Fig 2: Tissue expression of ANXA1 protein. ANXA1 protein expression was evaluated using IHC and double IF staining on resected tumors from 9 early-stage SCLC patients. ANXA1 expression was observed in the tumor cells of 2 patients, with scattered stromal cell positivity (A,B). Two additional patients showed ANXA1 expression in the stroma and in non-tumoral cells within tumor nests (C,D). Double IF staining revealed ANXA1 co-expression with CD45 on immune cells (E, asterisks), whereas cancer cells of tumor nests showed positivity only for ANXA1. ANXA1+ tumor cells (F, asterisks) do not colocalize with vimentin+ CAFs in the stroma (F), whereas ANXA1 shows expression on multiple CD68+ TAMs both in the stroma (G) and in tumor nests (G, asterisks in inset). In contrast, some CD68+ TAMs show no co-expression of ANXA1 (G, arrows in inset).

Fig 3: Differential expression and pathway analysis in the context of ANXA1 expression. Differentially expressed genes (DEGs) in ANXA1-high and ANXA1-low tumor subsets were analyzed using the HTG EdgeSeq oncological biomarker panel. n = 230 genes showed significantly increased expression in ANXA1-high tumors, while n = 16 genes were significantly overexpressed in ANXA1-low tumors (adjusted p-value < 0.05; LogFC > 2) (A). Pathway analysis using the Reactome database shows significant enrichment ratios (FDR ≤ 0.05) in blue for n = 3 pathways in ANXA1-high and n = 2 pathways in ANXA1-low tumors (B). Using ImmunDB’s Immunome panel, n = 117 genes were significant in ANXA1-high tumors and none in ANXA1-low tumors (C). Pathway analysis performed as described previously shows the 10 most enriched pathways (FDR ≤ 0.05) in ANXA1-high tumors. The affinity propagation method was used to reduce the redundancy of the gene sets in the enrichment (D).

Fig 4: Experimental protocol and annexin A1 silencing. Groups and experimental protocol (A). Western blot evaluation of ANXA1 relative protein levels of native, shCTRL, and shANXA1 cells (B). One-way ANOVA, Tukey’s post hoc test, n = 6–8; *** p < 0.001. Data are expressed as mean ± SEM.

Fig 5: Predictive characterization of the SCLC TME according to ANXA1 expression. ESTIMATE algorithm revealed significantly higher immune (p < 0.001) (A) and stroma scores (p < 0.001) (B) in ANXA1-high tumors and indicated higher estimate (p < 0.001) (C) and lower purity scores (p < 0.001) (D). EPIC algorithm shows the distribution of the main TME cell types in SCLC tumors derived from bulk RNA-seq data (E). ANXA1-high tumors were shown to harbor significantly more B-cells (p = 0.027), CAFs (p < 0.001), TAMs (p < 0.001), and NK cells (p < 0.001) compared to ANXA1-low tumors (F). Data are displayed on violin charts, where the median (solid line) and upper and lower quartiles (dashed lines) are indicated. Wilcoxon’s rank sum test was performed for pairwise comparisons, where no data showed a normal distribution according to the Shapiro–Wilk test. ns: non-significant, *: p < 0.05, ***: p < 0.001.