Safety, Pharmacokinetics, and Exposure–Response Modeling of Nedosiran in Participants With Severe Chronic Kidney Disease

Nedosiran is an investigational RNA‐interference therapeutic in development for primary hyperoxaluria (PH). Because nedosiran undergoes renal clearance, we assessed its pharmacokinetic profile in non‐PH participants with normal kidney function and Stages 4/5 chronic kidney disease (CKD), the latter with/without dialysis. Nedosiran exposure–response modeling in patients with PH Subtype 1 (PH1) with different renal function level was performed to recommend a nedosiran dose for this subpatient population. In this open‐label, single‐dose, Phase 1 study, 24 participants with estimated glomerular filtration rate <30 mL/min/1.73 m2 (CKD Stages 4/5; on hemodialysis [Groups 1a, 1b] and not on hemodialysis [Group 2]) and 10 participants with normal kidney function (estimated glomerular filtration rate ≥90 mL/min/1.73 m2; Group 3) received a single dose of subcutaneous nedosiran sodium 170 mg. Group 1a received nedosiran 8 hours before beginning hemodialysis, Group 1b received nedosiran 2 hours after completing hemodialysis; Group 2 was not on hemodialysis. Nedosiran population pharmacokinetic–pharmacodynamic analyses were conducted using pooled data from this study and 4 others. Nedosiran pharmacokinetic exposure in non‐PH participants with CKD Stages 4/5 was approximately 2‐fold higher versus participants with normal kidney function. Hemodialysis timing relative to nedosiran administration had no clinically significant impact on pharmacokinetics (Group 1a vs 1b). Nedosiran was well tolerated. Modeling indicated that in patients with PH1 with CKD Stages 4/5, lower nedosiran doses provide similar exposure and potential reduction in 24‐hour urinary oxalate to standard nedosiran doses in patients with PH1 with normal kidney function or CKD Stages 2/3. Nedosiran dosage reductions are recommended in patients with PH1 with CKD Stages 4/5; further adjustments are unnecessary if dialysis is started.

family of 3 known, genetically distinct, autosomal recessive, inborn errors of glyoxylate metabolism (Subtypes 1, 2, and 3 [PH1, PH2, and PH3]) characterized by the overproduction of hepatic oxalate, a metabolic end product that is eliminated almost exclusively by the kidneys. 1,22][3] Patients with PH who develop severe chronic kidney disease (CKD) are at increasing risk of extrarenal calcium oxalate accumulation (systemic oxalosis), a devastating complication associated with disability, poor quality of life, and multiple organ failure. 1,2,4,5n the Rare Kidney Stone Consortium PH Registry, the median estimated glomerular filtration rate (eGFR) at diagnosis was 48, 83, and 96 mL/min/1.73m 2 for patients with PH1, PH2, and PH3, respectively. 6By age 40 years, 63.8%, 34.2%, and 2.9% of patients with PH1, PH2, and PH3, respectively, progressed to kidney failure. 6Overall, PH accounts for 1% of pediatric kidney failure cases according to registries from Europe, the United States, and Japan. 7stablished treatment options for patients with PH, including hyperhydration, inhibitors of calcium oxalate crystallization, hemodialysis, and peritoneal dialysis, have difficulty in controlling plasma oxalate levels over the long term. 2 RNA-interference (RNAi) agents are an emerging class of PH therapies, of which lumasiran was approved for the treatment of patients with PH1. 8 It is plausible that the availability of lumasiran and other innovative therapeutics for PH1 may obviate the need for combined liver and kidney transplantation, hitherto the only potential lifesaving option in this subtype. 1,2,9edosiran is a liver cell-targeted investigational RNAi therapeutic comprising a double-stranded small interfering RNA oligonucleotide conjugated to N-acetyl-D-galactosamine (GalNAc) aminosugar residues. 10The GalNAc moieties bind to the asialoglycoprotein receptor in the liver and the entire conjugate is internalized.In the hepatocyte, nedosiran exploits the endogenous RNAi mechanism to silence the expression of the lactate dehydrogenase type A (LDHA) gene. 11s catalysis of glyoxylate to oxalate by LDH represents the last step of hepatic oxalate production, [11][12][13][14][15][16] nedosiran-induced silencing of hepatic LDHA may have clinical utility in the treatment of PH. 11,[17][18][19] The 24-hour urinary oxalate (Uox) excretion is the most important biomarker for clinical trials in PH, because lower 24-hour Uox excretion correlates with improved prognosis in PH, whereas high Uox excretion predicts poor kidney outcome. 20,21Population pharmacokinetic (PK)-pharmacodynamic (PD) modeling and simulation of PHYOX1 (randomized, single-ascending-dose, Phase 1 clinical trial; NCT03392896) data indicated that a monthly nedosiran subcutaneous dose of 170 mg (equivalent to 160 mg free acid [all doses hereinafter pertain to the sodium salt]) enables substantial reductions in 24-hour Uox excretion. 17Additionally, in the subsequent randomized, placebo-controlled, 6month PHYOX2 trial (randomized, placebo-controlled multiple-dose, Phase 2 study; NCT03847909), monthly nedosiran doses of 170 mg produced clinically meaningful reductions in 24-hour Uox excretion among patients with PH1. 18ubcutaneously administered nedosiran is systemically absorbed and reaches peak concentration (C max ) 6−12 hours after dosing. 17Data from PHYOX1 indicate that nedosiran is partially (approximately 26.6% of the administered dose) eliminated, unchanged by the kidney in addition to hepatic metabolism in humans. 17Patients with an eGFR less than 30 mL/min/1.73m 2 or on current dialysis were ineligible for inclusion in PHYOX1 and PHYOX2 and PHYOX3 (long-term open-label extension Phase 3 study; NCT04042402). 17,18Therefore, nedosiran disposition in patients with decreased kidney function requires characterization.Therefore, this Phase 1 study (PHYOX5) was conducted in healthy volunteers and non-PH patients with CKD Stages 4/5 (eGFR less than 30 mL/min/1.73m 2 ) to evaluate the effect of severe CKD (with and without dialysis) on the plasma and urine PK of single-dose nedosiran.The PK data obtained from this study, along with PK-PD data from PHYOX1, PHYOX2, and PHYOX3, as well as PK data from PHYOX6 (Phase 1, open-label, single-dose, ethnobridging study of nedosiran in healthy Japanese and White volunteers), were used in modeling and simulation to select a nedosiran dose regimen for patients with PH1 with different stages of CKD.

