Nipocalimab Phase 3 Dose Selection for Severe Hemolytic Disease of the Fetus and Newborn

Abstract

Nipocalimab, a neonatal Fc receptor (FcRn) blocker, is under evaluation for severe hemolytic disease of the fetus and newborn (HDFN). In the Phase 2 UNITY trial, weekly intravenous antenatal treatment with nipocalimab at dose regimens of 30 and 45 mg/kg prevented fetal anemia requiring intrauterine transfusion (IUT) in 54% of high‐risk pregnancies and delayed the need for IUTs versus their previous pregnancies in the remaining 46% of pregnancies. This analysis aimed to select a weekly dose regimen of nipocalimab for the Phase 3 study in severe HDFN (NCT05912517) that maintains FcRn blockade throughout antenatal treatment, including with an unplanned dosing delay of up to 3 days. Observed pharmacokinetic/pharmacodynamic (PK/PD) data from UNITY (i.e., nipocalimab concentrations, FcRn occupancy, and serum IgG) were analyzed using a model‐based approach. A PK/PD model originally developed in nonpregnant participants was updated to incorporate gestational weight gain. Nipocalimab PK and FcRn occupancy were described by a two‐compartment model with nonlinear, dose‐dependent PK, which captured longitudinal PK, FcRn occupancy, and IgG profiles during dosing and return toward baseline postpartum after discontinuation. Both 30 and 45 mg/kg achieved ∼80%–85% reductions in maternal IgG; however, 30 mg/kg showed greater variability in predose trough concentrations, increasing the risk of falling below concentrations required for full FcRn occupancy across antenatal treatment. Simulations incorporating PK/PD variability indicated that 45 mg/kg weekly per current weight maintained full FcRn occupancy in >95% of pregnant individuals, even with dosing delays up to 3 days. Exploratory exposure–response analyses supported 45 mg/kg for the Phase 3 HDFN study.

Introduction

Nipocalimab is a fully human immunoglobulin G (IgG) 1 monoclonal antibody that blocks the neonatal Fc receptor (FcRn). ¹ , ² , ³ , ⁴ As FcRn is the sole receptor for transplacental IgG transfer and serum IgG salvage, FcRn blockade by nipocalimab can both inhibit transplacental IgG transfer and lower maternal serum IgG levels, including alloantibodies. ² , ⁴ These effects are supported by preclinical evidence from an ex vivo term placental perfusion model and Phase 2 clinical data. ² , ⁴ Nipocalimab is approved for generalized myasthenia gravis and under evaluation for multiple IgG antibody‐mediated diseases, including severe and early‐onset severe (EOS) hemolytic disease of the fetus and newborn (HDFN). ¹ , ² , ³ , ⁴ , ⁵

Severe HDFN is a rare disease caused by maternal IgG alloantibodies that cross the placenta through FcRn, causing fetal anemia requiring intrauterine transfusions (IUTs), fetal hydrops, postnatal transfusions, and significant risks of mortality. ⁶ , ⁷ , ⁸ EOS‐HDFN, where fetal anemia develops at ≤24 gestational weeks, carries the highest risk of complications and fetal mortality. ⁷ , ⁹ , ¹⁰ , ¹¹

Nipocalimab demonstrated preliminary efficacy and safety in a Phase 2, single‐arm, open‐label trial in 13 pregnant participants at high risk of EOS‐HDFN (UNITY; ClinicalTrials.gov Identifier: NCT03842189). ⁴ Weekly intravenous (IV) nipocalimab administrations at 30 and/or 45 mg/kg prevented fetal anemia requiring IUTs with live birth at ≥32 gestational weeks in 54% of cases, exceeding the clinically meaningful margin of 10% based on a historical data set (P < .001). ⁴ Recent retrospective studies reported that 88%–100% of the subsequent pregnancy after prior EOS‐HDFN required ≥1 IUTs. ¹² , ¹³ , ¹⁴ The median gestational age at first IUT in UNITY was 27 weeks, later than the 20–22 weeks reported in these retrospective studies. ¹² , ¹³ , ¹⁴ No unusual infections were observed in both maternal participants and infants despite an 80%–85% reduction in maternal, fetal, and neonatal IgG. ⁴ , ⁵ These findings supported a larger randomized, controlled, Phase 3 trial of nipocalimab in severe HDFN (AZALEA; NCT05912517). ¹⁵

