In brief
- LuCaS+ is a single-cell RNA sequencing atlas of more than 2.5 million nuclei from 454 patients with NSCLC, all with clinical annotation and long-term follow-up.
- The atlas combines a discovery dataset from Berlin and Cologne (364 patients) with an independent validation dataset from Munich (90 patients).
- The authors identified tumour subpopulations with high metastatic potential, associated with oncogenic signalling, immune evasion and epithelial–mesenchymal transition; in early-stage lung adenocarcinoma these cells carried strong prognostic value.
- Combining tumour and microenvironment cell profiles improved survival prediction across all stages.
- The abstract gives no hazard ratios, accuracy measures or cell-population frequencies for these findings.
Study at a glance
- Design: Multicentre single-cell RNA sequencing atlas with clinical annotation
- Cohort: 454 patients with NSCLC, more than 2.5 million nuclei
- Discovery dataset: Berlin and Cologne, 364 patients
- Validation dataset: Munich, 90 patients
- Histologies analysed separately: Lung adenocarcinoma and lung squamous cell carcinoma
- Outcome studied: Metastatic potential and survival, with long-term follow-up
- Not reported: Hazard ratios, measures of predictive accuracy and length of follow-up
More than 2.5 million nuclei from 454 annotated patients
The authors built LuCaS+ to capture the cellular and molecular heterogeneity of NSCLC in patients whose outcomes are known. Each of the 454 patients is clinically annotated and was followed over the long term, so cell states can be set against survival.
The atlas has two parts. The discovery dataset came from Berlin and Cologne (364 patients) and the independent validation dataset from Munich (90 patients). The abstract does not state how many patients had each histology or stage, or which findings were reproduced in the Munich patients.
Tumour cells with high metastatic potential carried prognostic weight in early-stage adenocarcinoma
The authors describe tumour subpopulations with high metastatic potential. These cells were associated with three programmes: oncogenic signalling, immune evasion and epithelial–mesenchymal transition.
In early-stage lung adenocarcinoma, the same cells showed strong prognostic value. The abstract does not say whether this held in squamous cell carcinoma or in advanced disease, so the prognostic claim is confined, for now, to early-stage adenocarcinoma. Because the data are observational, the finding is an association with outcome and does not show that these cells cause the poorer survival.
Distinct tumour, immune and stromal patterns in adenocarcinoma and squamous disease
Adding the profiles of immune and stromal cells to those of the tumour cells improved survival prediction at every stage. Two of the interaction patterns were tied to a histology. In adenocarcinoma the group found inflammation driven by monocytes and fibroblasts. In squamous cell carcinoma it found angiogenic feedback loops mediated by VEGF and TGFβ. A third pattern, immune-permissive niches, was described as promoting antitumour immunity, with no histology attached in the abstract.
“LuCaS+ offers a clinically anchored single-cell resource to enable personalized, biology-driven therapies in NSCLC.”Jurmeister et al., Nature Cancer (2026)
A resource for biomarker work, not yet a clinical test
The abstract reports no testing of the atlas as a decision tool, and it does not compare its survival prediction with stage alone. Prospective validation would be needed before any of these cell states informed adjuvant treatment decisions in early-stage adenocarcinoma.
Sources
- Nature Cancer. A single-cell lung cancer atlas reveals cellular and molecular drivers of metastasis and clinical outcome (2026-11-12). doi.org
Featuring Lung Summit faculty
This study was co-authored by Lung Summit faculty Nikolaj Frost.
This article was produced independently by the Lung Summit editorial team, without industry funding or input.