Study Design
This Phase 1, open-label, single-dose study (PHYOX5) was performed at 2 inpatient clinical study sites (Clinical Pharmacology of Miami, Miami, FL; and Orlando Clinical Research, Orlando, FL).The study was approved by an independent institutional review board (IntegReview IRB, Austin, TX) and conducted in compliance with ethical principles of Good Clinical Practice according to the International Conference on Harmonization Tripartite Guideline.The primary objective was to evaluate the PK profile of nedosiran in participants with CKD Stage 4 (severe renal impairment, eGFR 15-29 mL/min/1.73m 2 ) and Stage 5 (kidney failure or end-stage kidney disease eGFR less than 15 mL/min/1.73m 2 ), with or without dialysis relative to healthy volunteers with normal kidney function (eGFR of 90 mL/min/1.73m 2 or greater).The secondary objective was to assess the relative PK profile of nedosiran when administered before versus after dialysis in participants with CKD Stages 4/5.The PK data obtained from this study were also used to update a prior population PK-PD model 17 and thus determine an appropriate nedosiran dose for adult and adolescent (aged 12 years or older) patients with PH1 with CKD Stages 4/5 or severe kidney impairment/kidney failure.
Participants were screened within 28 days of nedosiran administration (Day 1) and remained in the study center from Day −1 until Day 5 (Figure S1).Nedosiran sodium 170 mg was administered to participants on Day 1, and serial PK samples were collected before and after dosing.The total duration of participation from screening to the end-of-study visit (Day 29) was approximately 8 weeks.Four treatment groups were defined for treatment and analysis as follows: Group 1a comprised participants with CKD Stages 4/5 receiving nedosiran 8 hours before beginning dialysis; Group 1b comprised participants with CKD Stages 4/5 receiving nedosiran 2 hours after completing dialysis; Group 2 comprised participants with CKD Stages 4/5 not on dialysis; and Group 3 comprised participants with normal kidney function.