Given the small UNITY sample size in this ultra‐rare EOS‐HDFN population, Phase 3 dose selection was based on observed pregnancy pharmacokinetics (PK) and pharmacodynamics (PD), supplemented by PK/PD model‐based simulations that account for PK/PD variability in larger populations. Analyses aimed to identify an optimal dose that maintains FcRn blockade across the weekly dosing interval and allows for potential real‐world dosing delays. Exposure–response analyses were also employed to assess PK/PD‐related trends in efficacy and safety. These analyses of the 30 and 45 mg/kg IV weekly regimens based on baseline or current body weight evaluated in UNITY, as well as simulated theoretical 15, 30, 45, and 60 mg/kg IV weekly dose regimens support 45 mg/kg IV weekly based on current body weight as the regimen to best achieve these aims.

Methods

Study Design

The protocol and informed consent were approved by the Ethics Committee for each of the participating sites, and written informed consent was obtained from all subjects before study participation. UNITY was an open‐label, single‐group, Phase 2 study assessing treatment with weekly IV nipocalimab (30 or 45 mg/kg) administered from 14 to 35 gestational weeks in 13 pregnant participants at high risk for recurrent EOS‐HDFN. Study details, eligibility criteria, and the efficacy and safety outcomes have been previously published. ⁴ , ⁵ Briefly, eligible participants were adults (≥18 years) with a prior pregnancy complicated by severe fetal anemia, hydrops, or stillbirth with pathology consistent with HDFN, and maternal anti D titers ≥32 or anti Kell titers ≥4. ⁴

The initial study dose was 30 mg/kg based on baseline body weight (at treatment start) with protocol‐allowed, data‐driven increases in dose for all current and subsequent participants. Altogether, four dose groups were evaluated: 30, 30 and 45, or 45 mg/kg IV weekly, with the total dose calculated based on baseline body weight; or 45 mg/kg IV weekly, with the total dose calculated based on body weights measured every 2 weeks during antenatal treatment beginning at 14 gestational weeks through 35 gestational weeks (Figure S1). ⁴ Maternal PK, FcRn occupancy (FcRn blockade ³ , ⁵ , ¹⁶ ), and IgG were assessed at baseline and all predose trough visits. Maximum concentration of nipocalimab was assessed 45 ± 15 min after infusion at protocol‐specified visits across the treatment period. PK/PD measures were also assessed at weekly intervals after the last dose, including at delivery and at less frequent intervals postpartum.

Serum PK, FcRn Occupancy, and Total IgG Concentration Assays

Serum PK was measured using a validated ELISA assay employing paired anti‐nipocalimab antibodies for capture and detection in three healthy volunteer studies, ¹ , ² , ³ and using a validated sandwich electrochemiluminescence immunoassay on the Meso Scale Discovery platform in one healthy volunteer study (NCT04848558) and the HDFN UNITY study. ⁴ Cross‐validation analysis of identical serum samples analyzed by both methods demonstrated equivalent nipocalimab concentrations. FcRn occupancy was assessed in both healthy volunteer and UNITY studies using a fluorescence‐activated cell sorting assay that measured free FcRn in permeabilized monocytes from whole blood via fluor‐labeled nipocalimab. ⁵ Serum total IgG concentrations for all healthy volunteer and HDFN studies were determined using in vitro diagnostic tests performed at Clinical Laboratory Improvement Amendments of 1988–certified laboratories.

PK Model Development

The population PK data of nipocalimab in HDFN were characterized using an established quasi‐steady‐state target‐mediated drug disposition model originally developed in 114 nonpregnant healthy volunteers who received a single nipocalimab dose of 0.3, 3, 10, 30, or 60 mg/kg IV, or 300 or 1200 mg IV; multiple doses of 15 or 30 mg/kg IV weekly for 4 doses; or IV placebo infusion. ³ , ¹⁷ , ¹⁸ This model was updated with pooled EOS‐HDFN data from 14 pregnancies to account for gestational age–related physiological changes by incorporating time‑varying body weight as a covariate on clearance and volume parameters to create a mathematical model reflecting pregnancy‑related physiological change (Figure 1a). The model included two‐compartment disposition that simultaneously characterized time courses of nipocalimab serum concentrations and FcRn occupancy. The weight variation during pregnancy was used as a covariate in the model, where clearance (CL) and volume of distribution of the central compartment (V_c) increased with weight gain and weights up to 150 kg were evaluated by simulation. ¹⁹ Model equations, parameter estimates, goodness‐of‐fit plots, and visual predictive checks for the final nipocalimab HDFN PK model are provided in the Supplemental Methods.

