Escrever um comentário

Summary

The preprint “Tracking Inflammation in Real Time Following Vaccination: Validation of a Novel Individualized Digital Inflammatory Biomarker Relative to Serum Biomarkers” by Darpit Dave et al. presents an innovative approach to monitoring inflammation using wearable technology. The study introduces the inflammatory multivariate change index (iMCI), a digital biomarker calculated from continuous physiological data, including heart rate, skin temperature, respiration rate, and heart rate variability. Participants wore a sensor patch for 14 days surrounding vaccination, and the resulting physiologic signals were compared against traditional serum biomarkers such as C-reactive protein (CRP) and interferon-gamma (IFN-γ). The authors report strong correlations between iMCI and these established inflammatory markers, suggesting that digital, real-time monitoring may be a viable alternative or complement to blood-based assessments. Overall, the paper is well written, logically organized, and effectively frames its contribution within both personalized medicine and digital health research.

The authors also successfully contextualize the clinical importance of inflammation tracking, explaining how continuous physiological signals may detect subtle immune responses that traditional sampling might miss due to sparse blood-draw timing. The incorporation of similarity-based modeling (SBM) to construct individualized baselines, essentially digital physiological “twins”, is a compelling methodological feature. This personalized framework allows for the detection of deviations that are meaningful at the individual level, rather than relying on group-based averages. The study is presented as a proof-of-concept, and the writing appropriately balances innovation with scientific caution.

Suggested Revisions

Although the paper is strong overall, there are areas where the authors could improve clarity and reproducibility. First, the computational modeling methods are under-described. Because the iMCI calculation is central to the paper, the authors should provide more detail regarding how raw sensor data were processed, how features were selected or weighted, and whether the model underwent cross-validation or independent testing. Additional information about noise handling, missing data interpolation, and stability of the personalized baseline would help readers evaluate the methodological rigor. Without this, it is difficult to assess potential model bias or overfitting, especially since the dataset is relatively small.

Another area that would benefit from expansion is the relationship between physiological biomarkers and subjective symptom reporting. The authors note that correlations between iMCI and serum markers were strong, but correlations with self-reported reactogenicity were only moderate. This discrepancy deserves more discussion. For instance, individuals may not perceive subtle inflammation, or subjective symptoms may be delayed relative to physiological changes. Exploring these alternative explanations would strengthen the discussion and provide a more nuanced interpretation of the findings. It may also help clarify the potential clinical advantages of continuous physiological monitoring over self-reported symptoms.

Additionally, while the authors address limitations such as sample size and timing of blood draws, the discussion could be enhanced by explaining how these factors specifically influence the observed correlations. Clarifying the degree to which timing mismatches might under- or over-estimate associations would improve the reader’s understanding and increase transparency. A brief sensitivity analysis or hypothetical scenario comparing ideal versus actual sampling windows would further support their conclusions and demonstrate awareness of methodological constraints.

Overall Recommendation

Overall, this preprint represents a promising and forward-thinking contribution to the field of digital health and immunological monitoring. The use of wearable sensor systems to quantify inflammation in real time addresses a significant gap in current clinical practice, where immune responses are typically inferred from infrequent blood draws. The study presents compelling early evidence that multivariate physiological data can capture inflammatory dynamics comparable to established serum biomarkers. The manuscript is clear, well-organized, and scientifically cautious, making it a strong proof-of-concept.

However, the paper would benefit from greater methodological transparency, especially regarding the computational modeling pipeline and the interpretation of differences between physiological and subjective responses. These additions would not change the core findings but would meaningfully strengthen the credibility, reproducibility, and practical interpretability of the research. With these minor revisions, I would recommend the manuscript for publication, as it lays a solid foundation for future work in continuous health monitoring, digital biomarker development, and precision medicine.

Competing interests

The author declares that they have no competing interests.

Use of Artificial Intelligence (AI)

The author declares that they used generative AI to come up with new ideas for their review.