the time and costs associated with traditional clinical BE studies. Furthermore, this approach supports ethical principles by eliminating the need to expose healthy volunteers to investigational compounds without medical benefit. Intrasubject variability (intrasubject coefficient of variation, ICV) is a main factor for balanced cross-over design BE studies in calculating the confidence intervals for the geometric mean ratio of AUC and Cmax, which determines if bioequivalence criteria are met. Thus, it is essential to employ realistic and accurate ICV values for BE assessment through PBPK modeling. Published literature indicates that the drug properties that affect oral bioavailability exhibit a good correlation with ICV. The aim of this research is to expand drug databases to reevaluate the correlation between drug properties and clinically observed ICV for Cmax (ICV_Cmax). The results demonstrated that the dose number (dose to solubility ratio), permeability and elimination half-life are three main factors influencing ICV_Cmax. Permeability becomes a dominating factor when the value is relatively low (Papp < 0.8 × 10-6 cm/sec). In addition, PBPK modeling tools, Gastroplus and Simcyp were evaluated for their ability to predict ICV_Cmax based on drug properties and their built-in physiological variations. Both simulation tools estimated intersubject variability and ICV_Cmax similarly and underestimated ICV_Cmax compared with clinically observed ICV_Cmax.
