Innovative Clinical Pharmacology, Modeling, and Simulation Strategies for Accelerating Rare Disease Drug Development

Abstract

Clinical drug development for rare diseases continues to face significant challenges due to disease heterogeneity, fewer available patients, and incomplete understanding of pathogenesis, resulting in trials with limited clinical data, thus constraining traditional development pathways. Clinical pharmacology, modeling, and simulation‐based approaches can help address these challenges by informing decision‐making, mitigating uncertainty, and guiding optimal dose and regimen selection for the appropriate patient population. These approaches help streamline trial designs by reducing the scope and number of clinical trial evaluations, using exposure–response analyses to optimize dosing, the use of mechanistic‐physiologically based pharmacokinetics (M‐PBPK)‐based approaches for biopharmaceutical and formulation optimization, evaluations of drug–drug interactions, and organ impairment. These strategies increase development efficiency across all stages of drug development, thereby improving the probability of success. This review highlights case studies that applied innovative clinical and quantitative pharmacology approaches across early and late stages of drug development and regulatory decision‐making in rare diseases. The specific examples illustrate the application of pharmacokinetics/pharmacodynamics (PK/PD) and model‐informed drug development (MIDD) strategies to support dose and regimen selection, enabling efficient use of direct or adaptive trial designs, facilitating bridging across populations and indications, biopharmaceutics‐based transitions, and generating integrated PK/PD evidence to support labeling. Examples include drug repurposing, characterizing PK/PD in early phase to inform late‐phase development, population PK analysis to guide trial dosing and label recommendations, using phenotype‐targeted study design to address disease heterogeneity, expanding dosing regimen across indications using MIDD, quantitatively evaluating immunogenicity to support mitigation strategies, biomarker bridging, and applying M‐PBPK to predict clinical PK in organ impairment populations.

Introduction

Clinical pharmacology is a critical component in the design of efficient drug development. The foundation of clinical pharmacology is based on pharmacokinetics (PK), that is, what the body does to the drug, and pharmacodynamics (PD), what the drug does to the body. Integration of both PK and PD provides a set of quantitative tools that enable model‐informed drug development (MIDD) to the clinical pharmacologist, ¹ providing a strategic position in translational and clinical drug development to influence decision‐making for selecting the dose and the dosing regimen to the right patient population at the right stage of development.

In orphan drug development for rare diseases, the role of a clinical pharmacologist becomes increasingly vital as limited data generation makes quantitative pharmacologic modeling and simulation emerge as an essential, high‐impact tool for informed and optimized drug development. As the name implies, “rare diseases” are rare—they affect only a small portion of the population in a geographic region—less than 200,000 individuals in the United States ² and 5 in 10000 individuals or fewer in Europe ³ ; affecting more than 300 million people worldwide. ⁴

Clinical drug development of rare diseases is confronted by numerous challenges. These challenges begin with the etiology of each disease—new and unknown precursor and causal mechanisms, often characterized by chronic, complex, heterogeneous, and progressive disease leading to disabilities and premature death. Additionally, rare diseases often have limited or no established preclinical efficacy models or biomarkers, further exacerbating the narrow understanding of disease mechanisms. Lastly, varying prevalence of each rare disease geographically and the varying clinical manifestations and disease progression make recruitment and clinical drug development using the traditional clinical development (i.e., Phase 1, Phase 2, and Phase 3) pathway often severely challenging. Despite these challenges, the development of drugs to treat rare diseases is of paramount importance. As a result of these factors, strategies based on clinical pharmacology, modeling, and simulation methodologies enable the development and design of unique and novel clinical trials that maximize clinical data (safety, efficacy, PK, and PD) from a relatively limited patient pool.

In this review, we present case studies that have demonstrated the use of clinical pharmacology strategies, including the use of quantitative tools across early and late stages of drug development of rare diseases. Each example highlights a distinct strategy based on innovative clinical and quantitative pharmacology that accelerated the development of drugs in rare diseases (Table 1). The strategies presented here include the use of an integrated clinical pharmacology strategy early in drug development based on a comprehensive clinical pharmacology plan that provides guidance on dosing and administration (dosing regimen and food‐effect), as well as justification of dose and dosing regimen in ethnic populations, with drug–drug interactions and in organ impairment studies. Other strategies presented here include the use of MIDD, such as exposure–response (E‐R) analyses to optimize dose and dosing regimen across indications, extrapolation of dose and dosing regimen from adults to pediatrics, use of exposure–biomarker relationship to support dosing regimen across indications, quantitative assessment of immunogenicity to develop mitigation strategies for anti‐drug antibodies (ADAs), and a potential application of MIDD in understanding the role of phenotype‐driven patient‐reported outcomes to help inform pivotal trials. Additionally, quantitative tools such as mechanistic‐physiologically based pharmacokinetics/pharmacodynamics (M‐PBPK/PD) modeling framework help predict clinical PK and guide selection and justification of optimal dose and dosing regimen to other subpopulations, including patients with organ impairment. One example highlights the role of MIDD in enabling drug repurposing, where PK was evaluated from an unrelated therapeutic area, and the indication supported the development and approval of the drug in a new therapeutic area and an unrelated indication.

Table 1.

Examples Demonstrating Clinical Pharmacology, Modeling, Simulation Strategies Used in Rare Disease Drug Development