Study Population
Consenting male and female participants aged 18-80 years with a body mass index (BMI) 18 to 39 kg/m 2 were eligible for enrollment based on their kidney function at screening.Allocation into one of the 4 treatment groups was guided by the eGFR using the Chronic Kidney Disease Epidemiology Collaboration equation. 22Groups 1a and 1b had an eGFR less than 30 mL/min/1.73m 2 (with at least 50% having an eGFR less than 15 mL/min/1.73m 2 ), Group 2 had an eGFR less than 30 mL/min/1.73m 2 , and Group 3 had an eGFR of 90 mL/min/1.73m 2 or greater.Participants in Groups 1a and 1b were mean matched on the basis of age (±10 years) and weight (±20%) and had similar proportions of men and women.Participants in Group 3 were matched with participants in Groups 1b and 2 on a 1:1 basis based on sex, age (±10 years), and weight (±20%).Hence, the same subject in Group 3 could have been matched to more than 1 subject (ie, 1 in Group 1b and 1 in Group 2).
The screening eGFR was confirmed on Day −1 of admission to the clinics.In Groups 1a and 1b, participants must have been stable on a dialysis regimen of at least 3 times a week for at least 3 months and have had a clinically stable condition with respect to the underlying kidney function impairment.In Groups 1a, 1b, and 2, participants must have had no significant changes in medications for at least 14 days prior to Day 1 and must have been willing to remain on the same stable dose(s) throughout the study.
Participants who smoked were limited to no more than 5 cigarettes each day through the end of study.Volunteers were ineligible if they had any condition or comorbidity that interfered with study compliance, data interpretation, or could potentially impact participant safety.Exclusion criteria included severe intercurrent illness; known causes of active liver disease/injury or transaminase elevation (eg, alcoholic liver disease, nonalcoholic fatty liver disease/steatohepatitis); intake of drugs of abuse or excessive alcohol intake; history of cholecystectomy or splenectomy or any clinically relevant surgery within 6 months prior to screening; clinically significant abnormal findings in serum chemistry or hematology results; seropositivity for antibodies to HIV, hepatitis C virus, and hepatitis B virus; poorly controlled or unstable hypertension; severe intradialytic hypertension; and abnormal 12-lead electrocardiograms (ECGs).

Interventions
Nedosiran sodium 170 mg (160 mg free acid equivalent, manufactured by Dicerna Pharmaceuticals, Inc., a Novo Nordisk Company) was administered as a subcutaneous injection into the abdomen or thigh.Participants in Groups 1a and 1b underwent a 4-hour hemodialysis procedure on Days 1, 3, and 5 using the Fresenius Optiflux F180NR high-flux dialyzer.
Concentrations of nedosiran in plasma, urine, and dialysate matrices at each time point were analyzed by an anion-exchange high-performance chromatography (AEX-HPLC) method with a DNA-Pac PA100 column (Thermo Fisher) and eluted using a gradient with mobile phases composed of 25 mM Trizma buffer pH = 8, 30% acetonitrile, 1 mM ethylenediaminetetraacetic acid in water at a 1-mL/min flow rate.Detection was attained by the use of a fluorescence detector (20Axs; Shimadzu).The assay is based on the specific hybridization of the antisense strand of nedosiran with a complementary peptide nucleic acid strand.The peptide nucleic acid is labeled with an Atto425 fluorescence dye and yields a specific signal in the subsequent analysis by AEX-HPLC coupled to a fluorescence detector.The method was developed without the need for analyte extraction.Therefore, a recovery of 100% is maintained during the process and no internal standard is needed.The lower limit of quantitation (LLOQ) of the AEX-HPLC method was 1.0 ng/mL.

Pharmacokinetic Analysis
The PK analysis was performed on all participants who received nedosiran and for whom enough PK data points (ie, 3 or more consecutive postdose concentrations) were available to allow the calculation of parameters based on actual sampling times.Individual plasma concentration-time data were used to calculate nedosiran PK parameters using standard, noncompartmental methods with the R package "maNCA" (R Foundation for Statistical Computing).Systemic exposure to nedosiran was represented as C max and area under the plasma concentration-time curve (AUC) from the time of dosing to the last quantifiable plasma concentration (AUC 0-t ).AUC from the time of dosing to infinity (AUC 0-inf ) was calculated, if possible.
Urine concentrations of nedosiran for each nominal collection interval relative to the time of nedosiran administration were used to determine the following urine PK parameters: cumulative amount of nedosiran excreted in the urine (Ae); amount of nedosiran excreted in urine over a given time interval (Ae t1-t2 ) (mg); fraction of unchanged drug excreted in the urine (fe) represented as a percentage of dose (fe[%dose] = Ae/dose*100); and renal clearance from the time of dosing to the last quantifiable plasma concentration calculated as Ae 0-t /AUC 0-t (L/h).
Hemodialysis clearance (CLHD) was calculated as follows: where QBIN is the blood flow into the dialyzer, R is the blood to plasma drug concentration ratio, f (dialysis) is the fraction of the drug eliminated during dialysis, calculated as where AUC τ_in is the AUC in the dialysis interval based on the samples collected at the entry of the dialyzer, and AUC τ_out is the AUC in the dialysis interval based on samples collected at the exit of the dialyzer.

