Skip to main content
Molecular Therapy. Nucleic Acids logoLink to Molecular Therapy. Nucleic Acids
. 2026 May 6;37(2):102948. doi: 10.1016/j.omtn.2026.102948

Biodistribution of mRNA vaccines in rats: Enrichment in injection site and lymph tissues and rapid clearance without tissue persistence

Susan MG Goody 1,∗, Christopher Rowbottom 1, Yun Liu 1, Nancy Chen 1
PMCID: PMC13223946  PMID: 42233109

Abstract

Messenger RNA (mRNA) vaccines using lipid nanoparticles (LNPs) are globally approved with acceptable safety profiles for preventing respiratory diseases. Here, we describe tissue distribution and kinetics of mRNA-LNP vaccines after intramuscular dosing using three products formulated with same LNP matrix: mRNA-1273 (Spikevax), mRNA-1647 (a candidate cytomegalovirus [CMV] vaccine), and a reporter mRNA (nascent peptide-luciferase) drug product. Consistent biodistribution patterns were observed: tissues with highest exposures were the injection site, draining lymph nodes, and spleen, with minimal distribution to other organs. Levels of vaccine components declined rapidly in circulation and tissues, with model-based projections indicating concentrations at low or below quantifiable levels by ∼2 weeks. Unbound (free) spike protein was transient after mRNA-1273 vaccination (modeled decline < 5 days) and did not accumulate with repeated dosing. The ionizable lipid in the LNP, lipid H, underwent biotransformation and was excreted renally and hepatically, with no human-specific metabolites. Collectively, these results indicate similar biodistribution patterns across products within this dataset using a shared LNP composition, despite different mRNA cargo. Furthermore, in a SARS-CoV-2 infection-free model, no sustained spike protein was observed. Overall, the data establish a framework that can be used to scientifically justify leveraging data across a shared platform.

Keywords: MT: Clinical Applications, biodistribution, lipid nanoparticle, mRNA, spike protein, persistence, transient, clearance kinetics, intramuscular administration

Graphical abstract

graphic file with name ga1.jpg


Goody and colleagues show that intramuscular mRNA-lipid nanoparticle medicines exhibit conserved, lymphoid-focused biodistribution with rapid clearance of mRNA, lipid components, and expressed protein. These data establish LNP composition as the primary driver of tissue exposure, providing a scientific foundation for platform-level biodistribution bridging and reduced redundant animal testing.

Introduction

Lipid nanoparticle (LNP)-encapsulated mRNA vaccines are designed to deliver mRNA efficiently to cells adjacent to the intramuscular (IM) injection site and induce a robust immune response.1 LNPs typically comprise four components: an ionizable lipid, a PEGylated lipid, cholesterol, and a helper lipid such as distearoylphosphatidylcholine (DSPC).2,3 Following IM injection, histological and flow cytometry data from rats and/or non-human primates show that mRNA-LNP vaccines are taken up predominantly by immune cells, such as antigen-presenting cell (APC)-associated compartments including macrophages, dendritic cells, and fibroblasts at the injection site and draining lymph nodes.4 APCs expressing the vaccine antigen, along with mRNA-LNP, traffic to afferent lymphatics and initiate seeding of germinal centers in draining lymph nodes, which drive affinity maturation of antigen-specific B cell receptors and produce corresponding long-lived plasma and memory B cell antibodies.3,5

mRNA-LNP products are recognized worldwide for delivering clinically meaningful outcomes with well-characterized and acceptable safety profiles. Global licensure of four mRNA-LNP vaccines, including Moderna’s mRNA-1273 (Spikevax), mRNA-1345 (mRESVIA), mRNA-1283 (mNEXSPIKE), and Pfizer-BioNTech’s BNT162b2 (Comirnaty), are a testament to this.6,7,8 Beyond these licensed products, there are several examples of promising clinical data with other indications such as seasonal influenza9,10 and oncology.11,12 Collectively, these successes solidify mRNA-LNP vaccines and therapies as a versatile, regulatory-validated modality with a growing clinical footprint spanning licensed respiratory vaccines, late-stage prevention therapeutics, and personalized oncology medicines.

Drug developers follow internationally recognized guidelines for preclinical evaluations of mRNA-LNP vaccines,13,14 which provide recommendations on the studies designed to support their regulatory approval. These studies rigorously assess the preclinical immunogenicity, safety, metabolism, clearance, and biodistribution of the vaccines to support testing in humans. Biodistribution studies in rats are an accepted regulatory standard for mRNA-LNP vaccines15,16,17,18,19 and are designed to map the tissue localization of drug components and/or expressed antigen, measure their persistence and clearance, and to assess for potential accumulation with repeated administration if more than one dosing occasion is employed. For vaccines, recommendations for biodistribution studies are risk-based and product-specific rather than prescriptive, noting that bridging across products is acceptable when scientifically justified if LNP composition and route of administration remain unchanged.13,14 Detailed biodistribution datasets have been reviewed and deemed acceptable to support global licensure of mRNA-LNP vaccines; however, because these datasets are proprietary in regulatory submission dossiers, the broader scientific community has limited visibility to the details and the results from these studies.

To increase transparency of these data, we present a comprehensive description of the tissue distribution and systemic kinetics of three mRNA-LNP products formulated in the same LNP matrix following IM dosing in Sprague-Dawley rats. We also provide the first characterization of SARS-CoV-2 spike protein tissue biodistribution and clearance in a SARS-CoV-2 infection-naïve model, addressing a key limitation of human studies where pre-existing or subclinical SARS-CoV-2 infection cannot be reliably ruled out. Additionally, we describe the metabolism and excretion of the ionizable lipid used in the LNP formulations, lipid H. Within this dataset, consistent findings across products with differing mRNA cargo and immune responses suggest that LNP composition is the primary determinant of biodistribution. Moreover, LNP components and spike protein levels were transient, with no evidence of quantifiable persistence beyond approximately 1–2 weeks in the highest-exposure tissues after IM injection.

Results

Across drug products, tissues with the highest exposure of mRNA and lipid H are consistently the injection site, draining lymph nodes, and spleen

Across the three biodistribution studies, administered rat dose levels were within a similar range on a flat-dose basis and represented conservative high-margin exposures relative to human doses (when normalized to body weight). Study-specific dose levels and corresponding rat: human margins are summarized in Table S1. An evaluation of the area under the time-concentration curve (AUC; h∗ng/mL or h∗ng/g for plasma/serum or tissue, respectively) for mRNA and lipid H across drug products revealed that the tissues with the highest AUC were consistently the injection site, draining lymph nodes (particularly axillary, inguinal, and/or popliteal), and spleen (Tables 1 and 2). Differences in dose levels and/or final collection time points have the potential to introduce variability in the value of raw calculated AUC; therefore, tissue exposures were normalized to systemic exposure (based on AUC). The resulting values (otherwise referred to as the “tissue-to-systemic ratio, expressed as a percentage [%]) were also used to rank order tissue exposure relative to systemic exposure. Tissues with ratios >100% represent those with higher exposure relative to systemic exposure, whereas tissues with ratios <100% represent tissues with lower exposure compared with systemic exposure. The resulting distribution patterns of tissue to systemic ratios were consistent with the calculated AUC value (Tables 1 and 2) across the drug products, as demonstrated by ratios of >100% for lymphoid tissues and <100% for other distal tissues.

Table 1.

Summary of systemic and tissue mRNA AUC and tissue to systemic AUC ratios, sex combined

Matrix AUC (h∗ng/mL or h∗ng/g)f
Tissue AUC to systemic AUC ratio (%)b
mRNA from mRNA-1647a
mRNA from NPI-Luc mRNA
mRNA from mRNA-1273
mRNA from mRNA-1647
mRNA from NPI-Luc mRNA
mRNA from mRNA-1273
Day 1
Day 1
Day 1
Day 28
Day 1
Day 1
Day 1
Day 28
Male Combined Combined Combined Male Combined Combined Combined
Injection site 24,483 31,100 29,285 10,885 101,170 2,907 2,074 1,293
Lymph node (axillary) ND 8,260 13,185 14,860 ND 772 934 1,765
Lymph node (distal) 1,521 ND ND ND 6,285 ND ND ND
Lymph node (inguinal) ND NC ND ND ND NC ND ND
Lymph node (inguinal/popliteal) ND ND 26,500 18,000 ND ND 1,877 2,138
Lymph node (popliteal) ND NC ND ND ND NC ND ND
Lymph node (proximal) 4865 ND ND ND 20,103 ND ND ND
Spleen 315 51,000 71,250 68,300 1,302 4,800 5,046 8,112
Systemicb 24.2 1,070 1,412 842 NA NA NA NA
Bone marrow 2.70 14.0 ND ND 11.2 1.31 ND ND
Brain 0.41c NC 1.8d NC 1.70c NC 0.13d NC
Heart 3.52 NC 108 144 14.5 NC 7.65 17.1
Liver 11.9 438 908 599 49.2 40.9 64.3 71.1
Lung 3.97 75.4d 289 134 16.4 7.05d 20.5 15.9
Testes (males)e 4.71 NC ND ND 14.5 NC ND ND

AUC, area under the concentration versus time curve from the start of dose administration to the time after dosing at which the last quantifiable concentration was observed; LNP, lipid nanoparticle; Luc, luciferase; mRNA, messenger RNA; ratio is expressed in %; values less than 100% represent tissue exposure less than systemic; NA, not applicable as calculation was not performed because the parameter does not apply to this matrix; NC, not calculable due to insufficient quantifiable data to derive pharmacokinetic parameters (e.g., AUC, effective T1/2); ND, not determined as part of the pre-defined study design; NPI, nascent peptide imaging.

a

Average reported across six mRNA sequences.

b

Tissue AUC to plasma/serum AUC ratio (%) for mRNA-1647 used plasma, whereas NPI-Luciferase mRNA in lipid H LNPs and mRNA-1273 used serum.

c

Only three out of the six mRNA sequences had quantifiable values.

d

Quantifiable values only observed for females. Concentrations in males were all below LLOQ.

e

Insufficient ovary tissues from NPI-Luc study to evaluate mRNA.

f

mRNA values across products were generated using different qualified assay platforms (RT-qPCR for mRNA-1273; bDNA-based assays for NPI-Luc and mRNA-1647) and therefore support qualitative cross-product comparison of tissue distribution patterns more strongly than exact numerical equivalence of concentration metrics.

Table 2.