Clinical Pharmacology Strategy	Example (Disease; Modality)	Key Clinical Pharmacology Evidence Supporting Drug Development	Impact on Drug Development	References
Traditional clinical pharmacology	Avacopan (ANCA‐associated vasculitis; small molecule)	FIH/Phase 1 showed dose‐dependent PK and ex vivo PD. Food increased exposure, and CYP3A4 was identified as the dominant metabolic pathway.	Traditional Phase 1 PK/PD directly informed dose selection, BID regimen choice, and label recommendations for food and CYP3A4 interactions.	6 , 9 , 10 , 11 , 12 , 13
Clinical pharmacology adapted for novel modalities
In vivo gene therapy	Voretigene neparvovec (RPE65 retinal dystrophy; AAV gene therapy)	No conventional PK or DDI studies were performed. Instead, the clinical pharmacology package was built around vector–genome dose per eye and vector shedding and biodistribution monitoring.	Demonstrates how classical clinical pharmacology principles can be translated for ultra‐small rare‐disease gene‐therapy programs where standard ADME is not informative.	16 , 17
Ex vivo cell/gene therapy	Betibeglogene autotemcel (transfusion‐dependent β‐thalassemia; autologous gene‐modified HSC therapy)	Conventional PK/ADME were not applicable. PK‐related characterization included product, cellular, and vector exposure measures such as cell dose, drug product vector copy number, percentage of lentiviral vector‐positive cells, engraftment, persistence of transduced hematopoietic cells, and peripheral blood/PBMC vector copy number over time.		19
Leveraging clinical PK/PD to inform rare disease clinical development
PopPK/PD, exposure–response, and extrapolation framework	Inebilizumab (NMOSD, gMG, and IgG4‐RD; anti‐CD19 mAb)	Population PK/PD analyses supported dose selection for NMOSD; expanded dosing regimen to other indications allowing direct‐to‐Phase 3 transition; supported use in subpopulations including pediatrics.	One regimen using biomarker‐led dose optimization allowed bridging across indications without repeating a full dose‐ranging study in every small population, demonstrating that a mechanism‐linked PD biomarker can substitute for much larger comparative datasets when recruitment is limited.	30 , 31 , 32 , 33
Mechanism‐linked biomarker bridging across indications	Eculizumab (aHUS, gMG, NMOSD; terminal complement inhibitor)	Model‐based analyses in gMG and NMOSD showed the approved 900/1200‐mg regimen rapidly achieved sustained complement inhibition.		49 , 52 , 53
PRO‐integrated, phenotype‐specific development	Dazodalibep (Sjögren's disease; CD40L antagonist, investigational)	Phase 2 intentionally enrolled two distinct phenotypes: a systemic‐disease population using ESSDAI and a symptom‐burden population using ESSPRI. This links mechanism‐based pharmacology to phenotype‐specific outcomes.	Creates a platform for phenotype‐specific exposure–response and biomarker–response analyses. This is particularly valuable for heterogeneous, low‐prevalent immune diseases where the endpoint is not clearly defined.	56 , 57 , 60
Drug repurposing with model‐informed dose selection	Teprotumumab (thyroid eye disease; IGF‐1R mAb)	No dedicated dose‐ranging study was conducted in thyroid eye disease. Prior oncology Phase 1 PK plus target‐mediated clearance analysis, and PopPK modeling supported a regimen expected to maintain concentrations above the threshold for >90% saturation of target‐mediated clearance.	Reusing prior PK knowledge helped compress development timelines when the disease is rare and recruitment is challenging.	72 , 73
Mechanistic modeling	Clinical PBPK/PD	Use of PBPK‐PD modeling framework to predict PK in patients with organ impairment, allowing expansion to subpopulations, thus reducing need for dedicated trials in hard‐to‐treat populations.	This example provides supporting evidence for novel modalities when clinical datasets are small, healthy‐volunteer studies may be uninformative or infeasible, and dedicated subgroup studies are limited.	63
Immunogenicity integrated with PK/PD	Pegloticase + methotrexate (refractory gout;)	Higher anti‐pegloticase titers were associated with faster clearance, lower efficacy, and more infusion reactions. Serum uric acid was used as a downstream PD marker; co‐therapy with MTX reduced ADA formation and infusion reactions.	Supported prospective ADA assessment and label‐based serum uric acid monitoring demonstrating how routinely collected PD markers can help optimize benefit–risk for patients when sample sizes are small.	82
Biopharmaceutics/formulation bridge	Cysteamine bitartrate delayed‐release (PROCYSBI) (nephropathic cystinosis; small molecule)	A bioequivalence study along with assessment of biomarker supported bridging from immediate‐release to delayed‐release formulation, and enabled a less frequent dosing schedule while maintaining clinically relevant exposure/PD control.	The new dosing regimen helped improve patient compliance and adherence to treatment.	84

Collectively, these strategies have helped influence regulatory decision‐making at each stage of drug development, ensuring optimal dose and regimen selection for the right patient population, efficiently driving drug development and accelerating the overall development pathway, thereby increasing the probability of success. Eventually, the innovative strategies based on clinical pharmacology, modeling, and simulation have led to evidence‐based labeling recommendations (Figure 1).

Figure 1.

Innovative clinical pharmacology, modeling, simulation strategies in decision‐making across drug development. M‐PBPK, mechanistic physiologically based pharmacokinetic model; PK, pharmacokinetics; PRO, patient reported outcomes.

Application to Design Clinical Pharmacology Strategy to Support Clinical Development

Clinical pharmacologists play a pivotal role from the early stages of drug development, beginning with Phase 1, by defining and designing an integrated clinical pharmacology strategy that sets the initial clinical dose, guides dose escalation, and helps establish a clear pathway for Phase 2 dose and regimen selection in the appropriate patient population. This strategy extends beyond modeling and simulation to include fit‐for‐purpose clinical PK studies that directly inform dose selection, risk assessment, and provide a quantitative basis for key labeling decisions in the prescribing information.

Examples illustrating this application can be seen in the development and approval of the small molecule drug avacopan, where a comprehensive “conventional” clinical pharmacology strategy during drug development played an integral role in informing dose selection, administration, and labeling despite small sample sizes typical of orphan diseases. Voretigene neparvovec and betibeglogene autotemcel are two examples that demonstrate how clinical pharmacology principles were defined less by conventional systemic PK/PD and more by cell kinetics, vector biodistribution/shedding, and transfusion independence with local PD effect for voretigene neparvovec and population PD assessment for betibeglogene autotemcel.