Safety
Safety was evaluated by adverse event (AE) reporting, and performing clinical laboratory tests, changefrom-baseline 12-lead ECGs, vital sign measurements, physical examinations, and CLHD.All AEs, including treatment-emergent AEs, serious AEs, AEs leading to discontinuation, and treatment-related AEs were summarized for each treatment group using the Medical Dictionary for Regulatory Activities Version 22.1 coding system by system organ class, preferred term, relationship to nedosiran, and severity.A protocoldefined injection-site reaction (ISR), a disorder characterized by an intense adverse reaction (potentially immunologic) developing at the site of injection, was an AE of special interest.Individual signs or symptoms at the injection site (eg, erythema, swelling) reported within 4 hours of study intervention administration were recorded as AEs, not as ISRs, and graded in accordance with intensity categories.Signs or symptoms at the injection site occurring more than 4 hours after dosing were evaluated according to the Common Terminology Criteria for Adverse Events Version 5.0 criteria for ISR.Participants were also monitored for signs and symptoms of muscle weakness or pain, in addition to measurement of plasma creatine kinase at the time of or after single-dose nedosiran administration.

Data and Statistical Analysis
PK data were analyzed using R Version 3.2 or higher, and safety data were analyzed using SAS (SAS Institute Inc.).Plasma concentrations below the LLOQ were treated as missing if they occurred after the C max in PK parameter estimates but as zero while plotting the mean concentration-time profiles.PK parameters were estimated using noncompartmental analysis.The potential effects of CKD Stages 4/5 and dialysis on nedosiran PK exposure were assessed using an analysis of variance (ANOVA).The geometric least-square mean ratios and 90% confidence intervals (CIs) for C max and AUC 0-t between participants with CKD Stages 4/5 with or without dialysis and participants with normal kidney function were estimated on the basis of least-squares means and the mean square error obtained from the ANOVA.