Summary of systemic and tissue lipid H AUC and AUC to systemic AUC ratios, sex combined

Matrixa AUC (h∗ng/mL or h∗ng/g)
Tissue AUC to systemic AUC ratio (%)
NPI-Luc mRNA
mRNA-1273
NPI-Luc mRNA
mRNA-1273
Day 1
Day 1
Day 28
Day 1
Day 1
Day 28
Combined Combined Combined Combined Combined Combined
Injection site 15,2000 168,750 306,000 2,165 1,023 2,045
Lymph node (axillary) 13,700 54,950 89,100 195 333 595
Lymph node (inguinal) 83,100 ND ND 1,184 ND ND
Lymph node (inguinal/popliteal) ND 126,500 90,350 ND 767 604
Lymph node (popliteal) 71,200 ND ND 1,014 ND ND
Spleen 6,980 65,450 35,950 99.4 397 240
Systemic 7,020 16,495 14,965 NA NA NA
Bone marrow 591 ND ND 8.42 ND ND
Brain NC NC NC NC NC NC
Heart NC 21.7 28.2 NC 0.132 0.188
Kidney (left) 8.09 ND ND 0.115 ND ND
Kidney (right) 14.6 ND ND 0.208 ND ND
Liver 2,180 1,865 3,990 31.1 11.3 26.7
Lung 231 366 238 3.29 2.22 1.59
Ovaries (females) 457 ND ND 6.51 ND ND

AUC, area under the concentration versus time curve from the start of dose administration to the time after dosing at which the last quantifiable concentration was observed; LNP, lipid nanoparticle; Luc, luciferase; mRNA, messenger RNA; ratio is expressed in %; values less than 100% represent tissue exposure less than systemic; NA, not applicable as calculation was not performed because the parameter does not apply to this matrix; NC, not calculable due to insufficient quantifiable data to obtain pharmacokinetic parameters (e.g., AUC, effective T1/2 not calculable [insufficient data points above the limit of quantification]); ND, not determined as part of the pre-defined study design; NPI, nascent peptide imaging.

a

Concentrations in testis were all below LLOQ in NPI-Luc study and were not determined in mRNA-1273 study.

Of the distal tissues, the liver had the highest tissue-to-systemic ratios for mRNA and lipid H. When compared to the injection site, which serves as the primary depot of exposure after IM administration (Tables 1 and 2), liver exposure was significantly lower, representing only ∼3% relative to the injection site. Other distal tissues, such as bone marrow, brain, heart, lung, testes, or ovaries, had exposures that were <1% relative to the injection site (Tables 1 and 2). The concentrations in these tissues were highly variable and should be interpreted cautiously, as values were often near the lower limit of assay quantification (LLOQ; Table 8; Tables 1 and 2; Table S6) or reflected measurable mRNA in the absence of detectable lipid. For example, the detectable concentrations in brain were sporadic and inconsistent across sexes (e.g., only in females and at a few early time points) and mRNA concentrations were observed near the lower end of the assay limit, and sometimes without quantifiable levels of lipid H (Tables 1 and 2; Table S6). Summary data for selecting low-level distal tissues (e.g., brain and heart) are presented as examples in Table S6. Collectively, these findings demonstrate that exposures in the brain and other distal tissues are negligible and show limited biodistribution of mRNA-LNP components beyond the injection site and lymphoid tissues.

No evidence of persistence of mRNA, lipid H, or SARS-CoV-2 spike protein in the systemic circulation or tissues beyond the reported observation and modeling time frames

In the systemic circulation, mRNA and lipid H declined rapidly, with a serum half-life (T1/2) of approximately 3 and 8 h, respectively (Table 3). The T1/2 values of mRNA in lymphoid tissues with highest exposure (based on AUC) ranged from 24 to 48 h in lymph nodes and were ∼63 h in the spleen (Table 3). Spike protein concentration peaked at 24 h (the time after dosing at which the maximum observable pharmacodynamic effect occurred [TEmax]) following mRNA-1273 administration and declined rapidly with an effective serum T1/2 of 18 h (Table 4). The quantifiable levels in serum were observed only up to 3 days after the first dose and up to 7 days after the second (Table 4). While there were enough time points with quantifiable spike protein levels in the serum after the first dose to calculate the area under the pharmacodynamic (PD) effect (encoded protein)-time curve from time 0 to the last quantifiable concentration post dose (AUEClast), there were not enough time points with quantifiable levels after the second dose to calculate the AUEClast. Following the same distribution pattern as mRNA and lipid H, unbound (free) spike protein was detected mainly in the injection site, spleen and lymph nodes. Liver was the only distal tissue with measurable unbound (free) spike protein; however, values were near the LLOQ and were observed only up to 24 h. Unbound (free) spike protein was not detected in brain or heart or lung (Table 4).

Table 3.

Summary of systemic and tissue mRNA or lipid H effective T1/2, sex combined

Matrix Effective T1/2 (h) mRNA
Effective T1/2 (h) lipid H
mRNA from mRNA-1647a
mRNA from NPI-Luc mRNA
mRNA from mRNA-1273
NPI-Luc mRNA
mRNA-1273
Day 1
Day 1
Day 1
Day 28
Day 1
Day 1
Day 28
Male Combined Combined Combined Combined Combined Combined
Injection site 13.8–17.1a 4.88 15.2 10.1 14.4 15.9 15.6
Lymph node (axillary) ND 24.1 48.0 37.5 60.5 23.0 27.2
Lymph node (distal) 27.9–36.2a ND ND ND ND ND ND
Lymph node (inguinal) ND NC ND ND 37.5 ND ND
Lymph node (inguinal/popliteal) ND ND 42.8 46.4 ND 35.4 23.4
Lymph node (popliteal) ND NC ND ND 18.9 ND ND
Lymph node (proximal) 32.2–38.2a NC ND ND NC ND ND
Spleen 46.2–83.0a 45.8 49.9 57.9 22.8 27.2 27.6
Systemicb NC 2.95 2.39 2.83 8.44 5.93 5.61
Bone marrow NC 7.59 ND ND 7.94 ND ND
Brain NC NC 1.68c NC NC NC NC
Heart NC NC 2.62 2.50 NC 1.10 1.23
Kidney (left, right) NC NC ND ND 1.73 (left)
1.85 (right)
ND ND
Liver NC 2.98 6.97 6.55 6.33 6.10 7.32
Lung NC 3.07 3.28 7.62 4.80 5.14 4.85
Ovary (female) ND ND ND ND 27.8 ND ND
Testes (male) NC NC ND ND NC ND ND

Effective T1/2, half-life determination using mean residence time; LNP, lipid nanoparticle; Luc, luciferase; mRNA, messenger mRNA; NA, not applicable as calculation was not performed because the parameter does not apply to this matrix. NC, not calculable due to insufficient quantifiable data points to calculate pharmacokinetic parameters (e.g., AUC, effective T1/2); ND, not determined as part of the pre-defined study design.

In addition to the tissues included in the table, eye, jejunum, pancreas, stomach, thymus, and uterus were also evaluated but concentrations were insufficient to calculate half-life.

a

Range reported across six mRNA sequences.

b

Systemic mRNA exposure was evaluated in serum, whereas lipid H exposure was evaluated in plasma.

c

Insufficient concentrations in males to calculate half-life.

Table 4.

Summary of unbound/free SARS-CoV-2 spike protein quantification and calculated pharmacodynamics parameters following one or two administrations of mRNA-1273

Matrix Detection (TEmax/TElast)
(day 1 vs. day 28)
Quantification (Emax/AUEClast) Effective T1/2 (if calculable) Unbound (free) spike protein
Serum day 1: TEmax at 24 h, TElast at 72 h
day 28: TEmax at 168 h, TElast at 168 h
– day 1 day 28 ∼18 h (day 1 only) no sustained quantifiable levels beyond 72 h on day 1 and 168 h on day 2
Emax (ng/mL) 5.38 1.80
AUEClast (h∗ng/mL) 190 NC
Liver day 1: TEmax at 10 h, TElast at 24 h
day 28: TEmax at 24 h, TElast at 24 h
– day 1 day 28 ∼4–5 h (day 1 only) no sustained quantifiable levels beyond 24 h on day 1 and day 28
Emax (ng/g) 54 15
AUEClast (h∗ng/g) 789 NC
Brain not detected at either time point NC NC not detected under assay conditions used
Heart not detected at either time point NC NC not detected under assay conditions used
Lung day 1: not detected
day 28: TEmax at 10 h, TElast at 10 h
– day 1 day 28 NC no sustained quantifiable levels beyond 10 h on day 28
Emax (ng/g) NC 23
AUEClast (h∗ng/g) NC NC
Lymph nodes pooled day 1: TEmax at 14 h, TElast at 17 h
day 28: not detected
– day 1 day 28 NC no sustained quantifiable levels beyond 17 h on day 1
Emax (ng/g) 341 NC
AUEClast (h∗ng/g) NC NC
Injection site day 1: TEmax at 17 h, TElast at 48 h
day 28: TEmax at 10 h, TElast at 10 h
– day 1 day 28 ∼7 h (day 1 only) no sustained quantifiable levels beyond 48 h on day 1 and 10 h on day 28
Emax (ng/g) 288 29.8
AUEClast (h∗ng/g) 4,075 NC
Spleen day 1: TEmax at 10 h, TElast at 24 h
day 28: TEmax at 10 h, TElast at 10 h
– day 1 day 28 ∼5 h (day 1 only) no sustained quantifiable levels beyond 24 h on day 1 and 10 h on day 28
Emax (ng/g) 77 82
AUEClast (h∗ng/g) 1,350 NC

Emax, maximum observed pharmacodynamic effect (encoded protein); AUEClast, the area under the pharmacodynamic effect (encoded protein)-time curve from time 0 to the last quantifiable concentration post dose; NC, not calculable due to insufficient quantifiable data points to obtain pharmacokinetic parameters (e.g., AUC, effective T1/2); TElast, the time after dosing at which the last observable pharmacodynamic effect (encoded protein) occurred.

While these T1/2 values provide substantive evidence of rapid analyte clearance and lack of sustained persistence, the differences in timing for the final sample collection meant that it was not possible to precisely define the time point at which mRNA, lipid H, and/or encoded spike protein would fall below quantifiable levels in all matrices. Thus, first-order elimination modeling (using observed maximum concentration [Cmax] and serum/tissue T1/2 for mRNA, and TEmax and effective T1/2 for spike protein) was used to simulate the later phase of decline beyond the observed sampling window (Tables 5 and 6; Figures 1A and 1B). To estimate the most extreme circumstances, clearance for tissues with the highest exposure and corresponding T1/2 (spleen and lymph nodes) was simulated using the most sensitive analyte (mRNA). The simulation suggested >98% elimination of mRNA at approximately 2 weeks post-dose (Table 5; Figure 1A). In the case of unbound (free) spike protein in the systemic circulation, the simulation suggested >98% elimination within 5 days post-dose (Table 6; Figure 1B).