Avacopan

An orally available, selective antagonist of the complement 5a receptor (also called CD88), avacopan is approved as an adjunctive treatment of adult patients with severe active anti‐neutrophil cytoplasmic autoantibody (ANCA)‐associated vasculitis (AAV) (granulomatosis with polyangiitis [GPA] and microscopic polyangiitis [MPA]) in combination with standard therapy including glucocorticoids. ⁵ The first‐in‐human (FIH) single‐ and multiple‐ascending dose studies were conducted in healthy participants to characterize the PK and tolerability of avacopan. ⁶ The FIH study also evaluated ex vivo PD assessment of C5a‐induced upregulation of CD11b in circulating neutrophils using peripheral blood samples from selected cohorts, including 10‐mg single‐dose at 2 and 24 h postdose, 30‐mg and 100‐mg single‐dose cohorts at 2 and 12 h postdose, and 30‐mg twice daily (BID) cohort on Day 7 at 2 and 12 h postdose. The results from this early PK/PD assessment between avacopan plasma exposures and ex vivo PD effect demonstrated the efficacy of avacopan 30 mg BID in blocking C5aR by approximately 94% or greater throughout the day, based on assessments on Day 7 at 2 and 12 h postdose. ⁶ This provided a mechanistic rationale to evaluate the 30‐mg BID regimen in Phase 2 studies. ⁷ , ⁸ Building on this rationale, the clinical pharmacology plan to support later‐phase development and regulatory evaluation consisted of a food‐effect study, ⁹ thorough QTc assessment, ¹⁰ drug–drug interaction studies to assess the interaction between metabolic enzymes and avacopan, ¹¹ as well as studies in special populations including a hepatic impairment study ¹² and the impact of Japanese ethnicity ⁹ on the PK of avacopan. In addition, a human mass balance study was conducted to characterize the absorption, metabolism, and excretion (AME) of avacopan. ¹³ Results from this study demonstrated negligible elimination via urine (<0.1%), eliminating the need for a dedicated renal impairment study. Since renal impairment is a common manifestation of AAV, patients with varying degrees of renal impairment, except patients with end‐stage renal disease requiring dialysis, were enrolled in Phase 2 and Phase 3 studies. ⁷ , ⁸ , ¹⁴ Population PK analysis confirmed that renal function (eGFR), although a statistically significant covariate of avacopan PK, did not result in a clinically meaningful effect on avacopan exposures, ¹⁵ thus supporting a unified strategy for dosing regimen in patients with AAV and varying degrees of renal impairment. To further support the use of avacopan in patients with ESRD requiring dialysis, a dedicated Phase 1 study has been conducted (NCT06468826).

This case study presents a distinctive example of drug development for a rare, orphan autoimmune disease, illustrating how traditional clinical pharmacology findings directly informed key development decisions. These included selecting single‐ and multiple‐doses for evaluation in the FIH study, conducting an early ex vivo PD assessment to guide dose selection for Phase 2 studies, using early PK data to determine that a BID regimen was more appropriate than a once‐daily (QD) regimen, and informing prescribing information related to dosing and administration with respect to food, dose reduction with concomitant use of strong CYP3A4 inhibitors, and recommendations to avoid use with moderate and string CYP3A4 inducers.

Voretigene Neparvovec and Betibeglogene Autotemcel

Voretigene neparvovec, a first in vivo gene therapy, was approved for adult and pediatric patients with RPE65‐mediated retinal dystrophy in 2017 ¹⁶ at a recommended dose of 1.5 × 10^11 vector genomes per eye in 0.3 mL. ¹⁷ Given that voretigene neparvovec is an adeno‐associated virus serotype 2 (AAV2) therapy, the conventional clinical pharmacology mindset was adapted for an in vivo ocular gene therapy. Rather than relying on conventional plasma PK, PK characterization focused on dose exploration across low‐, middle‐, and high‐dose cohorts that resulted in selection of the recommended dose, subretinal route, and site of delivery, ocular and systemic biodistribution, vector persistence or clearance, and vector shedding assessed by qPCR in tears from both eyes, serum, and whole blood, and PD‐activity was evaluated through functional visual endpoints and biomarkers including retinal pigment epithelial cell transduction, functional RPE65 expression, restoration of visual‐cycle activity, and functional visual endpoints such as multi‐luminance mobility testing, full‐field stimulus threshold, visual acuity, and visual field assessments.

Betibeglogene autotemcel ¹⁸ is a one‐time gene therapy for adult and pediatric patients with transfusion‐dependent β‐thalassemia approved in 2022. It is an autologous CD34+ hematopoietic stem and progenitor cell therapy transduced ex vivo with the BB305 lentiviral vector encoding βA‐T87Q‐globin. ¹⁹ Here, conventional ADME and plasma PK were not applicable. Instead, key clinical pharmacology assessments included exposure‐related CD34+ cell dose, drug product vector copy number, percentage of lentiviral vector‐positive cells, engraftment, persistence of transduced hematopoietic cells, and peripheral blood or PBMC vector copy number over time. ¹⁹ PD was represented by the production of gene therapy–derived HbAT87Q, unsupported total hemoglobin, transfusion independence, transfusion reduction, and improvement in markers of ineffective erythropoiesis.

Both the examples above illustrate how clinical pharmacology principles are adapted when conventional plasma PK is not the main measure of exposure. Additionally, these examples show how modality‐specific exposures, PD, efficacy, and safety measures can support streamlined development in small rare‐disease populations.

Leveraging Clinical PK/PD to Inform Rare Disease Clinical Development

Translational PK/PD modeling integrates preclinical data with mechanism‐based models to predict drug behavior in humans ²⁰ by quantitatively linking drug exposure to pharmacological effects, thereby informing early clinical decision‐making. Crucial for the selection of the FIH dose, translational PK/PD approaches help balance safety with pharmacological relevance. Spanning from empirical to mechanistic, examples of these approaches include allometric scaling, ²¹ PBPK modeling, ²² quantitative systems pharmacology, ²³ and innovative approaches such as adaptive trial designs and Bayesian analyses.

Post‐translational stage, PK/PD modeling and simulation methodologies play a core role across all stages of clinical drug development, including extension to subpopulations such as hepatic impairment, renal impairment, ethnic populations and pediatrics, and life‐cycle management. Quantitative frameworks—PBPK, population PK, and E‐R modeling—support rational dose justification even when empirical data are sparse. In addition to characterizing the PK and PD of a drug, these models also evaluate the impact of covariates, including patient characteristics (e.g., age, gender, body weight, ethnicity, and the effect of organ impairment), sources of variability, compliance, drop‐out rates, and even dose adjustments based on adverse events or drug–drug interactions. When E‐R coherence and mechanistic plausibility are robust, clinical pharmacology‐based modeling has been shown to effectively provide an alternative to traditional “clinical trial phase” transitions, providing the evidentiary foundation for accelerated approval. Furthermore, regulatory agencies have institutionalized these methods through initiatives such as the United States Food and Drug Administration (US FDA) MIDD Pilot Program, underscoring the role of modeling as a regulatory decision tool, ²⁴ and more recently by the European Medicines Agency (EMA) for use of mechanistic models for regulatory purposes, including PBPK. ²⁵ , ²⁶ E‐R analyses can also support bridging between populations (e.g., adults to pediatrics), formulations, or indications, and often form part of the labeling justification for recommended dosing. PBPK models offer a complementary approach to clinical pharmacokinetic studies that may help in reducing the need for dedicated studies in certain cases. When integrated early in development, these analyses allow for evidence‐based decisions, reducing uncertainty at the time of regulatory review and facilitating approval by demonstrating that the selected regimen provides optimal benefit–risk balance.

