PO.CL05.10 · 临床研究
全基因组倍增诱导染色体不稳定性，塑造肿瘤-免疫微环境并损害非小细胞肺癌对免疫检查点抑制剂的反应
Whole-genome doubling induces chromosomal instability to shape the tumor-immune microenvironment and impair the response to immune checkpoint inhibitors in non-small cell lung cancer
作者与单位 Authors & Affiliations
摘要 Abstract
中文摘要
引言：全基因组倍增（WGD）见于30-40%的癌症，可驱动染色体不稳定性（CIN），但其在调节NSCLC免疫格局中的作用仍不清楚。我们研究了WGD如何塑造肿瘤内在和外在特征以影响免疫检查点抑制剂（ICI）的反应。
方法：我们分析了520份基线NSCLC肿瘤组织（WES，n=520；WTS，n=500）。WGD根据WES衍生的拷贝数谱定义为超过50%的常染色体基因组具有主要拷贝数≥2。在一个子集中（n=304）通过对全切片图像（WSI）的AI驱动空间分析对肿瘤浸润淋巴细胞（TIL）进行定量。统计比较采用Fisher精确检验和Wilcoxon秩和检验。
结果：在520例肿瘤中，202例（38.8%）为WGD+，318例（61.2%）为WGD-。TP53突变在WGD+肿瘤中显著富集（57.4% vs WGD-：40.3%，p=1.51×10⁻⁴）。而17p13.1（TP53）位点在WGD+队列中显著更高（76.7% vs WGD-：39.9%，p=8.82×10⁻¹⁷）。此外，这种TP53突变与LOH的耦合在WGD+中尤为突出：90.3%（93/103例WGD前突变体）获得LOH，而WGD-肿瘤为64.8%（83/128）。WGD+肿瘤还表现出显著更高的HLA I类LOH负荷（≥2个位点者占16.1% vs 6.3%，p=8.42×10⁻⁴）。WGD+肿瘤还显示出显著更低的T细胞浸润。使用倍性校正的T细胞比例（n=520），WGD+组的中位T细胞比例显著低于WGD-组（0.138 vs 0.184，p=4.28×10⁻⁴）。基因集富集分析（GSEA）显示WGD+肿瘤富集代谢/增殖程序。相比之下，WGD-肿瘤上调免疫激活通路。WGD+肿瘤中这种"免疫冷"特征通过显著更低的免疫评分得到证实；中位细胞溶解活性评分为2.53 vs 3.08（p=5.34×10⁻⁵），中位TLS评分为5.40 vs 6.51（p=1.05×10⁻¹²）。在空间上，AI驱动的WSI分析证实了这一点，将WGD+肿瘤与更少的炎症型和更多的免疫荒漠型表型相关联（p=2.74×10⁻³）。临床上，WGD状态与较差的结局显著相关。与非WGD患者相比，WGD患者表现出更差的总生存期（HR 1.284，95% CI 1.050-1.569，p=1.47×10⁻²）和无进展生存期（HR 1.286，95% CI 1.057-1.563，p=1.18×10⁻²）。
结论：在NSCLC中，WGD通过TP53突变和关键的HLA I类LOH促进高基因组不稳定性，从而实现免疫逃逸。这种基因组格局塑造了一个以代谢重编程、较低T细胞浸润和免疫荒漠型表型为特征的免疫抑制生态系统。因此，WGD定义了一种免疫难治性微环境，并可作为ICI反应不良的稳健预测生物标志物。
查看英文原文 English abstract
Introduction: Whole genome doubling (WGD), observed in 30-40% of cancers, drives chromosomal instability (CIN), but its role in modulating the immune landscape in NSCLC remains unclear. We investigated how WGD shapes tumor-intrinsic and -extrinsic features to influence immune checkpoint inhibitors (ICI) response.
Methods: We analyzed 520 baseline NSCLC tumor tissues (WES, n=520; WTS, n=500). WGD was defined from WES-derived copy number profiles as >50% of the autosomal genome having a major copy number ≥2. Tumor-infiltrating lymphocytes (TILs) were quantified in a subset (n=304) via AI-powered spatial analysis of whole slide images (WSIs). Statistical comparisons utilized Fisher's exact and Wilcoxon rank-sum tests.
Results: Among 520 tumors, 202 (38.8%) were WGD+ and 318 (61.2%) were WGD-. TP53 mutations were significantly enriched in WGD+ tumors (57.4% vs WGD-: 40.3%, p=1.51×10⁻⁴). While the 17p13.1 (TP53) locus was significantly higher in WGD+ cohort (76.7% vs WGD-:39.9%, p=8.82×10⁻¹⁷). Furthermore, this coupling of TP53 mutation and LOH coupling was particularly prominent in WGD+: 90.3% (93/103 pre-WGD mutants) acquired LOH, vs 64.8% (83/128) in WGD- tumors. WGD+ tumors also exhibited a significantly higher burden of HLA Class I LOH burden (≥2 loci in 16.1% vs 6.3%, p=8.42×10⁻⁴). WGD+ tumors also showed significantly lower T cell infiltration. Using ploidy-corrected T cell fractions (n=520), the median T cell fraction was significantly lower in the WGD+ group than in the WGD- group (0.138 vs 0.184, p=4.28×10⁻⁴). Gene set enrichment analysis (GSEA) showed WGD+ tumors enriched metabolic/proliferative programs. In contrast, WGD- tumors upregulated immune-activation pathways. This "immune-cold" profile in WGD+ tumors was confirmed by significantly lower immune scores; median cytolytic activity score was 2.53 vs. 3.08 (p=5.34×10⁻⁵) and median TLS score was 5.40 vs. 6.51 (p=1.05×10⁻¹²). Spatially, AI-powered WSI analysis corroborated this, linking WGD+ tumors to fewer inflamed and more immune-desert phenotypes (p=2.74×10⁻ 3 ). Clinically, WGD status was significantly associated with inferior outcomes. Compared to non-WGD patients, WGD patients demonstrated worse overall survival (HR 1.284, 95% CI 1.050-1.569, p=1.47×10⁻ 2 ) and progression-free survival (HR 1.286, 95% CI 1.057-1.563, p=1.18×10⁻ 2 ).
Conclusion: In NSCLC, WGD promotes high genomic instability through TP53 mutation and critical HLA Class I LOH, enabling immune escape. This genomic landscape shapes an immunosuppressive ecosystem characterized by metabolic reprogramming, lower T cell infiltration, and immune-desert phenotypes. As a result, WGD defines an immune-refractory microenvironment and serves as a robust predictive biomarker for unfavorable ICI response.
利益披露 Disclosure
C. Joe, None..
J. Kim, None..
H. Kim, None..
J. Choe, None..
M. Jang, None..
E. Oh, None..
N. Lee, None..
S. Kim, None..
S. Park, None..
H. Jung, None..
J. Sun, None..
J. Ahn, None..
M. Ahn, None..
J. Park, None..
S. Lee, None.