Population PK-PD (Exposure-Response) Modeling and Simulation
The population PK and PK-PD analyses were conducted using first-order conditional estimation method with interaction in Nonlinear Mixed Effects Modeling software (NONMEM Version 7.4, ICON).Perl-speaks-NONMEM (Uppsala University) was used for model diagnostics.Data management, simulations, computation of summary statistics, and graphical analyses were performed using R Version 3.3 or higher.
The analytic strategy and initial nedosiran population PK and PK-PD model have been described previously using adolescent and adult PK data collected from PHYOX1. 17In the current study, the population PK and PK-PD model was improved from the prior model.Briefly, the plasma PK of nedosiran was described by a 2-compartment disposition model, with a central and a peripheral distribution compartment.Nedosiran was eliminated through 2 pathways, one based on linear clearance and represented by apparent clearance (CL/F), and the other based on nonlinear clearance, represented by a Michaelis-Menten pathway described by maximum metabolic rate (V max ) and Michaelis constant (K m ) parameters.The linear clearance of nedosiran is mostly due to renal elimination of nedosiran, whereas the nonlinear component likely depicts nonlinear liver uptake and disposition for nedosiran, a GalNAc small interfering RNA. 10,23,24he absorption of nedosiran represented by an initial sharp increase in concentration followed by a slow and delayed increase in concentration after subcutaneous administration was described by a dual transitcompartment absorption model with a fast and a slow absorption component.Each of the fast and slow absorption pathways was described by transit compartments for which the optimal number of transit compartments required for each pathway were estimated (Figure S2).Weight-based allometric scaling components were included in the volume and clearance parameters to account for size differences and facilitate extrapolating into the pediatric population.The allometric scaling exponent was fixed to a value of 0.75 for clearance and intercompartmental clearance, whereas it was fixed to a value of 1 for the volume parameters.
The prior nedosiran population PK model was updated with additional PK data from PHYOX1, PHYOX2, PHYOX3, as well as PK data from PHYOX5 (the current analysis), and PHYOX6.
A covariate analysis on nedosiran population PK was performed using a stepwise forward addition and backward elimination process, which focused on age, body weight, body surface area (BSA), BMI, eGFR adjusted for BSA, creatinine clearance, alanine aminotransferase, aspartate aminotransferase, bilirubin, albumin, participant sex, primary hyperoxaluria type, hepatic function, vitamin B 6 use, and race.The influence of these covariates was tested for their statistical significance on PK parameters (CL/F, apparent volume of distribution for the central compartment [V c /F], and absorption transit rate constant [k a ]) in the model.Covariate selection in the model was based on decrease in objective function value (OBJ) at least 6.64 (P < .01)when adding a covariate and an increase in OBJ more than 10.8 (P < .001)when eliminating a covariate.
Previously, the 24-hour Uox excretion data were best described by an indirect response model in which the nedosiran concentration inhibited the production of Uox.This prior nedosiran population PK-PD model was then updated with more longitudinal 24-hour Uox excretion data in PH participants from PHYOX1, PHYOX2, and PHYOX3.A sequential PK and PD modeling approach was used for the PK-PD analysis, with an indirect response model as the best PD model to fit the 24-hour Uox excretion data.The PK of nedosiran was modeled first, as previously described.The PK-PD model used predicted individual plasma concentrations from the PK model to correlate the effect on 24-hour Uox excretion, with the drug effect of nedosiran modeled as a maximum inhibitory effect model inhibiting the production of 24-hour Uox excretion (Figure S2).For PK-PD modeling, individual PK parameters from PH1 participants were fixed to individual empirical Bayes estimates from the final PK model.
The bidirectional stepwise procedure used to identify covariates in the population PK model was also used in the nedosiran population PK-PD model.Covariates of interest were eGFR, body weight, body surface area, BMI, age, 24-hour Uox, 24-hour Uox/creatinine ratio, participant sex, and vitamin B 6 use.Individual estimated random effects for the maximum inhibitory effect, half maximal inhibitory concentration, and 24hour Uox excretion were plotted against the covariates, which were included in the model if the η shrinkage for the parameter of interest was less than 35%, a visually apparent relationship was identified, and OBJ was 6.64 or greater (P < .01 with 1 degree of freedom).
Once a final covariate PK-PD model was obtained, model evaluation was performed using graphical techniques, assessing random effects, evaluating collinearity and model stability, parameter uncertainty, and visual predictive checks (VPCs).VPCs were performed by simulating 1000 replicates to evaluate the predictive ability of the final model.
Finally, after the final nedosiran PK-PD model was selected, multiple-dose simulations were performed using that model to choose an appropriate nedosiran dosing regimen that could optimize reductions in 24-hour Uox excretion for adult and adolescent (aged 12 years or older) patients with PH1 with CKD Stages 4/5.
A nedosiran sodium dose range between 68 and 170 mg (in steps of 8.5 mg) was simulated to predict plasma nedosiran concentrations and 24-hour Uox excretion response following multiple-dose administration (ie, once monthly [Q1M]) of nedosiran in patients with PH.For each regimen, the proportion of participants with Uox excretions of less than 0.60 mmol/24 hour (near-normal range) and less than 0.46 mmol/24 hour (normal range) at 3, 6, 9, and 12 months was plotted.Additionally, the simulations were stratified by weight tier (either 50 kg or greater or less than 50 kg).
A data set of 1000 virtual adults/adolescents (aged 12 years or older; with weight of 50 kg or greater) and 1000 virtual adults/adolescents (aged 12 years or older with weight less than 50 kg) was sampled without replacement and uniformly across sex from the National Health and Nutrition Examination Survey (1999−2012) 25 to simulate nedosiran exposure and 24hour Uox excretion response following multiple doses of nedosiran.Based on the NHANES database, 77.6% of adolescents aged 12−18 years have a body weight greater than 50 kg. 25The virtual data sets were replicated for each of the following categories of CKD, with eGFR randomly generated for each virtual subject from a uniform distribution within each group.Based on the prior population PK-PD analysis for nedosiran, age was not found to be a significant covariate for nedosiran PK or PD. 17 Thus, the following representative nedosiran sodium dosing regimens were finally evaluated in the simulations: •