Figure 1.

Nipocalimab (a) PK model^a and (b) IgG model^b scheme. FcRn, neonatal Fc receptor; F_free, fraction of unoccupied receptors versus baseline; IgG, immunoglobulin G; IV, intravenous; k_deg, degradation rate constant of the free receptors; k_{deg_IgG}, degradation rate constant of IgG without FcRn‐mediated recycling; k_int, internalization rate constant of drug‐receptor complex; k_off, dissociation rate constant; k_on, binding rate constant; K_{rec_IgG}, FcRn‐mediated IgG recycling rate when free FcRn is fully available at time 0; K_syn, synthesis rate of the receptor; K_{syn_IgG}, synthesis rate of IgG; PD, pharmacodynamics; PK, pharmacokinetics; Q, intercompartmental flow. The PK and IgG model scheme accounts for several components of nipocalimab PK‐PD behavior. (a) Nipocalimab binding to and disassociation from FcRn is dependent upon nipocalimab concentrations in the central compartment (plasma) which partitions (Q) between the central (plasma) and peripheral compartment (tissues); the total pool of FcRn is determined by its endogenous synthesis and degradation rates. (b) The pool of FcRn not bound by nipocalimab (percent unoccupied FcRn) mediates IgG recycling and together with endogenous IgG synthesis and clearance determine the total plasma IgG concentration. This model also accounts for increased nipocalimab clearance at low nipocalimab concentrations (i.e., target‐mediated disposition), where the total mass of nipocalimab nears that of total FcRn pool and nipocalimab‐FcRn complexes are subject to degradation due to the short endogenous half‐life of FcRn and IgG recycling returns with new FcRn synthesis. ^aModel equations for the final nipocalimab PK model are provided in Equations 1, 2, and 3 (Supplemental Methods). ^bThe indirect response model equations for the F_free/IgG relationship are provided in Equation 4 and Equation 5 (Supplemental Methods).

IgG Model Development

The nipocalimab updated PK‐IgG model (for pregnancy) used a sequential modeling approach to describe observed longitudinal IgG data (healthy volunteers N = 130; pregnancies N = 14) following IV nipocalimab administration through an indirect response model that characterizes the total serum IgG turnover. In this model, the increase in FcRn occupancy following nipocalimab administration was assumed to decrease FcRn‐mediated IgG recycling, therefore increasing the total IgG degradation rate (Figure 1b), and no significant differences in PK/PD were observed between healthy and pregnant participants. Serum PK, FcRn occupancy, and total IgG concentrations were measured in the studies using validated bioanalytical assays ⁴ , ¹⁶ ; methodological details, model equations, parameter estimates, goodness‐of‐fit plots, and visual predictive checks for the final nipocalimab HDFN PK‐IgG model and FcRn occupancy/IgG relationship are provided in the Supplemental Methods.

Simulations of FcRn Occupancy with Dosing Delays

Based on the PK and IgG models, simulations of full FcRn occupancy when encountering unexpected dosing delays were conducted. Serum concentrations for 1000 pregnant individuals per dose regimen were simulated following 15, 30, 45, or 60 mg/kg IV weekly adjusted for weight during pregnancy from gestational Weeks 14–35. The dose selection was guided by the consistent reduction of maternal IgG, in addition to a consistent and full FcRn occupancy during dosing intervals. Weight gain during pregnancy was also considered in these simulations, where the simulated weight gain followed the same distribution as observed data (Figure S2). ¹⁹ Body weight at baseline was assumed to be sampled from a normal weight distribution (mean [SD]: 66 [15] kg). ²⁰

Exposure–Response Analysis

Exploratory exposure–response relationships were examined to correlate exposure (by concentration and tertile) with the primary efficacy and safety endpoints from UNITY. Exposure was defined as observed serum nipocalimab trough concentration or total IgG concentration at gestational Week 20, where the greatest number of samples were collected and nipocalimab or IgG concentration reached steady state. The primary efficacy endpoint was the proportion of pregnant participants with a live birth at ≥32 gestational weeks without an IUT. The selected safety endpoints included the proportion of maternal participants with ≥1 adverse event of infections requiring oral or IV antibiotic treatment and with ≥1 serious adverse event throughout the study, which were protocol‐prespecified and clinically significant. Infections not requiring treatment were excluded from this endpoint but were captured in the overall adverse events reported previously. ⁴

Software and Analysis

Population PK and PK/PD analyses were performed using the nonlinear mixed effects modeling software version 7.4 and higher (ICON plc, Hanover, MD) and the first‐order conditional estimation with interaction estimation option. Validated R version 4.2.0 or higher (Comprehensive R Network, http://cran.r‐project.org/) was used for all analyses in the study. Simulations were performed in mrgsolve version 1.0.6 in R version 4.2.0 or higher. This exploratory model‐based analysis was not powered to compare dose regimens or show superiority to historical controls.

