使用 GeoMx 數位空間分析技術進行子宮內膜異位症相關卵巢透明細胞癌的空間蛋白質體分析

Abstract

標題: 使用 GeoMx 數位空間分析技術進行子宮內膜異位症相關卵巢透明細胞癌的空間蛋白質體分析

背景: 子宮內膜異位症是一種慢性發炎與纖維化的婦科疾病，且長期以來被證實與卵巢透明細胞癌（ovarian clear cell carcinoma, OCCC）之發生風險增加密切相關。儘管兩者之間的流行病學關聯已被廣泛建立，目前針對辨識具有惡性轉化風險之子宮內膜異位症患者的策略仍然不足，且在疾病進展過程中，上皮、免疫與基質微環境層級的生物學變化仍存在顯著知識缺口。傳統的整體（bulk）分子分析方法缺乏空間解析度，無法保留組織內的空間脈絡，因而限制了對不同組織區室中特定生物學程式的理解，而這些程式可能正是子宮內膜異位症相關卵巢癌發生的關鍵基礎。

方法: 本研究採用 NanoString GeoMx® 數位空間分析技術（Digital Spatial Profiling, DSP），結合免疫腫瘤蛋白質體圖譜（Immuno-Oncology Proteome Atlas），進行空間蛋白質體分析，以比較子宮內膜異位症、原發性卵巢透明細胞癌與復發性卵巢透明細胞癌之蛋白質表現特徵。研究分析來自兩位患者的福馬林固定石蠟包埋（FFPE）組織切片，並依據空間分布劃分為三個明確的組織區室：PanCK⁺ 上皮區、CD45⁺ 免疫區，以及 PanCK⁻/CD45⁻ 基質區。於各區室內分別進行不同疾病狀態之蛋白質差異表現分析。蛋白質體結果進一步結合 NanoString 所提供之癌症標誌（Hallmarks of Cancer）主題性註解，透過關聯層級（association-level）分析架構，以辨識與疾病進展相關的主要生物學模式。

結果: 子宮內膜異位症呈現出一種以纖維化與發炎為主、並適應低氧環境的微環境特徵，表現為細胞外基質重塑、免疫參與，以及代謝壓力調適等生物學程式，且此現象橫跨所有組織區室。原發性卵巢透明細胞癌則顯示上述壓力適應程式的進一步放大，並伴隨顯著的上皮細胞增殖訊號、上皮結構穩定性，以及高度代謝需求的腫瘤相關免疫表型。復發性卵巢透明細胞癌則以「選擇性強化」之持續存活導向特徵為主要表現，包括增殖優勢的進一步主導、上皮收縮性增加，以及與溶酶體運輸及上皮—免疫介面互動相關的免疫調適機制。整體而言，多項癌症標誌相關主題在不同疾病狀態中持續存在，包括細胞能量代謝失衡、促腫瘤性發炎反應、侵襲相關程式的活化，以及免疫逃避機制。

結論: 本研究結果支持一個疾病連續體模型，顯示子宮內膜異位症雖屬非惡性狀態，但其微環境已具備生物學上被「預先啟動」的特徵，並與卵巢透明細胞癌共享多項關鍵分子特徵。惡性轉化與復發過程似乎並非源自全新生物狀態的出現，而是透過既有壓力適應程式的放大與選擇性強化而逐步形成。空間蛋白質體分析能夠提供對不同組織區室中微環境演化的關鍵洞見，並突顯以微環境為核心的研究策略，對於理解子宮內膜異位症相關卵巢透明細胞癌之發生與進展具有重要意義。

Title: Spatial Proteomic Profiling of Endometriosis-Associated Ovarian Clear Cell Carcinoma Using GeoMx Digital Spatial Profiling

Background: Endometriosis is a chronic inflammatory and fibrotic gynecologic disorder that has been consistently associated with an increased risk of ovarian clear cell carcinoma (OCCC). Despite this epidemiological link, strategies to identify endometriosis patients at risk of malignant transformation remain underdeveloped, and significant gaps persist in understanding how disease progression occurs at the level of the epithelial, immune, and stromal microenvironments. Traditional bulk molecular analyses obscure spatial context, limiting insight into compartment-specific biological programs that may underlie endometriosis-associated ovarian carcinogenesis.

Methods: In this study, spatial proteomic profiling was performed using NanoString GeoMx® Digital Spatial Profiling with the Immuno-Oncology Proteome Atlas to compare endometriosis, primary OCCC, and recurrent OCCC. Formalin-fixed paraffin-embedded (FFPE) tissue sections from two patients were analyzed across three spatially defined compartments: PanCK⁺ epithelial, CD45⁺ immune, and PanCK⁻/CD45⁻ stromal regions. Differential protein expression analyses were conducted for disease-state comparisons within each compartment. Proteomic findings were interpreted using an association-level framework based on NanoString’s Hallmarks of Cancer thematic annotations to identify dominant biological patterns associated with disease progression.

Results: Endometriosis exhibited a fibro-inflammatory, hypoxia-adapted microenvironment characterized by extracellular matrix remodeling, immune engagement, and metabolic stress adaptation across all compartments. Primary OCCC demonstrated amplification of these stress-adaptive programs, with marked enrichment of epithelial proliferative signaling, epithelial cohesion, and metabolically demanding tumor-associated immune profiles. Recurrent OCCC was distinguished by selective reinforcement of persistence-oriented features, including proliferation dominance, increased epithelial contractility, and immune adaptation linked to lysosomal trafficking and epithelial–immune interface association. Hallmark-associated themes were maintained across disease states, including deregulated cellular energetics, tumor-promoting inflammation, activation of invasion-associated programs, and avoidance of immune destruction.

Conclusions: These findings support a continuum model in which endometriosis represents a biologically primed but non-malignant microenvironment that shares key molecular features with OCCC. Malignant transformation and recurrence appear to arise through amplification and selective reinforcement of pre-existing stress-adaptive programs rather than the emergence of entirely novel biological states. Spatial proteomic profiling provides critical insight into compartment-specific microenvironmental evolution and underscores the importance of microenvironment-focused approaches for understanding endometriosis-associated ovarian clear cell carcinoma.