Application of Model‐Informed Approaches to Enable Indication Bridging and Extension to Subpopulations

MIDD approaches play a significant role across all stages of drug development and influence regulatory decision‐making, and have been routinely used to optimize doses and dosing regimens, and clinical trial designs by reducing the number of clinical trial evaluations. ²⁷ This is particularly important when the objective is to bridge dose and dosing regimens across indications. By integrating PK and PD, model‐based approaches offer a strategic foundation to clinical drug development by guiding selection of optimal sampling, optimal dose and dosing regimen, bridging across indications, and routes of administration, quantitatively assessing biomarkers and surrogate endpoints, and patient‐reported outcomes to trial design and selection of patient population, and use of model‐based meta‐analysis along with natural history and real‐world data to help establish efficacy and safety. ¹

Examples illustrating the use of a model‐based approach such as population PK analysis and E‐R analysis that led to efficient drug development and increased the probability of success are inebilizumab and eculizumab.

Inebilizumab

Inebilizumab is an afucosylated, humanized IgG1 kappa–based monoclonal antibody (mAb) against human CD19, expressed on a broad spectrum of pathogenic antibody–secreting B cells often involved in autoimmune disorders. ²⁸ Inebilizumab was initially approved for use in adults with NMOSD, a rare autoimmune disease characterized by demyelination due to pathogenic antibodies, after a single pivotal Phase 3 clinical trial. ²⁹ The rare incidence of NMOSD precluded the traditional approach of first evaluating signs of efficacy in a Phase 2 proof‐of‐concept study prior to conducting two well‐controlled Phase 3 studies. Instead, following two Phase 1 dose‐escalating trials in systemic sclerosis (single dose; 0.1–10 mg/kg IV) with dense PK and B‐cell count sampling ³⁰ and in relapsing‐remitting multiple sclerosis (multiple dose; fixed IV doses of 30, 100, or 600 mg 2 weeks apart, and two cohorts of single SC dose), ³¹ inebilizumab was evaluated directly in a single pivotal Phase 3 study in adults with NMOSD without any prior proof‐of‐concept or dose finding studies in this disease setting. The Phase 3 study in NMOSD evaluated a single dose level (300 mg) given intravenously on Days 1 and 15, followed by every 6‐month maintenance dosing, a dosing regimen not evaluated previously in clinical development. This venture into a new indication with a dosing regimen not evaluated before provided an enormous risk to the program; however, a PK/PD modeling and simulation framework was implemented to support and strengthen this strategy.

The PK of inebilizumab has been extensively evaluated across several autoimmune indications. Inebilizumab follows non‐linear PK at lower doses, presumably due to saturation of all available CD19 targets, before exhibiting predominantly linear PK at doses ≥ 1 mg/kg. ³⁰ , ³¹ , ³² For a drug demonstrating B‐cell depletion, both the depth and durability of depletion are relevant for B‐cell–mediated autoimmune diseases. Inebilizumab doses of 3 mg/kg or higher demonstrated deep and durable B‐cell depletion for 6 months. This crucial information was utilized as part of an MIDD strategy to pursue approval in NMOSD with a direct‐to‐Phase 3 approach using a single pivotal study. Population PK model–based simulations supported the selection of a fixed inebilizumab dose of 300 mg given intravenously on Days 1 and 15, then every 6 months, a regimen that provided at least 3 mg/kg dosing that previously demonstrated deep and sustainable B‐cell depletion in individuals up to 100 kg body weight. ³³ The selection of the dosing regimen, which had never been evaluated in the inebilizumab program, was also based on modeling of B cell kinetics, where population PK/PD modeling and simulation were utilized to optimize the dose schedule. ³²

Mechanistically, the initial dose on Day 1 rapidly decreases peripheral CD19+ B cells, with a subsequent induction (loading) dose on Day 15 that was timed to deplete the newly circulated B cells released from the lymphoid and other tissue repositories to replenish peripheral B cells following depletion from the Day‐1 dose. The maintenance dosing of every 6 months was also optimized based on population PK/PD simulations focused on keeping B‐cell repletion below a critical threshold. This MIDD‐based strategy led to successful approval of inebilizumab for treatment of patients with NMOSD by the US FDA in 2020 and by the EMA in 2022, ³⁴ with several countries following suit.

Following the approval of inebilizumab in adults with NMOSD, the next step included expansion to pediatrics (2–17 years of age) with NMOSD. Since NMOSD is a rare disease, recruitment is challenging. Furthermore, in pediatrics with rare diseases, such as NMOSD, there are added factors, including correct diagnosis, age of diagnosis, similarity of symptoms between adults and pediatrics, and parental consent to participate in clinical trials. The traditional approach of a larger randomized, placebo‐ or active‐controlled Phase 2 study to demonstrate efficacy (proof of concept and dose‐finding study) is therefore challenging. Instead, an open‐label Phase 2 PK/PD study, in an appropriate pediatric age group, ³⁵ was accepted by regulatory agencies. In this “extrapolation” approach, efficacy and safety can be extrapolated from adults (where it was demonstrated in the Phase 3 registrational study) to pediatrics if comparable PK exposure is observed in pediatrics. MIDD approaches play an influential role in the development of pediatric rare diseases by leveraging all available information to guide dose and regimen selection, thereby bridging gaps to inform decision‐making and reducing uncertainty. ³⁶ A Phase 2 open‐label, multicenter study to evaluate the PK, PD, and safety of inebilizumab in eligible pediatric participants 2 to < 18 years of age with recently active NMOSD who are seropositive for autoantibodies against aquaporin‐4 (AQP4‐immunoglobulin [Ig]G) is ongoing (NCT05549258).