Results

Study Populations

Demographic and baseline characteristics for the 14 UNITY pregnancies and over 100 nonpregnant healthy Phase 1 volunteers whose data informed the model are shown in Table 1. Among the pregnancy population, three pregnancies received the 30 mg/kg based on baseline body weight regimen, two received the 30 and 45 mg/kg based on baseline body weight regimen, five received the 45 mg/kg based on baseline body weight regimen, and four received the 45 mg/kg dose adjusted to current body weight regimen.

Table 1.

Demographics and Baseline Characteristics of Participants Included in the Population Analysis Data Sets for PK, PK/FcRn Occupancy, and IgG

Characteristic	PK/FcRn Occupancy Analysis Data Set	IgG Analysis Data Set
Number of participants/pregnancies, n	128	144
Pregnancies with HDFN	14 a	14 a
Healthy volunteers	114	130
Number of observations, b n	PK: 1513 c FcRn occupancy: 1209	2110
Median (range) age, years	35 (18–63)	35 (18–64)
Median (range) body weight, kg	71.0 (50.3–110)	70.5 (50.3–110)
Female, n (%)	83 (64.8)	95 (66.0)
Race, n (%)
White	97 (75.8)	106 (73.6)
Asian	21 (16.4)	27 (18.8)
Black	6 (4.6)	6 (4.2)
Other	4 (3.1)	5 (3.5)
Positive for antibodies to nipocalimab, n (%)	50 (39.1)	53 (36.8)

FcRn, neonatal Fc receptor; HDFN, hemolytic disease of the fetus and newborn; IgG, immunoglobulin G; PK, pharmacokinetics.

^a A total of 14 pregnancies from 13 pregnant participants in the UNITY study were included in the PK/FcRn occupancy/IgG analysis.

^b Represents the total number of samples across timepoints included in the data set.

^c A total of 514 samples was below the limit of quantification.

Observed Nipocalimab PK and PD

Maternal PK demonstrated dose‐dependent increases in both trough and maximum nipocalimab concentrations (Figure S3), ⁵ with trough concentrations declining over gestation, particularly in the 30 mg/kg dose group. However, all trough concentrations, except for one timepoint in the 30 mg/kg group, remained above the 10 µg/mL FcRn blockade threshold. ³ , ⁵ , ¹⁶

FcRn occupancy analyses indicated sustained blockade across weekly trough timepoints in all pregnant participants (Figure 2a). These results are consistent with the observed sustained 80%–85% reductions in serum IgG during antenatal nipocalimab treatment in UNITY and in nonpregnant Phase 1 participants. ³ , ⁴

Modeling of Nipocalimab PK, FcRn Occupancy, and IgG Lowering

The nipocalimab PK and PK/PD relationship has been shown to be consistent across nonpregnant populations in Phase 1 studies. ³ , ¹⁷ , ¹⁸ A PK/PD model was developed based on these data (Figure 1). Maternal PK and total IgG data from the Phase 2 UNITY study were then incorporated to update the model for gestational age–related physiological changes, in which time‑varying body weight was incorporated as a covariate on CL and V_c parameters to create a mathematical model reflecting pregnancy‑related physiological change.

The incorporation of pregnant participants at risk of EOS‐HDFN into the PK/PD model was supported by the high degree of consistency between PK profiles, FcRn occupancy, and IgG lowering observed in UNITY and in Phase 1 nonpregnant participants. The overlap between observed data in pregnant participants and model simulations for nipocalimab PK, FcRn occupancy, and IgG lowering also support the model's ability to characterize nipocalimab PK/PD (Figure 2).

Figure 2.