Table 5.

Simulated mRNA exposures in tissues with the highest AUC and longest T1/2 demonstrate a marked decline with >98% reduction approximately 2 weeks after administration of mRNA-1273

Half-lives (n) Time (h) Simulated mRNA (ng/g) remaining from Cmax
Simulated % eliminated relative to Cmax
Spleen Lymph nodes axillary
0 0 1,447.5 366.8 0
1 62 723.5 183.4 50
2 124 361.75 91.7 75
3 186 180.9 45.9 87.5
4 248 90.4 22.9 93.8
5 310 45.2 11.5 96.9
6 372 22.6 5.75 98.4

Table 6.

Simulated exposures of unbound/free SARS-CoV-2 spike protein in systemic circulation demonstrate a marked decline with >98% reduction within 5 days after administration of mRNA-1273

Half-lives (n) Time (h) Simulated spike protein (ng/mL) remaining from Emax Simulated % of eliminated relative to Emax
0 0 5.38 0
1 18.1 2.69 50
2 36.2 1.35 75
3 54.3 0.673 87.5
4 72.4 0.336 93.8
5 90.5 0.168 96.9
6 108.7 0.084 98.4

Figure 1.

Figure 1

Simulated concentration-time profiles illustrating modeled decline of mRNA and unbound/free spike protein beyond the observed sampling window

(A) Simulated mRNA remaining from Cmax in axillary lymph nodes and spleen. (B) Simulated unbound (free) spike protein remaining from Emax in plasma. Profiles were simulated using observed Emax/Cmax and estimated effective half-life (T1/2). Vertical dashed lines indicate sequential half-lives. At six half-lives, ∼1.6% of the initial concentration remains (∼98.4% decline from the starting value).

The expressed nascent peptide-luciferase protein (NPI-Luc) from the study with NPI-Luc mRNA in lipid H-containing LNPs was largely not quantifiable in either tissues or systemic exposure (data not shown).

Lack of accumulation of mRNA, lipid H, and spike protein with repeated dosing and no impact of immune response on tissue distribution

To further interrogate the potential for persistence of mRNA and lipid H in the systemic circulation or in tissues, and to assess whether development of an immune response impacted the biodistribution profile, exposures of mRNA and lipid H were evaluated following a second dose of mRNA-1273. In this experiment, a second dose mRNA-1273 was administered in the contralateral quadricep approximately 4 weeks after the first dose. While repeat-dose accumulation at a single injection site and adjacent draining lymph nodes could not be assess due to the contralateral design, the AUC and Cmax of mRNA and lipid H in the systemic circulation and in other tissues with highest exposure (e.g., spleen) were the same on day 1 as on day 28, with no statistically significant differences observed (Figure 2).

Figure 2.

Figure 2

Cmax and AUC of mRNA and lipid H in Sprague-Dawley rats following IM administration of mRNA-1273 on day 1 and day 28

Means Cmax and AUC were calculated based on composite animal data at N = 4 per time point per sex to derive concentration-time profiles. Individual animal Cmax and AUC values were therefore not available for display. (A) Cmax of mRNA-1273. (B) AUC of mRNA-1273. (C) Cmax of lipid H. (D) AUC of lipid H. Tissues analyzed included serum, injection site, axillary lymph nodes, and spleen. Bars represent mean ± SEM. Comparisons between day 1 and day 28 were performed using unpaired t tests; no statistically significant differences were observed (p < 0.05). ns, not significant.

Similarly for unbound (free) spike protein in systemic or tissues, maximum observed PD effect (Emax) values on day 1 and day 28 were either comparable or, in some cases, lower or could not be determined due to insufficient measurable response on day 28 (Table 4). Comparison of AUEClast between day 1 and day 28 was not feasible, as only a few quantifiable levels were available following the second dose to obtain reliable estimation.

To confirm development of an immune response to mRNA-1273 and further reiterate that immunogenicity does not impact the biodistribution profile, antibody responses against the SARS-CoV-2 spike protein S2 subunit (S2P) antigen were measured following the first and second dose of mRNA-1273. At baseline (day 1, pre-dose), antibody titers were at or near the LLOQ (<30 antibody units/mL), with no measurable pre-existing antibodies (Figure S2A). At 2 weeks (336 h) post-first-dose, all animals demonstrated a robust antibody response, with binding titers increasing by several orders of magnitude compared with pre-dose values. On day 28, prior to the second dose, antibody titers remained elevated relative to baseline, demonstrating persistence of the immune response from the primary immunization (Figure S2B). Following the second dose, titers were further boosted, reaching high and consistent levels across all animals by 2 weeks after the second dose (336 h).

Efficient lipid clearance via metabolism and excretion

Analysis of the lipid H metabolic profile confirmed rapid clearance from systemic circulation and excretion through both hepatic and renal pathways. In plasma, parent lipid H accounted for nearly all detectable signals at early time points, with only trace levels of metabolites present (Figure S3A). By contrast, metabolites dominated in urine and bile excreta (Figure S3B and S3C). In urine, multiple metabolites (mainly M1–M4) were observed within the first 24 h post-dose, with negligible amounts of parent lipid H detected, indicating efficient renal elimination through biotransformation (Figure S3B). A fraction of parent lipid H was observed in bile; however, metabolites (M1–M12) were the dominant species, representing the majority of the relative abundance at later intervals (Figure S3C). Only small amounts of parent lipid H were detected in excreta, indicating that clearance occurred primarily through metabolism rather than direct elimination of the parent lipid H.

Metabolite identification in rat plasma, urine, and bile identified 12 distinct metabolites derived from pathways including ester hydrolysis, β-oxidation, hydroxylation, and glutathione conjugation (Table S4). In vitro studies using hepatocytes from human, rat, and non-human primates confirmed that these pathways were conserved across species, with consistent detection of major metabolites M1, M3, M4, M6, and M7; no human-specific metabolites were detected (Table S5).

Discussion

mRNA-LNP vaccines are now licensed worldwide for the prevention of several respiratory diseases, and, as a modality, mRNA-LNP products are actively being pursued for other indications. The acceptable safety and effectiveness profiles of approved products such as Moderna’s mRNA-1273 (Spikevax) and Pfizer-BioNTech’s BNT162b2 (Comirnaty), have established mRNA-LNP vaccines as a reliable and adaptable platform, preventing hospitalizations and deaths due to COVID-19.20,21,22,23 Using rigorous methods that have been accepted in regulatory filings and adhere to international regulatory guidelines, we report a comprehensive, cross-product dataset quantifying the biodistribution of three IM-administered mRNA-LNP drug products formulated in an identical LNP matrix (mRNA-1273, mRNA-1647, and a reporter mRNA [NPI-Luc] formulated in lipid H LNPs), in Sprague-Dawley rats at conservatively high (>130-fold human) dose levels when normalized to body weight. Thus, the results represent a conservative and extreme estimate of tissue and systemic exposure of mRNA, LNP components, and/or SARS-CoV-2 spike protein.

Across the Moderna products evaluated here, which includes both single- and multi-valent vaccines, the tissues with the highest exposure were consistently the injection site, draining lymph nodes, and spleen, with minimal distribution to other organs. This biodistribution profile is generally consistent with results used to support other licensed IM vaccine platforms in rodents and/or rabbits, including Pfizer-BioNTech’s BNT162b2 mRNA-LNP vaccine,17 as well as adjuvanted protein, and viral-vector vaccines.24,25,26,27 This shared biodistribution pattern suggests a conserved mechanism of vaccine uptake and antigen presentation primarily within local and lymphoid compartments after IM administration to induce an immune response. The lymphoid tropism of mRNA-LNP vaccines also likely reflects the optimized nanoparticle size (60–120 nm), surface charge, and apolipoprotein-E-mediated interactions that promote retention within draining lymph nodes.28,29,30,31 A lymphoid-focused pattern of distribution was also observed in non-human primates following IM administration of mRNA-LNPs formulated in the same LNP as the product used here,32 indicating cross-species consistency and the translational relevance to higher-order species. At the same time, the present findings should be interpreted within the context of the shared LNP platform evaluated here, because alternate LNP compositions and formulation attributes have been reported to redirect tissue distribution.28,29 Consistent with that, differences in kinetics and tissue persistence have been observed in some tissues with other mRNA vaccines using a different ionizable lipid (e.g., BNT162b230) substantiating that the ionizable lipid and/or other LNP characteristics, including lipid composition and particle attributes, are likely critical determinants of distribution and persistence.31,32,33,34,35

mRNA and LNP components cleared rapidly (within 48 h in serum; <14 days in lymphoid tissues) and first-order elimination modeling of these analytes suggested that concentrations continued decline to below quantifiable or very low levels by ∼2 weeks in the tissue with the longest T1/2 (spleen). Furthermore, there was no accumulation of mRNA or lipid H in the systemic circulation or tissues following a second contralateral dose of mRNA-1273 administered 4 weeks after the first dose, providing support for lack of meaningful increase in overall exposure under the repeat-dose paradigm tested here; however, these do not address potential accumulation in tissues for all possible repeat-dose scenarios (e.g., when using the same injection site). When considering AUC, exposures of mRNA and lipid H in non-lymphoid tissues were minimal, representing only a small fraction of that observed at the injection site. Albeit at low levels, the liver exhibited the greatest exposure among distal tissues (<3% relative to the injection site). Other tissues displayed only trace amounts compared with the injection site, which serves as the primary depot following IM administration. In support of hepatobiliary clearance behavior, published data using a different ionizable lipid (lipid 5) via the intravenous (IV) route (which yields substantially higher systemic exposure than IM), demonstrated similar rapid distribution predominantly to the liver and intestines by 24 h post-dose, supporting the expected clearance behavior of LNPs upon systemic entry.36

