Spatial transcriptomics analysis to predict response to immune checkpoint blockade (ICB) in recurrent or metastatic head and neck squamous cell cancer (RM-HNSCC).
Bibliographic record
Abstract
6045 Background: Spatial transcriptomics (ST) revealed conserved malignant leading edge (LE) and tumor core (TC) architectures in primary oral squamous cell carcinoma (OSCC) with potential for biomarker discovery. Spatial organization of tumor cells, as well as composition and prognostic significance of neighboring stromal cells in RM-HNSCC remain unknown. Methods: 21 tumor biopsy samples (14 baseline, 7 paired on-treatment) from 14 ICB-naive RM-HNSCC patients (pts) treated with pembrolizumab in INSPIRE (NCT02644369) were profiled using 10x Visium. Spatial organization was refined by scoring LE and TC gene sets identified in OSCC (Arora and Bose et al. Nat Comm 2023). Malignant (2,671 spots) and nonmalignant (8,177 spots) subclusters were annotated, with the latter classified into five cell subtypes using canonical markers: tumor-associated macrophages (TAMs) ( CD68 , CD14 , SCF1R ), regulatory stromal cells (reg) ( KRT17 , COL10A1 , SRBP1 ), plasma cells ( CD38 , IRF4 , PRDM1 ), T cells ( CD3D , CD3E , PTPRC ), and cancer-associated fibroblasts (CAFs) ( FAP , COL1A1 , PDGFRB ). Neighborhood analyses compared normalized counts of stromal cells adjacent to LE and TC, accounting for variations in cell density and sampling differences. A signature was built through k -means clustering of the five cell subtypes. Pts were stratified into high/low signature-score groups using the median cutoff and tested for association with progression-free survival (PFS). Results: Spatial organization revealed conserved malignant subclusters (C0 and C1) in 19/21 samples from 13 pts (11 non-responders). Top C0 genes were COL21A1 , S1PR3 , LIFR , and ZEB1 . Top C1 genes were KRT6B , KRT6C , KRTDAP , and LCN2 . Pathway analysis predicted activation of cell cycle and glycoprotein 6 in C0, and keratinization and neutrophil degranulation in C1. Comparative expression of OSCC-related gene sets revealed LE correlation with C0, and TC correlation with C1 (both p < 0.0001); stronger overlap was seen with the latter highlighting TC as a more conserved feature in HNSCC. Among non-responders, dominant communication patterns in LE and TC included claudins, cadherins, WNT, and IL-6, linked to cell adhesion, migration/invasion, and immune evasion. Signature generated by neighborhood analysis was enriched in TAMs and T cells, depleted in CAFs and reg near LE and TC, while plasma cells were depleted near LE but enriched near TC. Pts with high signature scores (6/13) exhibited improved PFS compared to low scores (7/13), median PFS 6.0 months [95% CI, 2.3-NA] versus 1.9 months [95% CI, 1.8-NA] (p = 0.059). Conclusions: To our knowledge, this is the first report using ST analysis to characterize LE and TC architectures in RM-HNSCC, along with heterogeneous neighboring stromal cells with prognostic potential for ICB. Ongoing cohort expansion will elucidate the clinical significance of these findings.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".