Dose-normalized vaccine delivery to draining lymph nodes¶
What fraction of a vaccine dose reaches a draining node? The answer depends on what was labeled, where it was injected, which node was collected, when it was sampled, and what the denominator means. This reference list collects primary experiments near the vaccine-window question. It is a scale check for explicitly assumed scenarios, not a pooled estimate for an IM SLP core or a default transport-rate distribution.
| Study | Route and material | Node measurement | Why it is not a core-to-APC fraction |
|---|---|---|---|
| Wang et al., 2018 | Mouse tail-base SC; fluorescent iTEP polypeptide or albumin-binding ABD-iTEP | At 6 h, inguinal plus axillary nodes contained 0.36 ± 0.02% and 1.31 ± 0.11% of injected label, respectively (n = 3). | Whole-node carrier label is not intact vaccine core. |
| Roces et al., 2019 | Mouse quadriceps IM; 5 µg radiolabeled H56 protein adsorbed to DDAB:TDB liposomes | Popliteal-node antigen label was <0.3% of initial vaccine dose across four conventional formulations, sampled at 6–192 h. | Radioiodine tracks antigen-associated label, not a sequenced epitope in APCs. The result is formulation- and node-specific. |
| Zhu et al., 2017 | Mouse SC; radiolabeled DNA–RNA nanocapsules carrying peptide neoantigen | At 24 h, 0.13% injected dose was in inguinal and 0.11% in axillary nodes. | The quantitative label is on the carrier. Peptide and carrier colocalization does not yield peptide copies per APC. |
| Vicente et al., 2014 | Rabbit SC; indium-111-labeled polyglucosamine/squalene nanocapsules | Popliteal node had 3.8 ± 1.2% injected dose at 48 h by gamma scintigraphy. | This is radiolabeled adjuvant carrier, not antigen or core peptide; some label dissociated in serum. |
| Glycosylated OVA study, 2021 | Mouse hock SC; fluorescent OVA versus OVA–p(GluNAc), 5 µg antigen/hock | Calibrated whole-node imaging reports percent starting antigen dose/hock; conjugate gave 17-fold more node signal at its 8 h maximum and 10-fold greater node-signal AUC. | The fold changes compare formulations, not an IM dose fraction or intact-core/APC amount. |
| Backlund et al., 2022 | Mouse tail-base SC; fluorescent gp100 peptide versus cell-penetrating-peptide conjugates | At 48 h, inguinal-node fluorescence was 3.1–5.8-fold higher for most conjugates than free peptide. | Relative fluorescence has no injected-dose denominator; label-positive APCs are a cell fraction, not dose recovered in APCs. |
| Hassett et al., 2024 | Macaque IM; reporter mRNA–LNP, 300 µg RNA/site | Reporter RNA in draining node measured 155 ng/g at the first 8 h tissue sample; a separate cohort had 9.1% EGFP-positive cDCs at 24 h. | Tissue concentration and positive-cell fraction have different denominators and cohorts; neither is peptide dose in APCs. |
Wang and Roces are the closest antigen-associated, injected-dose
denominator checks here. Wang used subcutaneous carrier polypeptides rather
than a free SLP; Roces used an intramuscular liposomal protein formulation.
Both report material present in a node at a sample time. Our
node_apc_fraction_of_dosed_peptide instead counts cumulative first arrival
of core-bearing lineages inside nodal APCs by the chosen horizon. A node
snapshot may include extracellular material, may miss material that passed
through earlier, and cannot tell whether a label still contains the target
sequence. No numerical conversion between these endpoints is justified.
Related denominators that cannot be read as percent of dose¶
A subcutaneous mouse mRNA–nanoparticle study
reported a node peak of about 170% injected dose per gram at 8 h for
tritium-labeled RNA. This is a concentration normalized by tissue mass
(%ID/g); it is not 170% of the dose in one node. The paper also found
fewer than 5% nanoparticle-positive node cells, with up to 50% of those
positive cells classed as CD11c⁺/CD123⁺ DCs. Neither cell percentage supplies
RNA copies, translated peptide copies or a core-bearing APC fraction.
Roces et al. also tested an engineered biotinylated liposome with an avidin predose: node lipid-label recovery reached roughly 30% in inguinal and 40% in popliteal nodes. The deliberate trapping intervention and lipid analyte make those values a separate formulation experiment, not a generic IM protein baseline. It illustrates how sharply formulation can change node recovery.
What to measure next¶
To calibrate a window-specific probability, measure a known IM dose of the actual SLP and timed, sequence-confirmed intact core-bearing fragments in injection-site fluid, lymph, whole node, sorted APCs and blood. Report both percent injected dose in each compartment and absolute copies per APC subset; separately measure surface pMHC and cell counts. For mRNA, measure active RNA, produced antigen, direct APC expression, extracellular antigen transfer and preformed-pMHC transfer. Matched time courses are needed to separate production from loss. The remaining gap is tracked in calibration issue #549 and the calibration audit.