Across studies, trace but detectable levels of mRNA or lipid H in distal tissues, such as heart, lung, and brain, were near the assay LLOQ and highly variable. Quantifiable levels in these tissues fluctuated across studies, sexes, time points, and even within the same drug product. Interpretation of these low-level observations is further constrained by the analytical conventions used of noncompartmental summaries, including BLQ-to-zero handling, AUC calculation only to the last quantifiable time point, and study-specific use of plasma vs. serum as the systemic comparator, which may introduce conservative bias when analyte levels are near the assay limit. Consistent with this, following administration of mRNA-1647, concentrations of the six different mRNAs exhibited discordant levels across these low-concentration tissues. There were also occasional detections of mRNA without corresponding lipid H, which are unlikely to represent true exposure since unprotected RNA is rapidly degraded in biological fluids and thus not expected to persist in tissues without LNP encapsulation.35,37 These low-level signals are therefore attributed to analytical noise near the LLOQ or to residual vascular content after perfusion, rather than a physiologically relevant distribution. Perfusion substantially reduces intravascular content but cannot completely clear capillary beds; residual blood can thus account for trace detections near the assay limit. Additionally, trace brain mRNA signals are not considered indicative of blood-brain barrier penetration, given the restrictive barrier architecture, the LNP size (∼60–120 nm), and the lack of active transport ligands of LNPs.38,39,40,41

Discrepancies in tissue kinetics were occasionally noted between lipid H and encapsulated mRNAs, which can be attributed to both biological and technical factors. First, the intracellular degradation of mRNA and lipids differ: mRNA is inherently more labile than lipids, and is subject to nuclease degradation and endosomal processing,42,43 whereas lipid clearance follows distinct esterase-mediated pathways. Second, differences in the LLOQ of the bioanalytical methods used to measure mRNA and lipid could contribute to variable apparent kinetics. Direct quantitative comparison of mRNA exposure across products should be interpreted cautiously because different qualified bioanalytical methods (reverse transcription-quantitative polymerase chain reaction [RT-qPCR] or branched DNA [bDNA]) were used across studies. While these methods were each fit for purpose, differences in assay sensitivity and format may influence absolute concentration estimates. Accordingly, the strongest cross-product inference supported by the present work is the consistency of the overall biodistribution pattern rather than strict numerical equivalence of exposure metrics across products. Despite these differences, the tissues with the highest mRNA exposure were also always those with the highest lipid H exposure. Thus, for mRNA-LNP drug products, these results support the practical use of either analyte as a surrogate for the other when the goal is to compare tissue exposure across organs.

To our knowledge, this is the first characterization of SARS-CoV-2 spike protein exposure and clearance in a SARS-CoV-2-naïve model, addressing a key limitation of human studies where pre-existing or subclinical SARS-CoV-2 infection cannot be excluded. We found that in rats, where there were no prior existing antibodies against SARS-CoV-2 prior to administration mRNA-1273, unbound (free) spike protein is transient in the systemic circulation. This is demonstrated by a serum T1/2 of ∼18 h (albeit calculable only after a first dose), >98% simulated decline within 5 days post-dose, and lack of accumulation following a second dose of mRNA-1273. Recent claims that spike protein persists for months following vaccination are either not aligned with the broader body of scientific evidence or have technical limitations that may confound interpretation. For example, Bhattacharjee et al.44 relied on small, uncontrolled samples, and bioanalytical methods that could not distinguish vaccine-derived spike protein from infection-derived spike. Additionally, Patel et al.45 asserted persistence of spike protein in non-classical CD16+ monocytes for more than 200 days post-vaccination, despite the cells’ ∼7 days lifespan, suggesting latent or ongoing SARS-CoV-2 infection as a more plausible explanation.46 Altogether, our data, derived from a well-controlled animal study, along with a clinical study,47 do not support sustained unbound (free) spike protein persistence beyond 5 days after vaccination.

Since the ELISA used here was designed to detect only unbound (free) antigen, absence of detectable signal should not be interpreted as definitive absence of spike-derived antigen in all possible forms (e.g., bound to antigen). This distinction is particularly relevant following repeat dosing. AUEClast for systemic spike protein levels was estimable after the first dose, but not after the second dose, due to insufficient quantifiable serum time points. The most plausible explanation is assay-related rather than a true absence of protein expression following repeat dosing. Specifically, the robust, anti-SARS-CoV-2 S2P-specific polyclonal antibodies elicited after the first dose likely formed immune complexes with newly expressed spike protein and masked epitopes recognized by the capture and detection antibodies. This potential interference could have contributed to the apparent lack of detectable free spike protein after the booster dose and support the use of orthogonal detection methods (e.g., liquid chromatography-mass spectrometry [MS]) in future investigation.

Lipid H underwent rapid and predictable biotransformation with intact lipid representing the predominating circulating species after IV dosing in Sprague-Dawley rats. Ester hydrolysis and β oxidation were the primary metabolic pathways observed in vivo and in vitro, generating metabolites that were efficiently cleared through renal and biliary routes. Low-molecular weight, hydrophilic metabolites were especially enriched in urine, consistent with renal filtration. Taken together, the extensive metabolism of lipid H, rapid overall clearance (within 24 h) and the efficient elimination of both parent and its metabolites to <3% of the maximum level indicate that lipid H is unlikely to accumulate upon repeat IM dosing. In addition, the metabolic pathways of the ionizable lipid were conserved across species, with comparable metabolite profiles observed in rat, non-human primate, and human hepatocytes and no evidence of human-specific metabolites. This cross-species conservation supports the clinical safety assessment of the vaccine platform by reducing concern that humans generate unique or unexpected lipid H metabolites not represented in preclinical species. Together, these observations support the translational relevance of the rat model for describing the overall disposition of mRNA-LNP components.

Using Sprague-Dawley rats as the model system enabled a comprehensive characterization of the mRNA, lipid H, and expressed antigen exposure in both blood and tissues, the latter of which are measurements that are not feasible to evaluate in humans. However, several limitations should be acknowledged. The doses that were used were intentionally supratherapeutic (>130-fold human when normalized to species body weight) to support conservative biodistribution assessment under exaggerated exposures, rather than to establish direct quantitative equivalence to human tissue exposure. These doses were aligned with toxicology study dose ranges, consistent with regulatory expectations for correlating tissue exposure with toxicological findings. Accordingly, these studies are most informative for defining qualitative biodistribution patterns and upper-bound exposure scenarios. Recently published human PK data for lipid H and mRNA in systemic circulation following mRNA-1273 vaccination help contextualize our findings from rats. Kent et al.48 reported that both vaccine mRNA and the ionizable lipid from mRNA-1273 are detectable in plasma shortly after vaccination and decline progressively over time, with ionizable lipid concentrations approaching background levels within approximately 7–14 days, and low levels of vaccine mRNA remaining detectable in only a subset of subjects for up to 14–28 days post-vaccination. These findings are directionally consistent with the rat data, with faster clearance in rats than humans, as expected.49,50,51 The principal value of the rat studies is to define the overall tissue-distribution pattern, characterize relative persistence across tissues, and to determine potential risk of accumulation under the study conditions used here. Thus, the ∼2 weeks clearance observed in rats should be interpreted as a conservative, model-based estimate of exposure decline in the highest-exposure tissues under exaggerated dosing conditions, rather than a direct predictive timeline for humans. Moreover, given limitations in the final sampling time points, modeling was required to estimate the later phase of elimination, consistent with standard pharmacokinetic approaches for truncated datasets.52 While these model-based projections beyond the observed time frame should not be interpreted as direct observations, they match the overall patterns seen with reported human data (which was evaluated out to 4 weeks after vaccination),48 suggesting that the rat results remain reliable even with this limitation.

In conclusion, by harmonizing the dose level, route of administration, and LNP composition, and by incorporating an evaluation of distribution following repeat dosing, we provide a comprehensive view of the metabolism, clearance, and tissue distribution of Moderna mRNA-LNP vaccines. The different mRNA cargo or immune response did not appear to alter the overall tissue tropism, and exposures for all vaccine components were transient and declined over time, with no evidence of sustained quantifiable signal beyond the reported observation and modeling windows. Furthermore, the metabolism of lipid H favored clearance over storage. The broadly consistent distribution and clearance patterns observed across products (spanning both single- and multi-valent vaccines) provide supportive evidence for considering biodistribution bridging when products share the same LNP composition and route of administration. This approach aligns with the risk-based framework endorsed by major regulatory agencies (e.g., European Medicines Agency, United States Food and Drug Administration [FDA], Japan’s Pharmaceuticals and Medical Devices Agency, and the World Health Organization) and supports scientifically justified read-across for appropriately matched LNP-formulated nucleic acid medicines. It also advances efforts to reduce redundant in vivo testing consistent with the 3 Rs (replacement, reduction, and refinement), and the FDA’s roadmap promoting scientifically justified non-animal strategies to support drug development.53

Materials and methods

Animals and ethical compliance

All animal care and experimental procedures were conducted in compliance with the Guide for the Care and Use of Laboratory Animals and were approved by the Institutional Animal Care and Use Committee (Charles River Laboratories, Inc, Mattawan, MI, USA or Worcester, MA, USA) or by the ethical committee of Charles River Laboratories Safety Assessment Services.

mRNA-LNP products

Biodistribution studies were conducted using three mRNA-LNP products, where mRNA was encapsulated in the same four lipids (lipid H, PEG2000-DMG, cholesterol, and DSPC) with generally similar ratios of lipids to mRNA: (1) mRNA-1273 (Spikevax), Moderna’s licensed SARS-CoV-2 vaccine; (2) mRNA-1647, an investigational CMV vaccine candidate (containing six distinct mRNAs); and (3) an mRNA-LNP product that encodes a reporter protein, NPI-Luc. For metabolism and excretion studies, a product encapsulating a non-translating mRNA was used (non-translating factor IX or NT-FIX), which was formulated similarly to the other products in the same 4-lipid matrix. All dose levels listed in the in vivo studies are reflective of the mRNA dose (μg/dose).

The mRNA-LNP products were manufactured using methods described previously.32 In brief, mRNA was produced by in vitro transcription using established methods, followed by enzymatic capping and purification. LNP formulations were generated via nanoprecipitation by mixing an ethanolic lipid phase (containing ionizable lipid, DSPC, cholesterol, and PEG-lipid) with an aqueous mRNA phase under acidic conditions. The resulting mRNA-LNPs were buffer exchanged into a physiologically relevant buffer, sterile filtered, and characterized for key physicochemical attributes prior to in vivo use. All LNP formulations evaluated in this study are consistent with clinical approved mRNA-LNP vaccine compositions and exhibited comparable physicochemical properties.

The test materials were supplied preformulated. While the study with mRNA-1273 and NPI-Luc used the formulations as received, the study with mRNA-1647 used phosphate-buffered saline (PBS) dilution to achieve target dose concentration.