Study Population
Of

Pharmacokinetics
Following single-dose subcutaneous administration of nedosiran, mean plasma nedosiran concentrations showed a fast and slow absorption profile, with a small peak approximately between 0.5 and 1 hour after dosing and C max occurring around 10 hour after dosing for all groups (Figure 1). Figure 1a in a linear scale and 1b in a semilog scale show that mean plasma nedosiran concentrations in Groups 1a, 1b, and 2 (CKD Stages 4/5) were higher than that in Group 3 (normal kidney function).Figure 1c in a linear scale and 1d in a semilog scale show similar concentrations of nedosiran in plasma samples from the line in (presumed arterial blood) and plasma samples from the line out (presumed venous blood) in Group 1a during the 4hour hemodialysis period that started 8 hours after dosing.
Systemic nedosiran PK exposure metrics (ie, C max and AUC 0-t ) were approximately 2-fold higher in Groups 1a, 1b, and 2 than Group 3 (Table 2).Values for C max and AUC 0-t were generally comparable across Groups 1a, 1b, and 2. Mean CL/F appeared higher in Group 3 than in Groups 1a and 1b.Across the kidney function groups, there were no substantial differences in time to maximum concentration or apparent volume of distribution (Table 2).
Regarding the primary PK end point, ANOVA of geometric mean ratio and 90% CI for C max and AUC 0-t confirmed 2.81-and 2.37-fold higher plasma nedosiran exposure in Group 2 versus Group 3 (Table 3), suggesting nedosiran dose reduction in patients with CKD Stages 4/5.Similar magnitudes of increases in C max and AUC 0-t in Groups 1a and 1b versus Group 3 were also observed.Regarding the secondary PK end point, the 90% CIs of the geometric mean ratios for C max and  (c, d) Linear and semilog scale plots, respectively, of nedosiran concentrations in arterial blood (line in) and plasma (line out) in Group 1a, with emphasis on the 8-12 hours postdose period (the shaded area) when HD was undertaken.Group 1a predialysis = severe CKD (Stages 4/5) dosing 8 hours before HD; Group 1b postdialysis = severe CKD (Stages 4/5) dosing 2 hours after HD; Group 2 = severe CKD (Stages 4/5) no dialysis; Group 3 = normal kidney function.CKD, chronic kidney disease; HD, hemodialysis.
AUC 0-t in participants in Group 1a who received nedosiran 8 hours before hemodialysis relative to participants in Group 1b who received nedosiran 2 hours after hemodialysis included 100%.
Urine samples were collected routinely from participants in Groups 2 and 3 (n = 7 in each group), whereas only 2 participants in each hemodialysis group provided urine samples (1 participant each in Groups 1a and 1b).After nedosiran administration, the cumulative amount excreted over a 72-hour period was approximately 26 mg in Group 3 and less than 10 mg in Groups 1a, 1b, and 2. Mean fe (4.3% vs 15.2%) and renal clearance from the time of dosing to the last quantifiable plasma concentration (0.42 vs 3.15 L/h) were reduced in Group 2 relative to Group 3 (Table S1).
Nedosiran concentrations in all dialysate samples were below the LLOQ (1 ng/mL).Therefore, CLHD was estimated using multiplication of QBIN, R, and f (dialysis) to describe the effect of dialysis on nedosiran PK.The fraction of nedosiran eliminated during the 4hour hemodialysis period on Day 1 in Group 1a was minor, with an estimated mean value of 3.55%.CLHD was estimated at 6.83 mL/min (%CV 172).

