⚠️ Auto-extracted by Claude on 2026-06-09 — not verified against the full PDF (DOI 10.1038/s41586-026-10243-x; closed-access, no local full text). Effect sizes below are drawn from the abstract and reputable secondary coverage (ASCO Post); exact hazard ratios were not available and are described qualitatively. Verify against the primary source before relying on quantitative claims. gap/no-fulltext-access
Thymic health and immunotherapy outcomes in patients with cancer
TL;DR
Companion clinical-oncology paper to Bernatz 2026 (adults). The same deep-learning thymic health score (see thymic-health-score) was applied to routine pre-treatment CT in a pan-cancer cohort of n ≈ 3,476 patients receiving immune-checkpoint inhibitors (ICB). In non-small-cell lung cancer (NSCLC), higher thymic health was associated with reduced disease progression and lower all-cause mortality on immunotherapy — and these associations persisted within high-PD-L1 and high-tumor-mutational-burden (TMB) subgroups, i.e. thymic health added prognostic information beyond the two established ICB response biomarkers. Mechanistic anchoring in the TRACERx lung-cancer study linked thymic health to T-cell-receptor (TCR) repertoire diversity, T-cell-receptor excision circles (TRECs), and immune-signaling pathway activation — tying the imaging score to genuine adaptive-immune output.
Design
- Method: same deep-learning thymic-health scoring on routine CT as the adults paper.
- Cohort: pan-cancer, n ≈ 3,476 patients treated with immune-checkpoint inhibitors (PD-1/PD-L1 ± CTLA-4 blockade). Cancer types analyzed included NSCLC, melanoma, breast, and renal cancer; the strongest and most consistent signal was in NSCLC.
- Mechanistic validation: TRACERx lung-cancer cohort used to correlate the imaging score against molecular adaptive-immune readouts (TCR diversity, TRECs, immune-pathway transcriptional activation).
- Retrospective/observational; no randomization of thymic health.
Key findings (qualitative — exact HRs pending full-text verification)
- NSCLC on ICB: higher thymic health → lower risk of progression and lower all-cause mortality. gap/no-fulltext-access (hazard ratios / CIs not retrieved)
- Independent of established biomarkers: the association held within high-PD-L1 and high-TMB strata, indicating thymic health is not merely a proxy for tumor-side immunogenicity but reflects host adaptive-immune competence — a complementary, host-side axis.
- Biological grounding (TRACERx): thymic health correlated positively with TCR repertoire diversity, TRECs (a direct marker of recent thymic emigrant / naive-T-cell output), and immune-signaling pathway activation — evidence the CT score indexes real thymic function, not just anatomy.
Why it matters for this wiki
- Host immune fitness as an ICB response biomarker. Most clinical immunotherapy biomarkers are tumor-intrinsic (PD-L1 expression, TMB, MMR status). This paper adds a host-side, aging-driven predictor — the patient’s residual thymic output — and shows it is orthogonal to tumor-side markers. This is directly relevant to the disabled-adaptive-immunity “target immunogenicity” decision framework: a rejuvenated/intact adaptive system improves the payoff of surveillance-augmenting therapy.
- Connects immunosenescence to a clinical decision point. It operationalizes the long-standing hypothesis that immune aging blunts cancer immunotherapy, using a scalable imaging readout.
- Validates the thymic-health-score against molecular ground truth (TRECs, TCR diversity) — strengthening its claim to be a real immune-aging biomarker rather than an incidental radiological correlate.
Extrapolation / evidence quality
| Dimension | Status |
|---|---|
| Human evidence? | yes — human cancer cohort (n ≈ 3,476) + TRACERx molecular validation |
| Independent replication? | no — single group, single method, 2026; needs external cohort replication gap/needs-replication |
| Causal (vs. associative)? | no — observational; does not show that improving thymic health improves ICB outcomes gap/no-mechanism |
| Generalizes beyond NSCLC? | unclear — strongest in NSCLC; melanoma/breast/renal signal weaker or unreported gap/contradictory-evidence |
See also
- bernatz-2026-thymic-health-adults — companion paper: thymic health and mortality in asymptomatic adults
- thymic-health-score — the imaging biomarker
- disabled-adaptive-immunity — § Decision framework: target immunogenicity (host-side competence axis)
- thymus · immunosenescence