Overlaid simulated and observed (a) maternal serum PK, (b) percent unoccupied FcRn, and (c) percent change from baseline in maternal serum IgG after nipocalimab administrations. FcRn, neonatal Fc receptor; IgG, immunoglobulin G; PK, pharmacokinetics. Colored points are the observations, and colored lines and shaded areas are the model predictions and 90% prediction intervals. The yellow shaded area shows the threshold range of nipocalimab concentrations that correspond to full FcRn occupancy. For observed data, dosing start time was remapped to gestational Week 14, and time of last dose was remapped to gestational Week 35 to be overlaid with the simulation profiles.

Simulations of the 30 mg/kg based on baseline body weight regimen predicted trough PK <10 µg/mL FcRn blockade threshold with gestational age–associated weight gain, consistent with a single observed value <10 µg/mL in this group (Figure 2a). In contrast, simulations and observed data for 45 mg/kg adjusted to current body weight demonstrated consistently high trough PK values well >10 µg/mL. Simulations for 45 mg/kg based on baseline body weight are not shown, but the observed trough data exhibited greater variability and a downward trend with gestational age.

Simulated and observed FcRn occupancy indicated maintenance of FcRn blockade, and IgG lowering was sustained at 80%–85% below baseline during treatment (Figure 2b,c). After the last dose at gestational Week 35, both observed and simulated results indicated that nipocalimab was eliminated within 1–2 weeks, paralleling loss of FcRn occupancy (as early as gestational Week 36 in the 30 mg/kg based on baseline weight group). IgG recovery occurred in tandem with FcRn occupancy recovery, approaching baseline within 5 weeks after the last dose.

Additional simulations (not shown) were conducted for participants weighing up to 150 kg (approximately 330 lbs), including weights representative of morbid obesity (BMI >40, depending on height). Results demonstrated that, under the proposed 45 mg/kg dosing regimen for the Phase 3 study, predicted nipocalimab exposures remained within approximately 20% of those observed in a reference 70 kg participant.

Model Simulations of FcRn Occupancy with Dosing Delays

Model simulations incorporating baseline body weight and gestational weight gain across a large virtual population were used to identify the optimal nipocalimab dose for maintaining FcRn blockade throughout the weekly dosing interval and during potential dosing delays of up to 3 days. Simulated dosing delay scenarios showed that the percentage of pregnant participants with full FcRn occupancy would rapidly drop with nipocalimab dose regimens <45 mg/kg IV weekly due to the short half‐life and target‐mediated drug disposition of nipocalimab (Table 2). With a dosing delay of 3 days, the simulated percentage of pregnant participants with full FcRn occupancy was 5%, 71%, 96%, and 99% in the 15, 30, 45, and 60 mg/kg IV weekly dose groups, respectively. These results suggest that ≥45 mg/kg IV weekly doses of nipocalimab are more resilient to dosing delays and more suitable than the lower dose regimens.

Table 2.

Simulated Percentage of Maternal Participants with Full FcRn Occupancy

	Simulated Percentage of Maternal Participants with Full FcRn Occupancy
Dosing Delays, days	15 mg/kg IV Weekly a	30 mg/kg IV Weekly a	45 mg/kg IV Weekly a	60 mg/kg IV Weekly a
0	47	99.1	99.9	100
1	24.4	96.3	99.8	99.9
2	11.3	86.1	98.4	99.8
3	4.7	71	95.5	98.8

FcRn, neonatal Fc receptor; IV, intravenous.

^a Dose adjusted for body weight throughout antenatal treatment.

Exposure–Response Relationship

Exploratory exposure–response relationships were examined to correlate exposure (by concentration and tertile) with the primary efficacy and safety endpoints from UNITY. Nipocalimab across 30 and/or 45 mg/kg dose regimens achieved the primary efficacy endpoint: 54% of pregnancies resulted in live birth at ≥32 gestational weeks without an IUT versus 10% observed in a historical data set (P < .001). ⁴ No unusual maternal infections were observed, and serious adverse events were consistent with HDFN, pregnancy, or prematurity with no discernible pattern or relationship to nipocalimab dose. ⁴

A plateau in the exploratory exposure–response relationship was observed with increasing exposure to nipocalimab (30 and 45 mg/kg IV weekly) for either the primary efficacy endpoint or safety endpoints (Figure 3 and Table S1). Higher PK exposure or greater IgG reductions within this dose regimen range were not associated with differences in efficacy or safety, supporting the 45 mg/kg IV weekly regimen as achieving target efficacy without any safety trends identified.

Figure 3.

