diff --git a/inbox/queue/2026-03-30-eurheartj-select-mediation-analysis-esc-2024.md b/inbox/queue/2026-03-30-eurheartj-select-mediation-analysis-esc-2024.md
deleted file mode 100644
index d515166c..00000000
--- a/inbox/queue/2026-03-30-eurheartj-select-mediation-analysis-esc-2024.md
+++ /dev/null
@@ -1,68 +0,0 @@
----
-type: source
-title: "SELECT Mediation Analysis: Semaglutide's Cardiovascular Outcomes Not Explained by Measured Metabolic or Adiposity Parameters — ESC 2024"
-author: "Colhoun, Lincoff et al. (SELECT investigators)"
-url: https://academic.oup.com/eurheartj/article/45/Supplement_1/ehae666.2792/7835656
-date: 2024-09-01
-domain: health
-secondary_domains: []
-format: conference-abstract
-status: enrichment
-priority: medium
-tags: [GLP-1, semaglutide, SELECT-trial, cardiovascular, mediation-analysis, ESC-2024, inflammation, hsCRP, weight-independent]
-processed_by: vida
-processed_date: 2026-03-30
-enrichments_applied: ["semaglutide-cardiovascular-benefit-is-67-percent-independent-of-weight-loss-with-inflammation-as-primary-mediator.md"]
-extraction_model: "anthropic/claude-sonnet-4.5"
----
-
-## Content
-
-**Exploratory mediation analysis** presented at ESC Congress 2024. Published in European Heart Journal Supplement. Authors include Colhoun and Lincoff (Cleveland Clinic).
-
-**Study question:** Which measurable biomarkers and risk factors mediate semaglutide's cardiovascular benefit in SELECT?
-
-**Key findings (percent mediation estimates, all with wide 95% CIs):**
-- Waist circumference: 64.0% (widest CI — uncertain)
-- hsCRP (high-sensitivity C-reactive protein / inflammation marker): 42.1%
-- HbA1c: 29.0%
-- Body weight: **19.5%** (notably lower than waist circumference)
-- Joint mediation of ALL measured factors: **31.4%** (95% CI: -30.1% to 143.6%)
-- Statistically significant improvements in all mediators with semaglutide
-
-**Key conclusion:** "Neither change in body weight nor other measured cardiovascular risk factors fully explain the effect of semaglutide on MACE in SELECT. Substantial unmeasured pleiotropic effects of semaglutide on MACE not mediated through these risk factors remain possible."
-
-**The ~68.6% unexplained fraction** represents pleiotropic benefit not captured by weight, inflammation (hsCRP), glycemic control, or adiposity.
-
-**Note on confidence intervals:** The joint mediation CI (-30.1% to 143.6%) is extremely wide, reflecting the statistical difficulty of mediation analysis in this context. The individual estimates (hsCRP at 42.1%, body weight at 19.5%) are more interpretable as directional signals than precise measurements.
-
-## Agent Notes
-
-**Why this matters:** This is the ESC 2024 active thread from Session 14 that was outstanding. The analysis confirms that body weight accounts for less of the CV benefit than inflammation (hsCRP). The wide CIs limit precision but the directional finding is consistent with the Lancet 2025 prespecified analysis (Deanfield et al.), which confirms weight-independence with stronger study design.
-
-**What surprised me:** Body weight (19.5% mediation) is actually LOWER than hsCRP (42.1%). This means even among measured factors, inflammation is a more important mediator than weight. This inverts the public narrative that GLP-1s reduce CVD risk "because they cause weight loss."
-
-**What I expected but didn't find:** A clear, statistically precise decomposition of the mechanism. The wide CIs on the joint mediation estimate (−30.1% to 143.6%) show how statistically hard this question is to answer with a single trial. The Lancet 2025 prespecified analysis is the stronger evidence.
-
-**KB connections:**
-- Same cluster as the Lancet 2025 source (archive 2026-03-30-lancet-select-adiposity-independent-cv-outcomes-2025.md)
-- [[GLP-1 receptor agonists are the largest therapeutic category launch in pharmaceutical history but their chronic use model makes the net cost impact inflationary through 2035]]
-
-**Extraction hints:** This source should be extracted in conjunction with the Lancet 2025 prespecified analysis — they are complementary pieces of evidence for the same mechanism claim. The ESC 2024 abstract provides the inflammatory (hsCRP) mediator estimate; the Lancet 2025 prespecified analysis provides the weight-independence confirmation. Together they make the mechanism claim extractable. Do not extract as standalone — the wide CIs alone limit it.
-
-**Context:** Exploratory analysis from ESC Congress 2024. Less statistically rigorous than the Lancet 2025 prespecified analysis (Deanfield et al.). Use as supporting evidence, not primary evidence. Lincoff was co-investigator on the broader SELECT trial team.
-
-## Curator Notes (structured handoff for extractor)
-PRIMARY CONNECTION: [[GLP-1 receptor agonists are the largest therapeutic category launch in pharmaceutical history but their chronic use model makes the net cost impact inflationary through 2035]]
-WHY ARCHIVED: Complementary to Lancet 2025 prespecified analysis. Provides specific mediator percentages (hsCRP 42.1%, body weight 19.5%) that the Lancet analysis doesn't separately report.
-EXTRACTION HINT: Extract as SECONDARY EVIDENCE for the mechanism claim. The Lancet 2025 analysis (queue/2026-03-30-lancet-select-adiposity-independent-cv-outcomes-2025.md) is the primary source. Use this for the specific hsCRP/body weight percentage breakdown which the Lancet paper doesn't separately quantify. Wide CIs = flag for confidence calibration.
-
-
-## Key Facts
-- SELECT trial mediation analysis presented at ESC Congress 2024
-- hsCRP mediation estimate: 42.1%
-- Body weight mediation estimate: 19.5%
-- Waist circumference mediation estimate: 64.0% (widest confidence interval)
-- HbA1c mediation estimate: 29.0%
-- Joint mediation of all measured factors: 31.4% (95% CI: -30.1% to 143.6%)
-- All measured mediators showed statistically significant improvements with semaglutide