Exposure-Response Modeling and Simulation for Dose Selection
The population PK model included the plasma nedosiran concentrations and covariate data from the following studies: The structure of the population PK model is presented in Figure S2, with the addition of betweensubject variability on V c /F, absorption transit rate constant for the first pathway (k a1 ), absorption transit rate constant for the second pathway (k a2 ), and max-imum metabolic rate for the population PK model.Body weight was found to be a significant covariate on k a1 and k a2 , as participants with a higher body weight were shown to have reduced the rate of absorption, which is physiologically plausible given that nedosiran is administered subcutaneously.PH1 was found to be a significant covariate on k a1 solely based on the statistical evaluation of the PK model.Baseline eGFR and body weight were the 2 covariates that affected exposure to nedosiran.eGFR was included as a covariate on CL/F and V c /F. Allometric scaling was included on CL and volume terms.BSA, BMI, alanine aminotransferase, aspartate aminotransferase, bilirubin, albumin, sex, race, hepatic function, and concomitant vitamin B 6 use were not significant covariates on nedosiran PK.
Parameter estimates for the final population PK model are provided in Table 4.The estimated parameter values were 147 L for V c /F and 5.69 L/h for CL/F.Relative to a simulated patient with PH1 weighing 70 kg with normal renal function and receiving 170 mg nedosiran Q1M, the model predicted that steady-state nedosiran exposure (ie, AUC from time of administration over the dosing interval) would increase by 47%, 59%, and 97%, respectively, in a lowweight individual (40 kg), and in individuals with CKD Goodness-of-fit plots for the updated population PK model indicated a good fit to the observed data (Figure S3).The final population PK model adequately described the plasma concentration of nedosiran in PH participants with normal kidney function or relatively preserved kidney function (CKD Stages 2/3), healthy volunteers and non-PH participants with CKD Stages 4/5 with or without dialysis, with an even distribution of observations above and below the 5th and 95th prediction intervals in prediction-corrected VPC (Figure S4).The simulated median values agreed with the observed data.
Based on the simulation from the final population PK model, the total exposure of nedosiran in terms of AUC 0-τ,steady state following 170 mg nedosiran Q1M dose in patients with PH (aged 12 years or older) with eGFR of 30 mL/min/1.73m 2 or greater may be similar to those in patients with PH with eGFR less than 30 mL/min/1.73m 2 dosed with 110.5 mg nedosiran Q1M when body weight is more than 50 kg, or 85 mg nedosiran Q1M when body weight is less than 50 kg (Figure S5).An indirect response model was the best PK-PD model to fit the 24-hour Uox excretion data for adult and adolescent patients with PH1 (Figure S2).No significant covariate was identified in the final population PD model, including age.This enabled inclusion of adolescents with adults in the final population PK-PD analysis.The estimated parameter values were 64.5% for maximum inhibitory effect, 1.04 ng/mL for half maximal inhibitory concentration, and 0.338/week for the first-order elimination rate of 24-hour Uox excretion (Table 4).
Goodness-of-fit plots for the updated population PK-PD model indicated agreement between the observed data and the model predictions (Figure S6).To further validate the PK-PD model graphically, we generated prediction-corrected VPC of the prediction and variability of change in 24-hour Uox excretion reduction with nedosiran treatment time (Figure S7).The population PK-PD model adequately described the observed 24-hour Uox excretion data with an even distribution of observations above and below the 10th and 90th prediction intervals.Simulated median 24-hour Uox excretion values agreed with the observed data, indicating that the predictive performance of the model was acceptable.
Figure 2 shows comparable decreases in 24hour Uox excretion associated with the nedosiran Q1M dose regimens in these virtual populations.In adult/adolescent patients with PH1 weighing less than 50 kg with normal kidney function or CKD Stages 2/3, the decrease in 24-hour Uox excretion associated with nedosiran 136 mg Q1M dose (128 mg free acid equivalent) is likely comparable to that in adult/adolescent subjects with normal kidney function or CKD Stages 2/3 and weight of 50 kg or greater administered nedosiran 170 mg Q1M (Figure 2A).For adults/adolescents weighing less than 50 kg with CKD Stages 4/5, a further dose reduction to 85 mg Q1M (80 mg free acid equivalent) is expected to achieve a similar decrease in 24-hour Uox excretion as that in adults/adolescents weighing less than 50 kg with normal kidney function or CKD Stages 2/3 administered nedosiran 136 mg Q1M (Figure 2A).The simulations demonstrated that in adult/adolescent subjects (aged 12 years or older) weighing 50 kg or greater with CKD Stages 4/5, a nedosiran dose reduction to 110.5 mg Q1M (104 mg free acid equivalent) could result in a similar proportion attaining normal and near-normal 24-hour Uox excretion as adult subjects with normal kidney function or CKD Stages 2/3 receiving nedosiran 170 mg Q1M (160 mg free acid equivalent) (Figure 2B). Figure 3 summarizes the percentages of simulated subjects attaining normal and near-normal 24-hour Uox excretion with the representative monthly dosing strategies at the end of Weeks 13, 26, 39, 52, and 128.

Safety
Treatment-emergent AEs were reported by at least 1 participant in each group (Table S2).The 4 AEs considered related to nedosiran were the 2 reports of ISR (1 mild and 1 moderate in severity), 1 report of mild administration site swelling, and 1 instance of pancreatitis (in Group 1b).No clinically significant changes in cytokine levels were reported for the 2 participants experiencing ISRs.The case of pancreatitis occurred in a 44-year-old Black woman with long-standing kidney failure.The episode started with nausea, vomiting, and abdominal pain with associated worsening hypertension due to an inability to ingest oral antihypertensive medication.She was admitted to the local hospital on Day 4 and stayed for 1 week.Lipase was elevated and rising, and mild edema of the pancreas was noted on x-ray.The only factor relating pancreatitis to nedosiran was a temporal relationship with the time of administration: The episode started 3 days after receiving nedosiran (and resolved 10 days later), which is far earlier than the onset of nedosiran's PD effect (measured in weeks).In the entire study cohort, there were no clinically significant changes in clinical laboratory test parameters, physical exam findings, vital sign measurements, or 12-lead ECG values during the study.