Correlation of nipocalimab PK or maternal total IgG with the (a) primary efficacy and (b) safety endpoints. IgG, immunoglobulin G; IUT, intrauterine transfusion; PK, pharmacokinetics. Probability points represent a point estimate with 95% confidence intervals.

Discussion

Observed PK/PD data from UNITY indicated consistently higher predose trough nipocalimab concentrations with 45 mg/kg weekly IV nipocalimab dosing, whereas 30 mg/kg weekly IV dosing resulted in lower and greater variable trough PK concentrations, increasing the risk of losing FcRn blockade despite IgG lowering being maintained across dose levels. Model‐based simulated PK/PD results confirmed that the 45 mg/kg IV weekly nipocalimab regimen, adjusted for current body weight, achieved maintenance of full FcRn occupancy across antenatal treatment in contrast to lower dose regimens. Exploratory exposure–response relationships for both primary efficacy and safety endpoints across nipocalimab exposures plateaued, further supporting the 45 mg/kg IV weekly dose regimen for the Phase 3 trial in severe HDFN. ¹⁵

To our knowledge, this is the first PK/PD analysis using individual PK/PD clinical data of an FcRn blocker in pregnant participants and provides evidence for the potential use of nipocalimab in alloimmune pregnancy conditions, such as HDFN. The observed UNITY PK and PD results, as well as model‐based PK/PD and exposure–response analyses from the UNITY study, complement observations from other clinical studies of nipocalimab in multiple autoantibody‐mediated diseases, resulting in a consistent understanding of nipocalimab PK/PD behavior across indications and populations. ⁴ , ²¹ , ²² , ²³ , ²⁴ Across clinical studies to date, nipocalimab showed a rapid, substantial, sustained IgG reduction (up to 75%–85% depending on the dose regimen) that was associated with improved clinical outcomes, with a similar safety profile to those observed in Phase 1 healthy volunteer studies and Phase 2–3 studies across indications for which nipocalimab is being studied. ⁴ , ²¹ , ²² , ²³ , ²⁴ In UNITY, maternal IgG alloantibody titers decreased rapidly after treatment initiation with nipocalimab 30 and 45 mg/kg IV weekly regimens and increased after end of treatment similar to serum total IgG. ⁴ , ²⁵ Inhibition of transplacental alloantibody transfer by nipocalimab was also supported by low alloantibody titers and IgG in neonates when treated to near birth. ⁴ , ⁵ The maintenance of FcRn blockade reported here across all UNITY participants, together with PK/PD simulations, suggest that nipocalimab 45 mg/kg IV weekly regimens fully saturated FcRn, consistent with the decreased maternal total IgG and IgG alloantibodies and low fetal or neonatal alloantibody levels observed. ⁴ , ⁵ , ²⁵ Despite low IgG levels at birth, neonatal IgG concentrations followed the normal trend for physiologic nadir and were with the normal range for age though 2 years of life, with no unusual infections or impaired vaccine responses observed. ⁴ , ⁵

The PK/PD modeling results of this study also suggest that weight gain and other physiological changes should be considered when using nipocalimab in pregnant individuals, as these factors may impact nipocalimab PK, serum IgG levels, and transplacental IgG transfer. ²⁶ , ²⁷ , ²⁸ , ²⁹ , ³⁰ During pregnancy, an increase in blood volume, extracellular fluid, and organ size are observed, with an approximate 10–12 kg weight gain and 30%–45% plasma volume increase. ²⁶ Maternal IgG transfer also increases exponentially in the last trimester due to the higher expression of FcRn within the placenta and increased placental transfer activities. ²⁸ , ²⁹ , ³⁰ Therefore, nipocalimab doses adjusted for body weight during pregnancy would maintain sufficiently high concentrations over the weekly dosing interval to completely block FcRn throughout pregnancy.

Nipocalimab PK are governed by multiple processes, including its short half‐life and target‐mediated drug disposition, and the net impact has been empirically characterized through the clinical PK/PD analyses presented here. A dosing delay of even 1 day could impact nipocalimab concentrations and full FcRn occupancy. Simulations demonstrated that both the 45 and 60 mg/kg IV weekly regimens achieved full FcRn occupancy even with dosing delays up to 3 days. However, the safety, efficacy, and PK/PD for the 60 mg/kg IV weekly regimen has not been studied, whereas the reported UNITY results supported preliminary safety and efficacy of the 45 mg/kg IV weekly regimen, with no safety trends identified and 71% (5/7) of participants achieving live births at ≥32 gestational weeks without an IUT. ⁴ Both simulated and observed data support 45 mg/kg IV weekly as optimal, maintaining FcRn occupancy 2–3 days beyond the dosing interval and backed by preliminary safety and efficacy evidence, unlike the 60 mg/kg regimen. Further validation of these PK/PD models, alongside safety outcomes from larger Phase 3 trials, is warranted to characterize the effects of antenatal nipocalimab dosing in pregnancy.