Concurrently with the development of a pediatric NMOSD study, inebilizumab was subsequently developed in parallel in two additional rare autoimmune disorders, IgG4‐related disease (IgG4‐RD) ³⁷ and gMG, ³⁸ both identified as having disease pathophysiology that involved CD19+ B cells. These two new indications were subsequently evaluated with separate Phase 3 registrational trials using the same approved NMOSD dosing regimen (NCT04524273, NCT04540497). Again, each of these pivotal studies was launched directly to Phase 3, bypassing the traditional progression of evaluating efficacy and dose‐finding in Phase 2, with inebilizumab obtaining approval by the US FDA and the European Commission in 2025 and 2026 for both indications. ³⁹ , ⁴⁰

Given similar CD19+ B‐cell–mediated pathophysiology between IgG4‐RD, gMG, and NMOSD, robust PK/PD modeling and simulation, coupled with the relatively flat exposure–response relationship with NMOSD, ³² , ⁴¹ and similar PK across ⁴⁰ indications provided convincing justification to use the same NMOSD‐approved dosing regimen in both IgG4‐RD ⁴² and gMG. ⁴³

Eculizumab

Eculizumab is a first‐in‐class humanized mAb that binds to complement component C5, preventing its cleavage into C5a and C5b and thereby inhibiting terminal complement‐mediated cell lysis and activation. Eculizumab was first approved by the US FDA in 2007 for the treatment of paroxysmal nocturnal hemoglobinuria (PNH), a rare, progressive, and life‐threatening hematologic disorder characterized by uncontrolled intravascular hemolysis and a markedly increased risk of thrombosis due to complement overactivation. ⁴⁴ CH50, a measure of total serum hemolytic complement activity, was identified as the key PD surrogate endpoint, mechanistically linking intravascular hemolysis with circulating eculizumab concentrations, guiding dose selection and optimization. ⁴⁵ In a total of 140 evaluated patients, including 43 from the TRIUMPH trial ⁴⁶ and 97 from the SHEPHERD trial, ⁴⁷ following the first infusion of eculizumab, 49 of 135 patients (36%) had serum concentrations below 35 µg/mL; among these, 36 of 49 patients (73.5%) exhibited hemolysis over 20%. ⁴⁸ These findings indicated that maintaining trough eculizumab levels above approximately 35 µg/mL was necessary to achieve consistent inhibition of intravascular hemolysis in PNH. Further, the relationship between eculizumab exposure and complement suppression (ex vivo hemolysis) was characterized using a direct inhibitory E_max model. ⁴⁹ The approved dosing regimen of 600 mg weekly for 4 weeks followed by 900 mg for the fifth dose 1 week later, then 900 mg every 2 weeks thereafter, was optimized to maintain trough eculizumab concentrations above ∼35 µg/mL, the threshold required for sustained complement blockade and consistent inhibition of intravascular hemolysis in PNH. ⁵⁰ , ⁵¹ This biomarker‐driven approach also supported dose finding across indications. The same exposure and pharmacodynamic surrogate endpoint established in PNH patient population were applied to subsequent indications, including atypical hemolytic uremic syndrome (aHUS), gMG, and NMOSD, given their shared complement‐mediated pathology. ⁵² For example, in clinical trial for gMG trials, E‐R analyses identified a target eculizumab concentration of approximately 116 µg/mL, which corresponded to the level predicted to achieve 20% hemolysis. To ensure full complement inhibition, a higher clinical dose regimen (900 mg weekly for 4 weeks followed by 1200 mg every 2 weeks) was approved. ⁵³

Quantitative Assessment of Patient Reported Outcomes (PROs) as an Endpoint

PROs have long been known to play a critical role in understanding the impact of a drug directly on the health and quality of life of a patient, outside the bounds of clinical endpoints. ⁵⁴ PRO measures have been used as primary and secondary endpoints in labeling of orphan drugs, although their use has been limited, often due to challenges including inadequate data collection, validated PRO forms for the target population especially pediatrics, and the use of robust statistical analysis given the small sample sizes and heterogeneity in outcomes. An integration of PRO measures with clinical and quantitative pharmacology is still at its nascent stage. However, a quantitative assessment of MIDD using PRO measures holds potential to help guide dose and regimen selection, and drug tolerability, thus pairing and guiding benefit–risk assessment with patient experience, particularly for rare diseases with limited sample size. Dazodalibep presented below illustrates how phenotype‐targeted study design can help address questions related to disease heterogeneity, bringing patient experience into benefit‐risk assessment.

Dazodalibep

Dazodalibep is an investigational, non‐antibody fusion protein currently under evaluation in Phase 3 trials for the treatment of Sjogren's disease (NCT06245408), a chronic autoimmune condition characterized by dryness due to inflammation of the lacrimal and salivary glands, as well as extreme fatigue, muscular and joint pain, and systemic organ impairment. ⁵⁵ Dazodalibep acts by specific binding to human CD40L, thus inhibiting its interaction with human CD40 and presents yet another example of a drug repurposed in clinical development from rheumatoid arthritis ⁵⁶ to the treatment of Sjogren's disease. ⁵⁷ European League Against Rheumatism (EULAR) quantifies the symptomatic and systemic severity of Sjogren's with two tools: ESSPRI (EULAR Sjogren's Syndrome Patient Reported Index) and ESSDAI (EULAR Sjogren's Syndrome Disease Activity Index). ⁵⁸ ESSPRI is a validated PRO measure that quantitates dryness, fatigue, and pain, where dry mouth cause difficulty swallowing, talking, and eating thus leading to dental cavities, fatigue manifesting as extreme chronic exhaustion forcing lifestyle changes and affecting work, and joint and muscle pain often being mistaken for arthritis, occurring suddenly and severely. ⁵⁹ ESSDAI, on the other hand, is a physician assessed systemic disease activity index that quantitates 12 organ systems; combined domains assessed in ESSDAI and ESSPRI show the heterogeneity and complexity of treating Sjogren's disease. ⁵⁹