Discussion
The heterogeneous nature of PH complicates recognition of this disease in clinical practice, 2 with some patients remaining undiagnosed until the development of CKD Stage 4 or 5. 5,26 Although recent progress in the development of new treatments for PH will encourage early detection, many patients with PH will have advanced CKD when diagnosed.It is, therefore, imperative to characterize the PK profile and select an optimal dose of any candidate treatment for PH in this vulnerable patient population.
The current open-label Phase 1 study compared nedosiran PK in non-PH participants with CKD Stages 4 and 5 (eGFR less than 30 mL/min/1.73m 2 ) with or without dialysis, to healthy volunteers with normal kidney function (eGFR of 90 mL/min/1.73m 2 or greater).Participants with CKD Stages 4/5 showed a 2.37-fold higher nedosiran AUC 0-t versus those with normal kidney function.Because CKD Stages 4/5 had a significant effect on the PK of nedosiran, a dose adjustment is recommended for patients with PH and an eGFR less than 30 mL/min/1.73m 2 .Specifically, the population PK-PD (exposure-response) modeling and simulation revealed that nedosiran dose adjustment is not required in patients with PH1 with CKD Stage 2 or 3. A nedosiran dose reduction from 170 mg Q1M to 110.5 mg Q1M is required in adult/adolescent patients with PH1 aged 12 years or older with CKD Stages 4/5 (less than mL/min/1.73m 2 ) and weighing 50 kg or greater and from 136 mg Q1M to 85 mg Q1M is required in adults/adolescents weighing less than 50 kg with CKD Stages 4/5 to achieve and maintain normal or near-normal 24-hour Uox excretion similar to those patients with normal renal function or CKD Stages 2/3 administered with 170 mg nedosiran Q1M.The recommended dosage regimens apply to adolescents and adults considering that population PK-PD modeling included adolescent data and age was not a statistically significant covariate in either population PK or PK-PD models.
Effect of dialysis on the exposure of a drug is usually evaluated using a crossover design, in which the same set of individuals receive treatment before  and after dialysis and the 2 periods are separated by a washout period dependent on the half-life of the drug.A crossover design was not feasible in this study because of the need for an exceedingly long washout period (on the order of months).Since nedosiran is primarily taken up by the liver and has a long liver half-life (based on preclinical studies), 17 we conducted a parallel group design, with the potential for confounding mitigated by balancing Groups 1a and 1b on the basis of age, weight, and sex.In this study, the timing of dialysis relative to nedosiran administration had no clinically significant impact on the PK of nedosiran.It is noteworthy that the plasma nedosiran concentration-time profiles did not diverge during the 8-to 12-hour postdose dialysis period and that the concentration of nedosiran in plasma from arterial blood (line in) and plasma from venous blood (line out) during this period was almost overlapping.Furthermore, all dialysate samples were below the LLOQ for nedosiran, and during the 4-hour dialysis period, only 4% of nedosiran was eliminated, indicating that dialysis does not accelerate drug clearance from the body.As no significant quantities are removed by dialysis, dialysis-based dose adjustments are not recommended for nedosiran.
Nedosiran was generally safe and well tolerated in participants with relatively preserved kidney function and CKD Stages 4/5, including those on dialysis.However, our study was not designed to detect safety differences among the 4 treatment groups.The case of pancreatitis deemed possibly related to nedosiran due to a temporal association between the event and time of intervention seems to be an anomaly and is inconsistent with the known safety profile of nedosiran and its mechanism and onset of action. 17

Conclusions
Following a single dose of subcutaneous nedosiran sodium 170 mg (160 mg free acid equivalent), participants with CKD Stages 4/5 exhibited an approximately 2-fold higher nedosiran AUC 0-t than that in healthy volunteers with normal kidney function.Based on population PK-PD modeling and simulation, an adjustment to the nedosiran dosage is recommended in patients with PH1 aged 12 years or older who have CKD Stages 4/5 (less than 30 mL/min/1.73m 2 ), while a dosage adjustment in patients with PH1 with eGFR between 30 and 89 mL/min/1.73m 2 (CKD Stages 2/3) is not required.Since dialysis had no discernible effect on nedosiran PK profile, nedosiran dose adjustments are not recommended solely on the basis of dialysis.Future clinical studies will establish the safety and efficacy of the recommended nedosiran dosage in patients with PH with CKD Stages 4/5.

Table 1 .
Baseline Demographic and Clinical Characteristics of the PHYOX5 Participants