These results enable the first pivotal prospective randomized controlled study in this high‐risk HDFN population with significant unmet need for safe and effective medical treatment to potentially delay or prevent fetal anemia, fetal hydrops, and poor neonatal outcomes as well as to avoid complications, burden, and risk of IUTs. ¹¹ This study advances our understanding of nipocalimab PK, PD, and mechanism in pregnancy. The consistency in nipocalimab PK and PD observed in nonpregnant and pregnant individuals, together with the applicability of the PK/PD model in pregnancy, allow for improved estimation of nipocalimab dose, PK, and PD across both populations and may support research in additional indications in pregnancy.

This study was strengthened by using a diverse data set from the Phase 2 UNITY trial, four Phase 1 studies, and previously validated PK/PD models to predict PK/PD relationships in pregnant individuals. The study limitations include the small number of pregnant participants with HDFN treated with nipocalimab, which restricts the interpretation of the exposure–response relationship. This model‐based analysis provides preliminary PK/PD and exposure–response insights that informed dose selection for the ongoing Phase 3 AZALEA trial but does not constitute definitive evidence of efficacy or safety. Additionally, this model‐based study did not determine whether the nipocalimab 45 mg/kg IV weekly regimen, adjusted for body weight throughout antenatal treatment, would enhance the prevention of severe fetal anemia in HDFN. These limitations could potentially be addressed using data being collected in AZALEA. By evaluating the range of exposures due to individual differences in PK and IgG levels, future exposure–response analyses can support whether the selected dose regimen achieves optimal clinical responses.

Conclusions

The 45 mg/kg IV weekly dose regimen, adjusted for body weight throughout antenatal treatment, is predicted to provide more consistent PK exposures, IgG reductions, and the majority of pregnant participants with FcRn blockade than lower doses, especially given potential for real‐world dose delays. Given the absence of an exposure–response relationship, along with the achievement of FcRn blockade, this 45 mg/kg IV weekly dose regimen is the recommended Phase 3 dose for delay or prevention of fetal anemia and avoidance of poor outcomes due to severe HDFN or invasive rescue treatment with IUT.

Author Contributions

Jie Zhou: Conceptualization, data curation, formal analysis, methodology, software, validation, visualization, writing – review and editing. Yuan Xiong: Conceptualization, data curation, formal analysis, methodology, software, writing – review and editing. Edwin Lam: Conceptualization, data curation, formal analysis, methodology, writing – review and editing. Leona E. Ling: Investigation, methodology, writing – review and editing. Shumyla Saeed‐Khawaja: Investigation, methodology, writing – review and editing. Yosuke Komatsu: Investigation, methodology, writing – review and editing. An Vermeulen: Conceptualization, methodology, supervision, writing – review and editing. Mahesh N. Samtani: Conceptualization, methodology, supervision, writing – review and editing. Jocelyn H. Leu: Conceptualization, methodology, supervision, writing – review and editing.

Funding

This study was sponsored by Johnson & Johnson.

Conflicts of Interest

Jie Zhou, Yuan Xiong, Leona E. Ling, Shumyla Saeed‐Khawaja, Yosuke Komatsu, An Vermeulen, and Jocelyn H. Leu are employees of Johnson & Johnson and may hold stock/stock options from Johnson & Johnson. Edwin Lam and Mahesh N. Samtani are former employees of Johnson & Johnson and may hold stock/stock options from Johnson & Johnson.

Paper Presentation Information

The findings were presented previously, in part, at the Society for Maternal Fetal Medicine (SMFM) Pregnancy Meeting; February 10‐14, 2024; National Harbor, MD, USA.

Supporting information

Supplemental Information

Data Availability Statement

The data sharing policy of Johnson & Johnson is available at https://www.jnj.com/innovativemedicine/our‐innovation/clinical‐trials/transparency. As noted in this site, requests for access to the study data can be submitted through Yale Open Data Access (YODA) Project site at http://yoda.yale.edu.