Of the four lipids in the LNP formulation, lipid H was the only one evaluated in the biodistribution, metabolism, and excretion studies for the following reasons: (1) cholesterol is widely present throughout the body and therefore it is not possible to differentiate drug product-delivered cholesterol exposure from endogenous, naturally occurring cholesterol, and (2) DSPC and PEGylated molecules are lipids already in approved drug products54 at doses in excess of those in the clinical IM doses of the vaccines and/or as IV administered products (which would result in significantly greater systemic exposure than with IM administration).

Experimental designs

The biodistribution studies evaluating mRNA-1273 and NPI-Luc in lipid H-containing LNPs were conducted by Charles River Laboratories, Inc., Mattawan, MI, USA and used male and female Crl:CD(SD) Sprague-Dawley rats (Charles River Laboratories, Inc). Prior to initiation of dosing, the average age and weight range of animals in the mRNA-1273 study were 10.5–13.5 weeks and 225 g (females) and 408 g (males), respectively. In the NPI-Luc study, these biometrics traits ranged from 10 to 11 weeks of age and weights of 223 g (females) and 351 g (males). The biodistribution study evaluating mRNA-1647 was conducted by Charles River Laboratories Montreal ULC, Sherbrooke, QC, Canada and used male Crl:CD(SD) Sprague-Dawley rats (Charles River Laboratories, Inc). Prior to initiation of dosing, the age and average weight of animals in the mRNA-1647 study was 8 weeks and 323 g, respectively. The in vivo metabolism and excretion study with NT-FIX mRNA in lipid H LNPs was conducted by Charles River Laboratories, Inc, Worcester, MA, USA and used male Sprague-Dawley rats (Envigo EMS, average weight 259 g at initiation of dosing).

In all the studies, animals were acclimated to the facilities for at least 6–7 days and randomly assigned into groups. Rats were housed in microisolator cages under controlled environmental conditions (12-h light/dark cycle, 18°C–26°C, 40%–70% humidity) with ad libitum access to food (MI Nutrition International Certified Rodent Chow No. 5CR4, 14% protein or Teklad Global Diets, 18% Protein Rodent Diet 2018) and locally sourced municipal tap water that was passed through reverse osmosis, ultraviolet irradiated, and provided ad libitum.

In vivo biodistribution studies

The route of administration used to assess biodistribution was IM, consistent with the clinical route. All doses were administered via IM injection into the lateral thigh (quadriceps) in a 200-μL volume using a 27-gauge needle. In line with biodistribution regulatory guidance,55 the selected dose levels of the mRNA-LNP products bridged dose levels used in toxicology studies or were the maximum feasible dose based on the drug concentration and the maximum allowable dosing volume for IM administration (200 μL/dose).55 In addition, the use of intentionally high, supratherapeutic doses enabled quantification of low exposure tissue samples, which would be expected to fall below the limit of quantification at lower, clinically relevant exposure levels.

Table S1 summarizes the administered rat dose levels for each product on both a flat per dose basis (e.g., μg/dose) and body-weight normalized basis (μg/kg), together with example corresponding clinical comparator doses and the resulting “safety margin” relative to clinical doses. Rat doses were normalized using an assumed average rat body weight of 0.3 kg, and human doses were normalized using an assumed conservative body weight of 60 kg. Safety margins were calculated as the ratio of rat dose (in μg/kg) to human dose (in μg/kg). As shown in Table S1, this yielded body weight-normalized safety margins exceeding 130-fold relative to clinical doses. Once in circulation, interspecies disposition and PK of extravascularly administered drugs are governed primarily by body-size-related clearance.49,50,51 Therefore, given that these IM biodistribution studies are aimed at assessing exposure and kinetics systemically, it is most physiologically relevant to normalize dose based on body weight (μg/kg) when making rat-human comparisons. Thus, the studies intentionally exceeded clinical dose levels on μg/kg basis to enable comprehensive systemic biodistribution assessments at exaggerated exposures, in alignment with International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use S12 Guidance.

mRNA-1647 and NPI-Luc mRNA in lipid H-containing LNPs were administered as a single IM dose. mRNA-1273 was administered IM twice, once on day 1 and once contralaterally 4 weeks later (i.e., on day 28) to evaluate tissue distribution after repeat administration and when using the most compressed clinical dosing regimen.

Four animals/sex/time point/dosing occasion were administered mRNA-1273, and three and five animals/sex/time point were administered NPI-Luc in lipid H-containing LNPs and mRNA-1647, respectively. Animals were monitored daily for morbidity, mortality, clinical signs, and local injection site reactions. Body weights were recorded at baseline and prior to necropsy. In the study with mRNA-1273, body weights were also recorded weekly recordings post-dose of mRNA-1273. No unscheduled deaths occurred, and all animals completed the in-life phase as planned.

Tissue collection and processing

Following administration of 1 or 2 doses of mRNA-1273, whole blood and tissues were collected at 0.16, 1, 4, 10, 24, 72, 120, 168, and 336 h post-dose. Following administration of NPI-Luc mRNA in lipid H-containing LNPs, whole blood and tissues were collected at 0.16, 1, 4, 10, 24, 72, 120, and 168 h post-dose. Following administration of mRNA-1647, whole blood and tissues were collected at 0 (pre-dose), 2, 8, 24, 48, 72, and 120 h post-dose for mRNA-1647 quantitation. Animals were euthanized by deep isoflurane inhalation at predefined time points and were perfused with sodium chloride 0.9%, heparin (1,000 IU/mL), 1% sodium nitrite, and PBS until the fluid runs were sufficiently clear. Tissues were then excised rapidly, rinsed in cold PBS, blotted dry, weighed, and immediately snap-frozen in liquid nitrogen (within approximately 1 min of collection), maintained on dry ice, and stored at ≤−80°C until analysis. Samples were handled under RNase-free conditions using sterile instruments, with measures taken to avoid cross-contamination between samples. Immediately following collection, tissues were either immersed in RNA stabilization solution or snap-frozen in liquid nitrogen and maintained at ≤−80°C. All samples were kept on wet or dry ice during handling to minimize RNA and protein degradation. Whole blood was collected via cardiac puncture into K2-EDTA tubes for plasma or serum separation.

Collected tissues across the three studies primarily included: bone marrow, brain, eye, heart, injection site muscle, jejunum, kidney, liver, lymph nodes (axillary, inguinal, and popliteal), ovary, spleen, stomach, testes, thymus, and uterus. The selected tissues reflect both regulatory expectations for the IM route of administration and data derived previously from the developer demonstrating exposure (or lack thereof) in relevant tissues. Differences in tissues collected in the studies are noted in the “results” section in the appropriate tables.

Bioanalytical methods for biodistribution studies

A summary of the bioanalytical methods and assay rigor used for the biodistribution experiments is provided in Table S2 with detailed methods described below. In addition, analyte stability under relevant storage and processing conditions (including freeze-thaw stability) was assessed during assay qualification/validation, and methods demonstrated acceptable performance characteristics, supporting reliable quantification under the sample handling conditions used.

mRNA quantification

RT-qPCR was used for quantification in the mRNA-1273 study and reflects the most sensitive analytical method for detecting RNA in biological matrices. Single-plex bDNA hybridization and multiplex bDNA bead-based detection pre-dated use of RT-qPCR and were used for the NPI-Luc in lipid H-containing LNP and mRNA-1647 studies, respectively. Despite these differences in methodology, each method was qualified for precision, accuracy, selectivity, and sensitivity.

Quantification of the mRNA in mRNA-1273 was performed using a one-step RT-qPCR assay. Tissues were homogenized prior to RNA isolation in Maxwell RSC simplyRNA homogenization buffer. Total RNA was extracted from tissue homogenates and serum using the Maxwell RSC simplyRNA isolation procedure. Complementary DNA was synthesized using reverse transcriptase and amplified with TaqMan probe sets specific to the mRNA-1273 sequence on a real-time PCR system. Calibration curves were generated using known concentrations of mRNA-1273 drug product with curve range from 0.00085 to 170 ng/mL in rat serum and 0.00085 to 170 ng/g in tissue.

Levels of mRNA from NPI-Luc in lipid H-containing LNPs were quantified using a single-plex bDNA hybridization assay (QuantiGene 2.0 assay system, Thermo Fisher Scientific). Tissue homogenates and serum samples were lysed in the presence of proteinase K and hybridized with sequence-specific capture, label, and blocker probes targeting the luciferase transcript. Chemiluminescent signals were generated following amplification with alkaline phosphatase-conjugated probes and substrate and measured on a SpectraMax M3 luminometer. The calibration curve ranged from 0.125 to 8.00 ng/mL in rat serum. mRNA levels in each rat tissue homogenate were calculated against a single calibration curve prepared from pooled tissue homogenates with a calibration range of 0.125–16.0 ng/g. The pooled homogenates were generated by combining equal proportions of rat liver, spleen, heart, lung, kidney, brain, uterus, ovary, testes, eye, jejunum, axillary lymph node, tail, skeletal muscle, stomach, and thymus. The levels of the six different mRNA in mRNA-1647 were quantified using a multiplex bDNA hybridization assay (QuantiGene Plex, Thermo Fisher Scientific/Affymetrix) run on a Luminex Bio-Plex Suspension Array system. This assay simultaneously quantified all six mRNAs (gB, gH, gL, UL128, UL130, and UL131A). Plasma and perfused tissue samples were lysed, homogenized, and hybridized with construct-specific capture, label, and blocker probes conjugated to distinct bead regions. Chemiluminescent signals were amplified through pre-amplifier, amplifier, and streptavidin-phycoerythrin-labeling steps, then detected bead-by-bead on the Luminex platform. The multiplex design allowed assessment of each construct in parallel and demonstrated highly concordant PK across constructs. The calibration curve ranged from 0.01 to 5 ng/mL in plasma and tissue homogenates, depending on the construct.

Lipid H quantification

Quantification of lipid H in plasma and tissue samples was conducted using a qualified liquid chromatography-tandem MS method. Tissues were homogenized in PBS and subjected to protein precipitation using a 1:1 mixture of acetonitrile and methanol. After centrifugation, the supernatant was transferred to 96-well plates for analysis.

Chromatographic separation was achieved on a C8 reversed-phase column (50 mm × 2.1 mm, 5-μm particle size) using gradient elution with mobile phases consisting of aqueous ammonium formate and acetonitrile. A 10-μL injection volume was used, and each run had a total time of 4 min. The detection was performed using a Sciex API 5000 mass spectrometer equipped with electrospray ionization in positive ion mode.