One of the primary challenges in the treatment of Sjögren's disease is the substantial clinical heterogeneity of the disease and the difficulty in interpreting treatment effects across patients with markedly different disease manifestations. Patients could either experience severe symptoms of dryness, fatigue, and pain despite limited systemic inflammatory involvement, or could exhibit substantial systemic disease activity with a different symptom profile. When such heterogeneous populations are analyzed together, clinically meaningful treatment effects within specific patient subgroups may be diluted or obscured. A key innovation of the dazodalibep Sjögren's program is the prospective recognition that systemic inflammatory activity and patient symptom burden represent partially dissociable therapeutic dimensions. The Phase 2 trial ⁵⁷ grouped patients into two phenotype‐defined populations: one with moderate‐to‐severe systemic disease activity (ESSDAI total score ≥5) and one with high symptom burden but limited systemic organ involvement (ESSPRI total score ≥5 and ESSDAI total score <5). Results from Phase 2 study demonstrated significant improvements in ESSDAI and ESSPRI scores in Population 1 and Population 2, respectively. This phenotype‐based design enabled evaluation of treatment response within clinically distinct subgroups rather than a single heterogeneous Sjögren's disease population. Patients with moderate‐to‐severe systemic disease activity achieved significant improvements in ESSDAI, whereas patients with high symptom burden and limited systemic involvement demonstrated significant improvements in ESSPRI, suggesting that clinically meaningful treatment benefits may manifest differently across disease phenotypes. The methodological novelty here was therefore not the use of a PRO endpoint itself, but the incorporation of a validated PRO measure into prospective patient stratification and treatment evaluation, generating insights into disease heterogeneity that may not have been apparent in a pooled analysis. Such classification and assessment provide potential MIDD‐based pathway to assess phenotype‐based exposure–response relationship for responders using mechanism‐based PD biomarkers linked to distinct downstream clinical outcomes. Prior to evaluation of dazodalibep in Sjögren's disease, MIDD including population PK analysis has supported the bridging of PK from patients with RA to Sjögren's disease, as well as PK/PD analysis to support Phase 3 doses for the ongoing Phase 3 trials. ⁶⁰

Application of Mechanistic PBPK/PD Framework to Predict Clinical PK in Participants With Organ Impairment

Rare disease drug development often requires greater reliance on model‐informed approaches because clinical data are intrinsically limited by small patient populations, disease heterogeneity, and the practical or ethical constraints of conducting dedicated studies in all relevant subgroups. This reliance becomes especially important for novel modalities that may not be feasible or informative to study in healthy volunteers, where pharmacology may depend on target expression, tissue distribution, intracellular uptake, or disease‐relevant biology.

Nucleic acid therapeutics, including siRNA and antisense oligonucleotides, are increasingly represented in rare disease pipelines, particularly for genetically defined disorders. For these modalities, plasma concentrations may be transient and may not fully reflect tissue pharmacology; therefore, dosing is often guided by target mRNA knockdown and downstream protein or biomarker changes rather than systemic PK alone. In this context, mechanistic PBPK/PD models can provide an important quantitative bridge from preclinical and limited clinical data to support clinical study design, dose selection, and subgroup assessment.

GalNac‐based siRNAs are designed to specifically bind to the asialoglycoprotein receptor (ASGPR), which is predominantly expressed in the liver. ⁶¹ This receptor plays an important role in targeted delivery and gene‐silencing effects downstream. ⁶² Since dedicated organ impairment studies for GalNAc‐conjugated siRNA's are limited or lacking, a prospective mechanistic modeling can inform clinical study design and dosing decisions, thus accelerating efficient development of siRNA therapeutics.

In a PBPK/PD based‐modeling framework, the authors used inclisiran and vutrisiran as case studies to develop and calibrate a modality‐specific PBPK/PD platform in cynomolgus monkeys to evaluate the effect of organ impairment on the PK and PD of siRNAs. ⁶³ The model incorporated key intracellular and tissue‐level processes that govern siRNA disposition and gene silencing, including ASGPR‐mediated hepatic uptake, endosomal trafficking, limited endosomal escape, RISC loading, and RISC‐bound siRNA turnover. Translation of the model from monkeys to humans successfully captured clinical plasma PK and PD biomarker time courses, demonstrating that preclinical tissue‐level data can be quantitatively leveraged to predict human efficacy drivers for this modality. Importantly, this translational PBPK/PD approach enabled a mechanistic evaluation of hepatic and renal impairment without reliance on extensive dedicated clinical studies. The modeling showed that clinically observed changes in plasma exposure under organ impairment did not translate proportionally to pharmacodynamic effects, as PD is driven by intracellular hepatic siRNA and RISC kinetics rather than systemic PK. In addition, the model also showed that reductions in ASGPR alone are insufficient to alter PD, with modest changes in functional hepatocyte mass and hepatic uptake clearance potentially explaining observed clinical trends. ⁶³

Thus, although PBPK/PD approaches to organ impairment are not unique to rare diseases, this example illustrates their particular value in rare disease development: They support evidence generation for novel modalities when clinical datasets are small, healthy‐volunteer studies may be infeasible, and dedicated subgroup studies are limited. This PBPK‐PD‐based platform model, developed integrating kinetic and mechanistic understanding at cellular, tissue and physiological level, provided strong quantitative justification for the lack of dose adjustment across organ‐impaired (hepatic and renal) populations for inclisiran and vutrisiran. ⁶³

Application of Non‐Traditional Adaptive and Hybrid Clinical Trial Designs for Drug Repurposing

“Traditional” or “conventional” drug development generally consists of a sequential process starting with a FIH study to establish safety and to understand the PK of a drug, followed by proof‐of‐concept/dose‐finding Phase 2 studies, and subsequently conducting two adequate, well‐controlled confirmatory Phase 3 studies demonstrating effectiveness and safety. ¹ Although evidentiary expectations are broadly similar for rare and common diseases, this sequence is often difficult to execute in rare diseases. Low prevalence, geographic dispersion of patients, few expert centers, heterogeneous presentation, and variable disease progression can make recruitment slow and sometimes infeasible. Multiple sequential trials or extensive dose exploration may therefore use a substantial portion of the available patient population without clearly improving the benefit–risk assessment. Hybrid, adaptive, and model‐informed designs have emerged as both exploratory and confirmatory stages of development. ⁶⁴ These designs are useful in this setting not because they are exclusive to rare diseases, but because they address constraints that are more acute in rare‐disease development. These designs can reduce exposure to noninformative doses or prolonged placebo treatment, preserve scarce patient resources, and allow greater learning from each participant through integrated efficacy, safety, PK, PD, and biomarker data. Innovative and alternative trial designs, such as parallel dose‐ranging designs, seamless Phase 1/2 or Phase 2/3 designs, cross‐over trials, n‐of‐1 trials, randomized placebo‐phase trials, and randomized withdrawal designs with “enrichment strategies” have been evaluated for rare diseases. ⁶⁴

