This analysis reexamined public transcriptomic datasets from three human peptide-vaccine cohorts to determine molecular correlates of peripheral immunogenicity and clinical outcome. In the MUC1 plus Poly-ICLC PBMC RNA-seq cohort (C1), a summary measure of baseline immune-readiness showed a positive association with ordered anti-MUC1 IgG response classes (beta = 0.301, p = 0.0072, q = 0.093). Permutation testing supported this finding (permutation p = 0.0088). Independently, in the personalized-peptide vaccination cohort tied to castration-resistant prostate cancer survival (C3), higher baseline immune-readiness was associated with longer overall survival (hazard ratio 0.662, 95% CI 0.532–0.823, p = 0.000206, q = 0.00126).
These concordant associations across distinct cohorts indicate that a preexisting, transcriptionally defined immune-competent state in peripheral blood is linked to both measured immunogenicity and longer-term clinical outcome in the analyzed datasets.
A contrasting baseline signal characterized as an erythroid/inflammatory drag displayed associations opposite to immune-readiness. In C1 this module associated negatively with response class (beta = −0.258, p = 0.031, q = 0.202), with permutation testing consistent with the directionality (permutation p = 0.0324). In C3 the same composite signature correlated with inferior overall survival (HR = 1.390, 95% CI 1.181–1.636, p = 0.0000755, q = 0.000982).
Together, these results suggest that elevated erythroid-linked or inflammatory transcriptional programs at baseline may act as a negative correlate of effective vaccine-linked immunity and clinical benefit in these retrospective cohorts.
Analyses of manufactured dendritic-cell vaccine preparations (C2) compared product-state gene expression to TARP ELISpot response. Strong ELISpot responders exhibited lower expression of a tolerogenic/myeloid dendritic-cell product-state signature. At the gene level, 8 of 19 prespecified focused genes reached q < 0.05 in this comparison, indicating a reproducible difference in product-state transcriptional features between stronger and weaker cellular responders.
This finding highlights product-intrinsic transcriptional characteristics of dendritic-cell preparations that correlate with antigen-specific T cell readouts measured by ELISpot in the studied cohort.
At the C1 week-2 timepoint, a composite module characterized as priming/costimulation/mTOR-AKT showed a significant cross-sectional association with antibody response class (permutation p = 0.0026). However, when within-person paired changes were examined between baseline and week 2, the change in this module did not reach significance.
This pattern indicates that while the priming-associated program is linked to response class at an early post-vaccination timepoint, measurable within-subject modulation over that interval was not robust in the available data.
The analysis used three publicly available human peptide-vaccine cohorts: C1/GSE278476 (MUC1 with Poly-ICLC; PBMC RNA-seq with ordered anti-MUC1 IgG classes), C2/GSE85698 (manufactured dendritic-cell vaccine products linked to TARP ELISpot responses), and C3/GSE53922 (baseline PBMC expression associated with overall survival after personalized peptide vaccination in castration-resistant prostate cancer).
Prespecified gene modules were summarized as mean standardized scores. Cohort-appropriate endpoint models were applied at the cohort level, and multiple testing was controlled using within-family false-discovery rate. Permutation testing was reported where applied to support cross-sectional associations. The source reports the statistical outcomes quoted above; no additional methodological details beyond what is stated in the source were introduced here.
Across these retrospective datasets, the author proposes a phase-resolved model in which distinct transcriptional layers contribute to measurable vaccine response phenotypes: (1) a favorable host readiness state at baseline that predicts immunogenicity and survival; (2) an adverse erythroid/inflammatory drag that associates with poorer responses and survival; (3) dendritic-cell product-state transcriptional features that correlate with cellular ELISpot responses; and (4) early post-vaccination priming signals, including a priming/costimulation/mTOR-AKT module, that align with response class cross-sectionally.
The aggregated evidence supports the idea that multiple, temporally linked transcriptomic factors — some host-derived and some product-related — are measurable and associated with outcome metrics in these public peptide-vaccine cohorts.
The source explicitly states limitations: these are retrospective cohort analyses and do not establish causality, biomarker qualification, clinical utility, or evidence of durable tumor control. The reported associations are observational and cohort-specific; external validation, prospective biomarker testing, and mechanistic experiments would be required before translating these signatures into clinical decision tools or interventional strategies.
Readers should note that the source-reported results and effect sizes are limited to the datasets and models analyzed and that no new interventional or clinical recommendations are claimed in the original report.