Calibration curves were generated in matrix-matched standards and covered a concentration range of 0.5–500 ng/mL (or ng/g), with linear regression and appropriate weighting.

Encoded protein quantification

Levels of the encoded SARS-CoV-2 spike protein and NPI-Luc protein were quantified using a qualified ELISA method. For the SARS-CoV-2 spike protein assay, the ELISA method was designed to measure the concentration of free spike protein present in the serum and tissue samples. Accordingly, the assay was not intended to quantify spike-derived antigen in all possible forms, and detectability may be reduced if antigen is bound in immune complexes or otherwise masked from the capture and/or detection of antibodies.

Quantification of expressed NPI-Luc protein was performed using qualified electrochemiluminescence-based ligand-binding assays. Tissue homogenates were prepared and diluted to a minimum required dilution (MRD) in assay buffer before analysis. Plates were coated with hamster anti-V5 antibody, followed by sequential incubation of the sample with anti-firefly luciferase and SulfoTag-labeled secondary antibodies to generate an electro-chemiluminescent signal proportional to analyte concentration. The assay demonstrated acceptable accuracy, precision, and selectivity, with an LLOQ of 200 ng/g in tissue homogenates.

Quantification of the expressed SARS-CoV-2 spike protein following mRNA-1273 administration was performed using qualified colorimetric ligand-binding (sandwich ELISA) assays. The assays employed mouse anti-SARS-CoV-2 whole spike protein as the capture antibody and biotin-conjugated rabbit anti-spike antibody for detection, followed by streptavidin-peroxidase for signal generation. Serum samples were analyzed following an MRD of 1:20, while tissue homogenates were analyzed following an MRD of 1:10. The assays demonstrated acceptable accuracy, precision, and selectivity, with LLOQ of 4.57 ng/mL in serum and 91.4 ng/g in tissue homogenates.

Antibody titers against SARS-CoV-2

The immunogenicity of mRNA-1273 was evaluated by assessing binding antibody titers against SARS-CoV-2. High-binding 96-well plates (MaxiSorp) were coated overnight (2°C–8°C) with SARS-CoV-2 S2P protein (GenScript; 1.5 μg/mL in Dulbecco’s PBS), washed 3× with Dulbecco’s PBS-0.05% Tween 20, and blocked with StartingBlock (PBS) for ≥45 min at room temperature. Rat serum standards and samples, diluted in StartingBlock T20 (PBS), were added (100 μL/well) and incubated for 2 h. After washing, goat anti-rat immunoglobulin(H + L)-horseradish peroxidase (KPL/SeraCare; 1:7,000 in StartingBlock T20) was applied for 1 h, plates were washed, and 3,3′,5,5′-tetramethylbenzidine (SureBlue Reserve) was developed for 30 min protected from light. Absorbance was read at 650 nm (VersaMax), and antibody units (AU)/mL were calculated by 4-parameter logistic interpolation from a standard curve generated using serial dilutions (1:102 to 1:107) of a pooled hyperimmune rat serum reference standard. The analytical LLOQ was 0.059 AU/mL, corresponding to the lowest dilution of the standard producing signal ≥3.5-fold above background (≈1:3.2 × 105 dilution). Samples were tested at an MRD of 1:500. Values below the analytical LLOQ (0.059 AU/mL) correspond to a reportable LLOQ of 30 AU/mL in undiluted serum. Samples exceeding the upper calibration standard were further diluted as necessary to fall within the validated range, consistent with standard ELISA practice.

Metabolism and excretion studies

A summary of the bioanalytical methods and assay rigor used for the metabolism and excretion experiments is provided in Table S2 with detailed methods described below.

In vitro metabolite identification and profiling of lipid H

Primary rat, monkey, and human hepatocytes were thawed and resuspended in Williams’ E medium at 1 × 106 cells/mL, with 250 μL of cell suspension or media used per condition; heat-denatured controls were prepared by boiling hepatocytes at 100°C for 15 min. A lipid H LNP stock solution was diluted in media to 100 μM and further diluted in cell suspensions to achieve a final concentration of 10 μM. Samples were collected at 0, 4, and 24 h, and lipid H and its metabolites were extracted with 250 μL cold acetonitrile, centrifuged at 10,000 × g for 10 min at 10°C, and the supernatant was removed for analysis. Metabolite profiling was performed on an Agilent 1290 UPLC coupled to an Agilent 6550 QTOF mass spectrometer using a reverse-phase C18 column, with data acquired in positive ion mode and instrument parameters optimized with a lipid H standard; both data-dependent and targeted tandem MS were used for metabolite identification.

In vivo metabolism and excretion study

The IV route of administration was used in the in vivo study to maximize systemic exposure and ensure the ability to detect lipid H parent and metabolites in plasma, bile, and urine. Route of administration would not be expected to influence metabolism or elimination routes; thus, the data derived using IV dosing is also considered supportive of IM dosing. One group of male Sprague-Dawley rats (n = 3; bilateral jugular-vein cannulated/bile duct cannulated) was administered NT-FIX mRNA in lipid H LNPs at a dose of 0.672 mg/kg (dose concentration 0.48 mg/mL; dose volume 1.4 mL/kg, diluted in buffer with Tris/sucrose/acetate, pH7.5) via a 10-min infusion followed by a 0.5 mL post-dose flush. Serial blood samples were collected from each animal prior to dosing and at 2, 6, and 24 h post-infusion start. Blood samples, which were collected into tubes containing K2-EDTA, were processed for plasma. Urine and bile were collected from 0 to 2, 2 to 6, and 6 to 24 h post-infusion start. Metabolite profiling was conducted by liquid chromatography-high-resolution MS.

Kinetic analysis, reporting, and simulation of concentration-time profile

PK and PD evaluation

PK and PD parameters were estimated using noncompartmental analysis in Phoenix PK software. For PK assessments (mRNA and lipid H), concentration-time data in plasma or serum, respectively, and all tissues except the injection site were analyzed using an extravascular model. The IV bolus model was applied for the injection site. PK parameters were appropriately calculated based on either lipid or mRNA dose. For the kinetic evaluation of mRNA-1273, all parameters were generated from composite lipid H, and mRNA concentrations collected on days 1 and 28, whenever available, assuming a pre-dose concentration of zero on day 28 for the extravascular route of administration. At each time point, samples from four animals per sex were collected and combined to generate a composite concentration value for each sex/group. Composite concentration values across time were then used to construct one concentration-time profile per sex/group, from which PK parameters such as Cmax and AUC were derived. Since these parameters were calculated from composite group-level profiles rather than individual animals, individual animal Cmax and AUC values were not available.

For the PD assessment (expressed protein), concentration-time profiles of the encoded protein (NPI-Luc or SARS-CoV-2 spike protein) in serum and tissues were analyzed using the same non-compartmental approach consistent with the extravascular route of administration. For the study involving mRNA-1273, the day 28 pre-dose concentration was also assumed to be zero since the previous collection concentration was below the LLOQ.

For both PK and PD analyses, concentrations below the LLOQ were treated as zero (including assignment of BLQ values to zero), and concentrations above the calibration range were treated as the upper limit of quantitation (ULOQ) for the purposes of the PK data analysis. Tissue-to-systemic exposure ratios were calculated assuming equivalence of 1 g tissue to 1 mL plasma or serum, consistent with standard approaches for estimating tissue exposure. Systemic exposure was evaluated using either plasma or serum depending on the study and analyte. These methodological choices were applied consistently across datasets to enable comparison but represent simplifying assumptions that may introduce conservative bias, particularly for low-level measurements. Comprehensive estimation of primary or secondary PK parameters (including Cmax, time to maximum concentration [Tmax], area under the concentration-time curve from time 0 to time t [AUC(0-t)], area under the concentration-time curve to the last measurable concentration [AUClast], area under the concentration-time curve from time 0 to infinity [AUC(0-inf)], time of the last measurable concentration [Tlast], effective T1/2, mean residence time to the last measurable concentration [MRTlast], clearance normalized by bioavailability [Cl/F], apparent volume of distribution during the terminal phase after non-IV administration [Vz/F], tissue to plasma/serum ratio [%] of AUC, tissue to plasma/serum ratio [%] of Cmax), and PD parameters (including Emax, TEmax, area under the effect-time curve from time 0 to time t [AUEC(0-t)], AUEClast, time of the last measurable effect [TElast], effective T1/2, MRTlast) were calculated for each of the studies. While all PK and PD parameters were calculated in the studies to support regulatory submissions, the following parameters were selected for reporting in this manuscript as they best characterize tissue exposure and relative persistence and clearance: AUC(0-t), AUClast, tissue to plasma/serum ratio (%) of AUC and effective T1/2 for PK, and Emax, AUEClast, TElast, and effective T1/2 for PD.

Tissue-to-systemic exposure ratios (%) were calculated as (AUCtissue/AUCsystemic) × 100. AUC was defined as the area under the concentration versus time curve from the start of dose administration to the last observed quantifiable concentration using the linear trapezoidal method to calculate.

Within the datasets reported here, there were some occasions where data are presented as “not determined” (ND) values or “not calculable” (NC). ND values indicate tissues or time points that were not collected as part of the pre-defined study design(s). NC values denote pharmacokinetic parameters that could not be reliably estimated due to insufficient data to calculate parameters, for example, AUC or effective half-life, because there were fewer than three consecutive time points with quantifiable values.

In tissues with sparse values near the assay limit, derived AUC, Effective T1/2, and tissue-to-systemic ratios may be sensitive to BLQ handling and may underestimate low tail exposure. In addition, systemic comparators were not identical across all studies: plasma was used for mRNA-1647, whereas serum was used for NPI-Luc and mRNA-1273. These study-specific matrix differences, together with residual vascular content despite perfusion, should be considered when interpreting distal tissue signals or making precise quantitative comparisons across products.

Combined sex kinetic data reporting

To characterize broad-stroke kinetic estimates with the mRNA-1273 and NPI-Luc mRNA in lipid H-containing LNPs dataset, we pooled male and female PK and PD data (otherwise referred to as “sex combined”), except where tissues were sex-dependent (e.g., testes or ovaries). Combining male and female data increased sample size, effective power and avoided over-partitioning the dataset. Sex differences were, however, always evaluated as a potential covariate in the studies. There were no overall sex differences in individual serum, plasma, and or tissue mRNA concentrations in the study with NPI-Luc mRNA in lipid H-containing LNPs (Table S3), which was only administered as a single dose. In the study with mRNA-1273, where two dose administrations were administered over the course of 1 month and significant sex-dependent weight difference existed, mRNA exposures in some matrices were higher in females than males (Table S3). The differences in exposure are attributed to differences in body weight: female body weights were approximately 1.5- to 2-fold lower than males across dosing days (Figure S1). As a result, females received proportionally 1.5- to 2-fold higher doses of mRNA-1273 per kg of body weight than males, providing the most likely explanation for the sex-related differences in exposure.