Drug development is a time‐consuming, resource‐intensive, and complicated process, and for rare diseases, this process becomes more challenging. Given the extremely small sample size, varying clinical manifestations, and variable course of disease progression, it is imperative to develop an alternative and efficient strategy to augment development of therapeutics for treatment of rare diseases. However, developing de novo drug therapies for extremely small patient populations is particularly challenging if the developed therapy seems to work in only a subset of the patient population, leaving a persistent unmet need for the broader patient population. ⁶⁵ In this context, drug repurposing, also called drug repositioning, offers an alternative. ⁶⁶ Unlike routine indication expansion, where prior experience is often used to broaden development into a related population or therapeutic area, repurposing in rare diseases can reduce the need for extensive new clinical experimentation. Here, a drug previously evaluated in a therapeutic area or an indication is redeveloped or reinvestigated for a new unrelated therapeutic area or indication. This helps conserve time and resources, thereby accelerating development. ⁶⁶ The central theme to drug repurposing is that the target, for example, receptor, pathway, or disease‐driving mechanism, is shared between the original and new indications. When biologically plausible, clinical pharmacology‐based modeling and simulation can translate prior knowledge to the rare‐disease setting. Prior safety, tolerability, PK, target‐engagement, and exposure–response information can be integrated with disease biology to establish a target concentration and support dose and regimen selection for pivotal efficacy trials. This does not lower the evidentiary standard; rather, it provides a fit‐for‐purpose way to address key uncertainties, including whether prior PK and safety data apply to the new population, whether the mechanism is relevant in the rare disease, and whether the selected dose will achieve pharmacologically active exposures.

Teprotumumab presents an example of a successful drug repurposing wherein the drug, first evaluated and developed for cancer, was repurposed for the treatment of thyroid eye disease (TED).

Teprotumumab

Teprotumumab is an insulin‐like growth factor‐1 receptor (IGF‐1R)–inhibiting mAb. The initial clinical development of teprotumumab evaluated it as a therapeutic candidate for the treatment of solid tumors. However, the development program was terminated due to lack of efficacy. ⁶⁷ Teprotumumab was also evaluated for the treatment of diabetic macular edema (NCT02103283), but the program was stopped ⁶⁸ as well. Following this, teprotumumab was repurposed for active TED and became the first drug available for the treatment of active TED in 2020. ⁶⁹

The approved dosing regimen for TED is an initial intravenous (IV) infusion of 10 mg/kg followed by 20 mg/kg every 3 weeks for an additional seven infusions. This dosing regimen was first evaluated in a Phase 2 study ⁷⁰ in adult patients with active, moderate‐to‐severe TED, followed by a confirmatory Phase 3 study. ⁷¹ The selection of this dosing regimen was based on the PK data collected from the Phase 1 dose‐escalation study (doses: 1 to 16 mg/kg) in oncology patients with solid tumors (NCT00400361 and NCT00642941), and the findings from an initial population PK analysis that characterized the nonlinear PK of teprotumumab, and identified a concentration that would result in >90% saturation of IGF‐IR of target‐mediated clearance of teprotumumab—a concentration that became the target trough concentration of teprotumumab in patients with TED. ⁷² From a rare disease perspective, it is noteworthy that no dose‐ranging studies were conducted in patients with TED. Given the rarity of the disease and limited pool of eligible patients, only one dose and dosing regimen was evaluated in Phase 2 and Phase 3 studies. Prior oncology PK data, IGF‐1R target‐mediated disposition, and population PK modeling were used to select a biologically justified regimen while avoiding a conventional multi‐dose TED dose‐ranging program, this illustrating the application of quantitative tools, such as population PK approach, using data from an early‐phase program in oncology patients and adapting the findings to estimate target occupancy and guide dose selections for drug repurposing in other disease populations. ⁷² Importantly, the PK and target‐saturation information generated in oncology patients were translated to patients in TED. The approved dose of teprotumumab was justified by the overall clinical safety and efficacy and the lack of a meaningful exposure–response relationships for both efficacy and safety endpoints, including all patients in the Phase 2/3 studies. ⁷³

Thus, teprotumumab illustrates how MIDD‐enabled repurposing can be particularly useful in rare diseases: prior human experience can reduce redundant clinical experimentation, focus limited enrollment on the most informative regimen, and support dose selection when conventional dose‐ranging is not practical.

Quantitative Assessment of Clinically Relevant Immunogenicity on Drug Exposures

It is a well‐established concept that binding and/or neutralizing ADAs can meaningfully alter the PK, PD, efficacy, and safety of therapeutic proteins and other biologics. Immunogenicity is not unique to rare diseases but its assessment can be particularly challenging in rare‐disease development given small sample size, substantial disease heterogeneity, and limited opportunities to conduct extensive clinical characterization. In these settings, regulatory and clinical decision‐making may sometimes need to rely on more limited immunogenicity data than would be typically available in larger development programs. However, this should not be interpreted as acceptance of uncharacterized immunogenicity risk. Instead, it underscores the importance to proactively characterize immunogenicity through fit‐for‐purpose assays, planned ADA sampling and serum exposure sampling, and integrated PK, PD, efficacy, and safety analyses whenever feasible.

Clinical pharmacology and MIDD approaches have been extensively used to assess the clinical impact of immunogenicity on PK, efficacy, and safety. The objective of these approaches is not only to report ADA incidence but also establish clinical relevance by quantitatively linking ADA formation to drug exposure, pharmacologic effect, and clinical outcomes. ⁷⁴ The example of pegloticase described below highlights the interaction between clinical pharmacology and ADA incidence, and its impact on efficacy. A “cotherapy” strategy of administering pegloticase with an immunomodulator, methotrexate (MTX), reduces the development of ADAs against pegloticase, thus increasing its efficacy. However, finding the right dose combination can be challenging. It is here that a strategy based in clinical pharmacology‐driven strategy, supported by quantitative tools and methodologies can rigorously help assess the impact of ADAs and inform selection of an appropriate combination dose and dosing regimen. This approach was first tested in a proof‐of‐concept study, before being confirmed in a pivotal trial.

