Follow-up availability and the robustness of immune transcriptomic recovery after myocardial infarction

In 111 ST-segment elevation myocardial infarction patients in GSE59867, the six-month inflammatory-response transcriptomic rank-score change stayed negative (about -0.009 to -0.010) under multiple missing-assay assumptions.

• Why it matters: Later immune transcriptome assays may not reflect the admission-sampled population, so recovery summaries could be biased by selection and by assumptions about unavailable follow-up measurements.
• What they did: The authors reanalyzed GSE59867 with seven Hallmark programs scored on a shared 17,590-gene vocabulary, using paired-completer and marginal contrasts plus cross-fitted availability/outcome models and completion bounds.
• The result: With 83 observed six-month participants, paired and marginal estimates were -0.00893 and -0.00999, and augmented inverse-probability weighting was -0.01004; direction remained negative across prespecified {+/-}2-SD missing-mean scenarios, but unrestricted completion did not.

AI summary of the abstract · may contain errors · not medical advice · read the paper.