Simulation of mRNA and spike protein elimination profiles

Simulated concentration-time profiles for mRNA and spike protein were generated using observed Cmax or Emax values and tissue- or plasma-specific effective T1/2. First-order elimination kinetics were assumed, with concentrations estimated using Ct = Cmax or Emax × e−kel∗t and kel = ln (2)/t1/2. Profiles were simulated over six half-lives to illustrate the projected later phase of analyte decline beyond the observed sampling window, corresponding to ∼1.6% of Cmax or Emax remaining (98.4% eliminated). Simulations and curve fitting were conducted in GraphPad Prism (GraphPad Software, San Diego, CA).

Statistical analysis

To assess differences in exposure between the first and second dose of mRNA-1273, data are presented as mean values ±standard deviation (SD) from two independent dosing days (Figure 2). Comparisons between dosing days were performed using an unpaired Student’s t test. Statistical analyses were conducted using GraphPad Prism (GraphPad Software, San Diego, CA, USA). A p value of less than 0.05 was considered statistically significant.

Data and code availability

Supporting data may be made available from the corresponding author upon reasonable request

Acknowledgments

We acknowledge John Wickwire, Harkewal Singh, Saara Mansouri, Matthew Gorman, Shyam Kumar Gudey, Kelin Wang, and Charles River Laboratories. Medical writing and editorial assistance were provided by MEDiSTRAVA in accordance with Good Publication Practice guidelines and was under the direction of the authors. This work was supported by Moderna, Inc.

Author contributions

S.M.G.G., C.R., Y.L., and N.C. contributed to the study concept and design; S.M.G.G., C.R., Y.L., and N.C. analyzed and interpreted the data. All authors contributed to the drafting and critical review of this manuscript and approved the final draft.

Declaration of interests

S.M.G.G., C.R., Y.L., and N.C. are employees of and shareholders in Moderna, Inc.

Declaration of generative AI and AI-assisted technologies in the writing process

During manuscript drafting, the authors used ChatGPT for language editing and phrasing suggestions only. No AI tools were used for data analysis, interpretation, or generation of scientific conclusions, and all final text was reviewed and approved by the authors.

Footnotes

Supplemental information can be found online at https://doi.org/10.1016/j.omtn.2026.102948.

Supplemental information

Document S1. Figures S1–S3 and Tables S1–S5
mmc1.pdf (252.3KB, pdf)
Table S6. Example data demonstrating low level distal-tissue detections in brain and heart across the three biodistribution datasets: number of quantifiable samples and mean quantifiable concentrations by time point
mmc2.xlsx (14.2KB, xlsx)
Document S2. Article plus supplemental information
mmc3.pdf (2.2MB, pdf)

