Ageing promotes metastasis via activation of the integrated stress response
Summary
This Nature study uses aged versus young KRAS/TP53 (KP) genetically engineered mice and matched primary tumour cultures to show that physiological ageing shifts lung cancer behaviour: older hosts develop smaller, less proliferative primary tumours but show earlier and more frequent metastasis. The authors identify sustained activation of the integrated stress response (ISR), centred on PERK–eIF2α–ATF4, driven by age-related chromatin changes at the Atf4 locus. ATF4 is necessary and sufficient for the ageing-associated metastatic phenotype and reprogrammes metabolism towards greater glutamine-dependent anaplerosis. Importantly, that metabolic rewiring creates a druggable vulnerability—glutaminase inhibition (CB-839) blocks metastasis in the aged tumour context without shrinking primary tumours. Human NSCLC cohorts (western Sweden, TCGA and others) recapitulate key ageing patterns: older KRAS-mutant patients present with more advanced-stage disease despite smaller primary tumours, and high ATF4 associates with poorer survival in advanced disease.
Key Points
- Ageing in KP mice reduces primary lung tumour burden but increases early incidence of local and distant metastasis.
- Primary cultures from old mice (KP‑O) show enhanced EMT markers, anoikis resistance, larger spheroids and invasive behaviour in 3D assays.
- ATAC‑seq and RNA‑seq reveal open chromatin at Atf4 and enrichment of UPR/ISR and EMT programmes in KP‑O tumours.
- PERK–eIF2α–ATF4 signalling is amplified in aged tumour cells; ATF4 protein and target genes remain elevated and are transcriptionally driven.
- Genetic or pharmacological suppression of ATF4 (CRISPR/shRNA or ISRIB) reduces anoikis resistance and metastatic seeding in KP‑O cells; ATF4 overexpression drives metastasis in young cells and human A549 cells.
- Metabolically, ATF4 increases glycolytic flux and glutamine‑derived anaplerosis; KP‑O cells are selectively sensitive to glutaminase inhibitors (CB‑839) and glutamine depletion.
- In vivo CB‑839 nearly abolishes distant metastasis from KP‑O tumours while not markedly affecting primary tumour size—implying a metastasis‑specific vulnerability.
- Clinical datasets mirror the mouse findings: older KRAS‑mutant NSCLC patients (65–75 years) are likelier to have advanced disease, smaller primary tumours, higher ATF4 levels and worse survival when ATF4 is high.
Content summary
The authors induced lung tumours simultaneously in young (2–3 months) and old (18–19 months) KP mice. Although KP‑Old animals developed fewer and smaller primary tumours, they showed earlier metastatic spread and reduced survival. Primary tumour lines established from old mice were more metastatic in multiple transplantation assays and displayed EMT signatures and anoikis resistance in vitro.
Multi-omic profiling (RNA‑seq, ATAC‑seq) pointed to two dominant, overlapping programmes in KP‑O cells: EMT and the unfolded protein response/integrated stress response. Chromatin at the Atf4 locus was more open in KP‑O cells; stress (low glucose) triggered stronger and more sustained PERK–eIF2α–ATF4 activation. ATF4 levels were higher in old tumours and directly correlated with metastatic behaviour.
Functional studies showed ATF4 is necessary and sufficient for the ageing-associated metastatic state. Pharmacological ISR inhibition (ISRIB) or genetic ATF4 depletion reduced anoikis resistance and metastatic colonisation. Metabolic tracing demonstrated ATF4-driven increases in glycolysis and glutamine anaplerosis; KP‑O cells depended on glutaminolysis and were sensitive to GLS inhibitors (CB‑839, BPTES) and ASCT2 inhibition. CB‑839 treatment in vivo strongly reduced distant metastasis from KP‑O tumours without shrinking the primary lesion, highlighting a metastasis-specific therapeutic opportunity.
Context and relevance
This paper resolves an apparent paradox: ageing limits primary tumour expansion but increases metastatic risk. By showing that age‑associated epigenetic states prime sustained ISR–ATF4 activation, the work links physiological ageing to tumour cell plasticity, metabolic rewiring and metastatic competence. Clinically, most lung cancer patients are older; the study therefore argues for including biological age in preclinical models and trial design. For researchers and clinicians, ATF4 and glutaminolysis emerge as biomarkers and actionable targets specifically in the aged/metastatic setting of NSCLC—especially KRAS‑mutant cases.
Author take (punchy)
This is a big one. The team proves that ageing actively reprogrammes tumour cells — not just the microenvironment — flipping the switch from primary growth to metastatic spread through an epigenetically primed ISR/ATF4 axis. That switch creates a metabolic Achilles’ heel: glutamine dependence that you can drug. If you work on metastasis, lung cancer or age‑related therapy stratification, the detailed mechanistic and translational data here are essential reading.
Why should I read this?
Quick: it explains why older patients often die from metastasis even when primary tumours are smaller, reveals ATF4 as the ageing‑driven culprit, and shows a concrete way to hit it (glutaminase blockers). Saves you sifting through mouse, omics and clinical data — they did the heavy lifting and found a druggable weakness.