Pegloticase

Pegloticase, a PEGylated recombinant uricase enzyme, is an FDA‐approved treatment for adult patients with uncontrolled gout. ⁷⁵ Pegloticase provides a useful example for rare‐disease drug development as it demonstrates how mechanistically informative clinical pharmacology data can support decision‐making when patient numbers are limited, unmet need is high, and residual uncertainty must be bounded and communicated. This case study provides an example of a framework wherein a downstream PD marker serum urate (sUA), routinely measured in patients, showed correlation to drug exposure and efficacy.

Monotherapy with pegloticase in pivotal Phase 3 trials demonstrated sustained urate‐lowering response in patients receiving pegloticase every 2 weeks, while a substantial proportion of patient population experienced secondary loss of response and infusion reactions. ⁷⁶ Analyses of clinical PK data, serum urate levels, and safety data established treatment‐emergent antibodies, predominantly directed against the PEG moiety rather than the uricase enzyme itself, as a potential cause for accelerated clearance of pegloticase, thus reducing systemic exposure and leading to loss of urate‐lowering efficacy and patient discontinuation. ⁷⁷ Hence, there was a need to mitigate ADA formation. Immunomodulation via co‐therapy had been evaluated in real‐world setting with drugs such as MTX ⁷⁸ and leflunomide. ⁷⁹ To establish an efficacy‐ and safety‐based proof‐of‐concept, MTX was concomitantly administered with pegloticase in an exploratory, open‐label trial in 14 patients (NCT03635957). Positive results from this study paved the way for a larger randomized controlled trial. ⁸⁰ A comparison of systemic exposures between pegloticase administered alone versus pegloticase + MTX demonstrated that pegloticase serum concentrations were higher in combination with MTX than the predicted median value of pegloticase administered as monotherapy, with ADA data consistent with pegloticase PK and efficacy (MTX co‐therapy significantly increased month‐6 response rates [71% vs 39%]). ⁸¹ Results from PK analyses demonstrated higher pegloticase exposures and lower incidence of anti‐PEG antibodies in patients with MTX co‐therapy. Available data support a relationship between pegloticase exposure (C_{min, ss}) and urate response, while no clinically meaningful exposure–safety relationship has been observed. ⁸²

The clinical development of pegloticase illustrates a proactive comprehensive framework for evaluation of immunogenicity when clinical datasets are small: Early identification of ADA risk, development and implementation of sensitive and specific bioassays, prospective collection of ADA and exposure data, and application of clinical pharmacology and quantitative approaches can help identify whether immunogenicity is mechanistically and clinically meaningful, inform mitigation strategies, and clearly describe remaining uncertainties. For rare‐disease drug development, this example reinforces that limited sample size and high unmet need do not reduce the expectation to characterize immunogenicity. Instead, they require a structured plan to anticipate ADA risk; prospectively measure ADA and drug exposures; interpret ADA‐related effects on PK, PD, efficacy, and safety; and clearly communicate the remaining limitations of the available evidence. These findings informed an evidence‐based, patient‐centric dosing and management strategies, including monitoring algorithms that directly shaped the prescribing information.

Application of Clinical Pharmacology–Based Approach to Bridge Formulations

Cysteamine Bitartrate

The original immediate‐release (IR) formulation, approved in 1994, effectively depleted lysosomal cystine accumulation in patients with nephropathic cystinosis, a rare autosomal recessive disorder caused by mutations in the gene for cystinosis, CTNS. However, its short plasma half‐life (∼90 min) required dosing every 6 h, leading to significant gastrointestinal intolerance and poor adherence. ⁸³ The pivotal randomized, open‐label crossover trial directly compared RP103, enteric‐coated, delayed‐release cysteamine bitartrate capsules, to the IR formulation. The primary endpoint was steady‐state white blood cell (WBC) cystine concentration, a validated surrogate biomarker, measured at 3 weeks instead of long‐term clinical outcomes. PK bridging demonstrated equivalent AUC between formulations at 70%–80% of the total IR daily dose. The E‐R analysis ⁸⁴ showed that plasma cysteamine concentrations above about 1 mg/L were associated with near‐maximal reduction of WBC cystine levels (E_max model). These findings demonstrated comparable pharmacologic activity between formulations through similar drug exposure and WBC cystine suppression. WBC cystine was an established surrogate of disease control and its relationship with cysteamine exposure was well characterized, thus, comparable PK and biomarker responses were considered sufficient to infer preservation of clinical benefit across formulations without additional long‐term efficacy studies.

The development of delayed‐release cysteamine bitartrate (RP103) represents an example of how clinical pharmacology–based approaches including bioequivalence studies, E‐R analyses, and surrogate endpoint bridging can support regulatory approval for a new formulation in a rare disease setting, where limited patient populations and urgent unmet medical need often make additional long‐term efficacy studies impractical.

Concluding Remarks

Clinical pharmacology has remained a critical component of drug development since the beginning of modern medicine. Despite such advancements, safe and effective therapies for rare diseases are still an unmet need. In the 43 years since the enactment of the Orphan Drug Act in 1983 by the US FDA, there have been 1122 total approvals which included approval of new molecules, indications, and formulations, with 882 initial approvals (i.e., first approval falling under one orphan designation) representing at least 392 rare diseases. ⁸⁵ Clinical pharmacology–based modeling and simulation methodologies play an important and foundational role in integrating the understanding of the rare disease biology, its progression, clinical presentation, and efficacy and safety outcomes to design an optimal dose and administration regimen, bridging not just across ethnicities, special populations including pediatrics, geriatrics, age, organ‐impaired populations, and formulations but also help bridge across indications. The examples presented in this article are just a few of the many where clinical pharmacology and quantitative tools were used to overcome the challenges encountered during clinical development and life‐cycle management of drugs for the treatment of rare diseases. There are still other methodologies not discussed in this article, which have played an important role in accelerating approval of drugs in rare diseases such as the use of model‐based meta‐analysis, an informative toolkit within the MIDD framework, which uses internal data along with external data to inform key drug development decisions, and the use of natural history and real‐world data, which not only helps characterize natural disease progression but can add to the overall sample size as control arms in clinical trials for rare and ultra‐rare disease populations, thus accelerating drug development.

Conflicts of Interest

All authors are employees and shareholders of Amgen Inc.

Funding

All authors are employees of Amgen Inc. and may receive stocks or stock options. This work received no additional funding.

Data Availability Statement

Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.