References

  • 1.Hassett K.J., Benenato K.E., Jacquinet E., Lee A., Woods A., Yuzhakov O., Himansu S., Deterling J., Geilich B.M., Ketova T., et al. Optimization of lipid nanoparticles for intramuscular administration of mRNA vaccines. Mol. Ther. Nucleic Acids. 2019;15:1–11. doi: 10.1016/j.omtn.2019.01.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Gebre M.S., Brito L.A., Tostanoski L.H., Edwards D.K., Carfi A., Barouch D.H. Novel approaches for vaccine development. Cell. 2021;184:1589–1603. doi: 10.1016/j.cell.2021.02.030. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Pardi N., Hogan M.J., Porter F.W., Weissman D. mRNA vaccines—a new era in vaccinology. Nat. Rev. Drug Discov. 2018;17:261–279. doi: 10.1038/nrd.2017.243. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Espeseth A.S., Cejas P.J., Citron M.P., Wang D., DiStefano D.J., Callahan C., Donnell G.O., Galli J.D., Swoyer R., Touch S., et al. Modified mRNA/lipid nanoparticle-based vaccines expressing respiratory syncytial virus F protein variants are immunogenic and protective in rodent models of RSV infection. NPJ Vaccines. 2020;5:16. doi: 10.1038/s41541-020-0163-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Victora G.D., Nussenzweig M.C. Germinal centers. Annu. Rev. Immunol. 2022;40:413–442. doi: 10.1146/annurev-immunol-120419-022408. [DOI] [PubMed] [Google Scholar]
  • 6.US Food and Drug Administration Summary basis for regulatory action - SPIKEVAX. 2022. https://www.fda.gov/media/155931/download
  • 7.US Food and Drug Administration Summary basis for regulatory action - MRESVIA. 2024. https://www.fda.gov/media/179634/download?attachment
  • 8.US Food and Drug Administration Summary Basis for regulatory action - COMIRNATY. 2021. https://www.fda.gov/media/151733/download
  • 9.Soens M., Ananworanich J., Hicks B., Lucas K.J., Cardona J., Sher L., Livermore G., Schaefers K., Henry C., Choi A., et al. A phase 3 randomized safety and immunogenicity trial of mRNA-1010 seasonal influenza vaccine in adults. Vaccine. 2025;50 doi: 10.1016/j.vaccine.2025.126847. [DOI] [PubMed] [Google Scholar]
  • 10.Moderna, Inc. Moderna announces positive Phase 3 results for seasonal influenza vaccine. https://feeds.issuerdirect.com/news-release.html?newsid=4899326521164266&symbol=MRNA Press Release.
  • 11.Weber J.S., Carlino M.S., Khattak A., Meniawy T., Ansstas G., Taylor M.H., Kim K.B., McKean M., Long G.V., Sullivan R.J., et al. Individualised neoantigen therapy mRNA-4157 (V940) plus pembrolizumab versus pembrolizumab monotherapy in resected melanoma (KEYNOTE-942): a randomised, phase 2b study. Lancet. 2024;403:632–644. doi: 10.1016/S0140-6736(23)02268-7. [DOI] [PubMed] [Google Scholar]
  • 12.Rojas L.A., Sethna Z., Soares K.C., Olcese C., Pang N., Patterson E., Lihm J., Ceglia N., Guasp P., Chu A., et al. Personalized RNA neoantigen vaccines stimulate T cells in pancreatic cancer. Nature. 2023;618:144–150. doi: 10.1038/s41586-023-06063-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.World Health Organization Evaluation of the quality, safety and efficacy of messenger RNA vaccines for the prevention of infectious diseases: regulatory considerations. Annex. 2022;3 https://www.who.int/publications/m/item/annex-3-mRNA-vaccines-trs-no-1039 [Google Scholar]
  • 14.World Health Organization WHO guidelines on non-clinical evaluation of vaccines. 2005. https://www.who.int/publications/m/item/nonclinical-evaluation-of-vaccines-annex-1-trs-no-927
  • 15.Li S.D., Huang L. Pharmacokinetics and biodistribution of nanoparticles. Mol. Pharm. 2008;5:496–504. doi: 10.1021/mp800049w. [DOI] [PubMed] [Google Scholar]
  • 16.European Medicines Agency Assessment Report. mRESVIA. Moderna. 2024. https://www.ema.europa.eu/en/documents/assessment-report/mresvia-epar-public-assessment-report_en.pdf
  • 17.European Medicines Agency . Pfizer; 2021. Assessment Report. COMIRNATY.https://www.ema.europa.eu/en/documents/assessment-report/comirnaty-epar-public-assessment-report_en.pdf [Google Scholar]
  • 18.Therapeutic Goods Administration Nonclinical Summary and Recommendations. SPIKEVAX. Moderna. 2021 https://www.tga.gov.au/sites/default/files/2023-07/FOI%204320_0.pdf [Google Scholar]
  • 19.Pharmaceuticals and Medical Devices Agency . SPIKEVAX. Moderna; 2021. Report on the Deliberation Results.https://www.pmda.go.jp/files/000243267.pdf [Google Scholar]
  • 20.Tenforde M.W., Weber Z.A., Natarajan K., Klein N.P., Kharbanda A.B., Stenehjem E., Embi P.J., Reese S.E., Naleway A.L., Grannis S.J., et al. Early estimates of bivalent mRNA vaccine effectiveness in preventing COVID-19-associated emergency department or urgent care encounters and hospitalizations among immunocompetent adults - VISION Network, Nine States, September-November 2022. MMWR. Morb. Mortal. Wkly. Rep. 2023;71:1637–1646. doi: 10.15585/mmwr.mm7153a1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.DeCuir J., Payne A.B., Self W.H., Rowley E.A.K., Dascomb K., DeSilva M.B., Irving S.A., Grannis S.J., Ong T.C., Klein N.P., et al. Interim effectiveness of updated 2023-2024 (monovalent XBB.1.5) COVID-19 vaccines against COVID-19-associated emergency department and urgent care encounters and hospitalization among immunocompetent adults aged ≥18 years - VISION and IVY Networks, September 2023-January 2024. MMWR. Morb. Mortal. Wkly. Rep. 2024;73:180–188. doi: 10.15585/mmwr.mm7308a5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Link-Gelles R., Chickery S., Webber A., Ong T.C., Rowley E.A.K., DeSilva M.B., Dascomb K., Irving S.A., Klein N.P., Grannis S.J., et al. Interim estimates of 2024-2025 COVID-19 vaccine effectiveness among adults aged ≥18 years - VISION and IVY Networks, September 2024-January 2025. MMWR. Morb. Mortal. Wkly. Rep. 2025;74:73–82. doi: 10.15585/mmwr.mm7406a1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Surie D., DeCuir J., Zhu Y., Gaglani M., Ginde A.A., Douin D.J., Talbot H.K., Casey J.D., Mohr N.M., Zepeski A., et al. Early estimates of bivalent mRNA vaccine effectiveness in preventing COVID-19-associated hospitalization among immunocompetent adults aged ≥65 years - IVY Network, 18 States, September 8-November 30, 2022. MMWR. Morb. Mortal. Wkly. Rep. 2022;71:1625–1630. doi: 10.15585/mmwr.mm715152e2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Segal L., Wouters S., Morelle D., Gautier G., Le Gal J., Martin T., Kuper F., Destexhe E., Didierlaurent A.M., Garçon N. Non-clinical safety and biodistribution of AS03-adjuvanted inactivated pandemic influenza vaccines. J. Appl. Toxicol. 2015;35:1564–1576. doi: 10.1002/jat.3130. [DOI] [PubMed] [Google Scholar]
  • 25.Zhang W., Cui H., Xu J., Shi M., Bian L., Cui L., Jiang C., Zhang Y. Biodistribution and mechanisms of action of MF59 and MF59-like adjuvants. J. Contr. Release. 2025;378:573–587. doi: 10.1016/j.jconrel.2024.12.044. [DOI] [PubMed] [Google Scholar]
  • 26.Gómez-Mantilla J.D., Trocóniz I.F., Garrido M.J. In: ADME Translational Pharmacokinetics/pharmacodynamics of Therapeutic Proteins: Applications in Drug Discovery and Development. Zhou H., editor. John Wiley and Sons, Inc; Hoboken, NJ: 2016. ADME process in vaccines and PK/PD approaches for vaccination optimization; pp. 1–22. [Google Scholar]
  • 27.Marquez-Martinez S., Salisch N., Serroyen J., Zahn R., Khan S. Peak transgene expression after intramuscular immunization of mice with adenovirus 26-based vector vaccines correlates with transgene-specific adaptive immune responses. PLoS One. 2024;19 doi: 10.1371/journal.pone.0299215. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Zhang L., Seow B.Y.L., Bae K.H., Zhang Y., Liao K.C., Wan Y., Yang Y.Y. Role of PEGylated lipid in lipid nanoparticle formulation for in vitro and in vivo delivery of mRNA vaccines. J. Contr. Release. 2025;380:108–124. doi: 10.1016/j.jconrel.2025.01.071. [DOI] [PubMed] [Google Scholar]
  • 29.Kong W., Wei Y., Dong Z., Liu W., Zhao J., Huang Y., Yang J., Wu W., He H., Qi J. Role of size, surface charge, and PEGylated lipids of lipid nanoparticles (LNPs) on intramuscular delivery of mRNA. J. Nanobiotechnol. 2024;22:553. doi: 10.1186/s12951-024-02812-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Ministry of Health, LaW Report on the Deliberation Results - Cominarty Intramuscular Injection - Pfizer Japan Inc. 2021. https://www.pmda.go.jp/files/000243206.pdf
  • 31.Reddy S.T., Rehor A., Schmoekel H.G., Hubbell J.A., Swartz M.A. In vivo targeting of dendritic cells in lymph nodes with poly(propylene sulfide) nanoparticles. J. Contr. Release. 2006;112:26–34. doi: 10.1016/j.jconrel.2006.01.006. [DOI] [PubMed] [Google Scholar]
  • 32.McCright J., Naiknavare R., Yarmovsky J., Maisel K. Targeting lymphatics for nanoparticle drug delivery. Front. Pharmacol. 2022;13 doi: 10.3389/fphar.2022.887402. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Sebastiani F., Yanez Arteta M., Lerche M., Porcar L., Lang C., Bragg R.A., Elmore C.S., Krishnamurthy V.R., Russell R.A., Darwish T., et al. Apolipoprotein E binding drives structural and compositional rearrangement of mRNA-containing lipid nanoparticles. ACS Nano. 2021;15:6709–6722. doi: 10.1021/acsnano.0c10064. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Chen J., Ye Z., Huang C., Qiu M., Song D., Li Y., Xu Q. Lipid nanoparticle-mediated lymph node-targeting delivery of mRNA cancer vaccine elicits robust CD8+ T cell response. Proc. Natl. Acad. Sci. USA. 2022;119 doi: 10.1073/pnas.2207841119. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Hassett K.J., Rajlic I.L., Bahl K., White R., Cowens K., Jacquinet E., Burke K.E. mRNA vaccine trafficking and resulting protein expression after intramuscular administration. Mol. Ther. Nucleic Acids. 2024;35 doi: 10.1016/j.omtn.2023.102083. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Ci L., Hard M., Zhang H., Gandham S., Hua S., Wickwire J., Wehrman T., Slauter R., Auerbach A., Kenney M., et al. Biodistribution of Lipid 5, mRNA, and its translated protein following intravenous administration of mRNA-encapsulated lipid nanoparticles in rats. Drug Metab. Dispos. 2023;51:813–823. doi: 10.1124/dmd.122.000980. [DOI] [PubMed] [Google Scholar]
  • 37.Garneau N.L., Wilusz J., Wilusz C.J. The highways and byways of mRNA decay. Nat. Rev. Mol. Cell Biol. 2007;8:113–126. doi: 10.1038/nrm2104. [DOI] [PubMed] [Google Scholar]
  • 38.Wu D., Chen Q., Chen X., Han F., Chen Z., Wang Y. The blood–brain barrier: Structure, regulation and drug delivery. Signal Transduct. Targeted Ther. 2023;8:217. doi: 10.1038/s41392-023-01481-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Shilo M., Sharon A., Baranes K., Motiei M., Lellouche J.P.M., Popovtzer R. The effect of nanoparticle size on the probability to cross the blood-brain barrier: an in-vitro endothelial cell model. J Nanobiotechnol. 2015;13:19. doi: 10.1186/s12951-015-0075-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Betzer O., Shilo M., Opochinsky R., Barnoy E., Motiei M., Okun E., Yadid G., Popovtzer R. The effect of nanoparticle size on the ability to cross the blood-brain barrier: an in vivo study. Nanomedicine (London) 2017;12:1533–1546. doi: 10.2217/nnm-2017-0022. [DOI] [PubMed] [Google Scholar]
  • 41.Sasson E., Anzi S., Bell B., Yakovian O., Zorsky M., Deutsch U., Engelhardt B., Sherman E., Vatine G., Dzikowski R., Ben-Zvi A. Nano-scale architecture of blood-brain barrier tight-junctions. Elife. 2021;10 doi: 10.7554/eLife.63253. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Chander N., Basha G., Yan Cheng M.H., Witzigmann D., Cullis P.R. Lipid nanoparticle mRNA systems containing high levels of sphingomyelin engender higher protein expression in hepatic and extra-hepatic tissues. Mol. Ther. Methods Clin. Dev. 2023;30:235–245. doi: 10.1016/j.omtm.2023.06.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Hou X., Zaks T., Langer R., Dong Y. Lipid nanoparticles for mRNA delivery. Nat. Rev. Mater. 2021;6:1078–1094. doi: 10.1038/s41578-021-00358-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Bhattacharjee B., Lu P., Monteiro V.S., Tabachnikova A., Wang K., Hooper W.B., Bastos V., Greene K., Sawano M., Guirgis C., et al. Immunological and antigenic signatures associated with chronic illnesses after COVID-19 vaccination. medRxiv. 2025 doi: 10.1101/2025.02.18.25322379. Preprint at. [DOI] [Google Scholar]
  • 45.Patel A.A., Zhang Y., Fullerton J.N., Boelen L., Rongvaux A., Maini A.A., Bigley V., Flavell R.A., Gilroy D.W., Asquith B., et al. The fate and lifespan of human monocyte subsets in steady state and systemic inflammation. J. Exp. Med. 2017;214:1913–1923. doi: 10.1084/jem.20170355. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Yang C., Zhao H., Espín E., Tebbutt S.J. Association of SARS-CoV-2 infection and persistence with long COVID. Lancet. Respir. Med. 2023;11:504–506. doi: 10.1016/S2213-2600(23)00142-X. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Ogata A.F., Cheng C.A., Desjardins M., Senussi Y., Sherman A.C., Powell M., Novack L., Von S., Li X., Baden L.R., Walt D.R. Circulating severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) vaccine antigen detected in the plasma of mRNA-1273 vaccine recipients. Clin. Infect. Dis. 2022;74:715–718. doi: 10.1093/cid/ciab465. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Kent S.J., Li S., Amarasena T.H., Reynaldi A., Lee W.S., Leeming M.G., O'Connor D.H., Nguyen J., Kent H.E., Caruso F., et al. Blood distribution of SARS-CoV-2 lipid nanoparticle mRNA vaccine in humans. ACS Nano. 2024;18:27077–27089. doi: 10.1021/acsnano.4c11652. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Boxenbaum H. Interspecies scaling, allometry, physiological time, and the ground plan of pharmacokinetics. J. Pharmacokinet. Biopharm. 1982;10:201–227. doi: 10.1007/BF01062336. [DOI] [PubMed] [Google Scholar]
  • 50.Huh Y., Smith D.E., Feng M.R. Interspecies scaling and prediction of human clearance: comparison of small- and macro-molecule drugs. Xenobiotica. 2011;41:972–987. doi: 10.3109/00498254.2011.598582. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Huang Q., Riviere J.E. The application of allometric scaling principles to predict pharmacokinetic parameters across species. Expert Opin. Drug Metab. Toxicol. 2014;10:1241–1253. doi: 10.1517/17425255.2014.934671. [DOI] [PubMed] [Google Scholar]
  • 52.Kowalski K.G. An algorithm for estimating the terminal half-life in pharmacokinetic studies. Comput. Methods Progr. Biomed. 1994;42:119–126. doi: 10.1016/0169-2607(94)90048-5. [DOI] [PubMed] [Google Scholar]
  • 53.US Food and Drug Administration Roadmap to reducing animal testing in preclinical safety studies. 2025. https://www.fda.gov/files/newsroom/published/roadmap_to_reducing_animal_testing_in_preclinical_safety_studies.pdf
  • 54.US Food and Drug Administration Inactive Ingredient Search for Approved Drug Products. https://www.fda.gov/drugs/drug-approvals-and-databases/inactive-ingredients-database-download
  • 55.International Council for Harmonisation Nonclinical biodistribution considerations for gene therapy products S12. 2023. https://database.ich.org/sites/default/files/ICH_S12_Step4_Guideline_2023_0314.pdf

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Document S1. Figures S1–S3 and Tables S1–S5
mmc1.pdf (252.3KB, pdf)
Table S6. Example data demonstrating low level distal-tissue detections in brain and heart across the three biodistribution datasets: number of quantifiable samples and mean quantifiable concentrations by time point
mmc2.xlsx (14.2KB, xlsx)
Document S2. Article plus supplemental information
mmc3.pdf (2.2MB, pdf)

Data Availability Statement

Supporting data may be made available from the corresponding author upon reasonable request


Articles from Molecular Therapy. Nucleic Acids are provided here courtesy of The American Society of Gene & Cell Therapy

RESOURCES