肺淋巴上皮癌的預後因子及預測腫瘤復發風險的新分級系統

Abstract

肺淋巴上皮癌是非小細胞肺癌的一種罕見亞型，主要在東南亞地區流行。從組織學上看，其具有未分化的腫瘤細胞，核為泡樣核、核仁明顯，通常伴隨著大量的淋巴細胞浸潤。此外，肺淋巴上皮癌常與人類第四型疱疹病毒（EBV）感染相關。由於這種疾病的罕見性和地區限制性，對於肺淋巴上皮癌的研究較少，其相關預後因子仍未清楚。本研究旨在通過分析各種潛在因素，包括臨床特徵 (年齡、性別、吸菸史、胸膜浸潤、腫瘤大小、腫瘤分期、手術方式)、組織病理特徵(腫瘤型態、腫瘤&淋巴球比值、TILs、TLS、STAS、肉芽腫、腫瘤壞死、淋巴管侵犯、腫瘤出芽)、以及免疫組織化學染色標記(甲基硫腺苷酸磷酸酶、D 型細胞週期素、生長抑制素受體2A、細胞程式死亡-配體1)，來確定PLEC的有效預後因子。我們回顧了手術切除的I-III期肺淋巴上皮癌病例（n = 56）的腫瘤切片，並使用Kaplan-Meier方法進行生存分析，評估每個因子對總體生存（OS）和無復發生存期（RFS）的影響。 我們的結果顯示，幾個與OS相關的重要負面預後因素包括性別（p = 0.007）、吸煙史（p = 0.016）和STAS（p = 0.029）。對於RFS，顯著的負面預後因素包括胸膜浸潤（p = 0.003）、腫瘤大小（p = 0.026）、分期（p = 0.001）、STAS（p = 0.040）和淋巴管侵犯（p = 0.002）。為了進一步驗證我們的結果，我們分別對OS和PFS進行了多變量Cox回歸分析。多變量Cox回歸模型顯示，所有分析的因素在OS方面都沒有顯著相關。然而，STAS和分期（III期相對於I期）在PFS方面被證明是獨立的預後因素，其風險比率分別為3.748（p = 0.036）和10.631（p = 0.007）。 基於在本研究中發現STAS和PFS之間的顯著相關性，我們針對肺淋巴上皮癌提出了一個新的分級系統，該系統基於腫瘤邊界和腫瘤是否表現STAS來進行分級。根據這個以腫瘤邊界型態(tumor border pattern, TBP)的分級系統，腫瘤被分為三個等級：TBP1級腫瘤（15例，26.8%）具有明確完整的腫瘤邊界；TBP2級腫瘤（23例，41.1%）具有棘刺狀邊界但沒有STAS；TBP3級腫瘤（18例，32.1%）具有棘刺狀邊界和STAS。這個分級系統顯示出顯著的預後價值，如每個分級觀察到的PFS率有顯著的差異 (p = 0.036)：TBP1腫瘤為92.9%，TBP2腫瘤為65.7%，TBP3腫瘤為45.0%。 總結來說，我們的研究表明STAS和腫瘤分期（III期相對於I期）作為肺淋巴上皮癌PFS的獨立預後因素。此外，基於腫瘤邊界和STAS新提出的腫瘤邊界型態(tumor border pattern, TBP)的三級分級系統顯示出顯著的預後價值，為肺淋巴上皮癌的臨床決策提供了有價值的信息。

Pulmonary lymphoepithelial carcinoma (PLEC) is a rare subtype of non-small cell lung cancer (NSCLC) primarily prevalent in Southeast Asia. Histologically, PLEC is characterized by undifferentiated tumor cells with vesicular nuclei and prominent nucleoli, often accompanied by extensive lymphocyte infiltration. Additionally, PLEC is commonly associated with Epstein-Barr virus (EBV) infection. Given the rarity of the disease and regional limitations, there are limited studies on PLEC, and prognostic factors of PLEC have still remained unclear. In this study, we aimed to identify prognostic factors for PLEC by analyzing various factors, including clinical characteristics (age, gender, smoking history, pleural invasion, tumor size, stage, surgical method), histopathological features (tumor pattern, T: L, TILs, TLS, STAS, granuloma, tumor necrosis, LVI, tumor budding), and immunohistochemical (IHC) markers (MTAP, cyclin D1, SSTR2A, PD-L1). We retrospectively reviewed tumor slides from surgically resected stage I-III PLEC cases (n=56) and conducted survival analysis using Kaplan-Meier method to assess overall survival (OS) and recurrence-free survival (RFS) to find out prognostic factors of PLEC. Our results revealed several factors significantly associated with shorter OS, including gender (p = 0.007), smoking history (p = 0.016), and STAS (p = 0.029). For RFS, significant negative prognostic factors included pleural invasion (p = 0.003), tumor size (p = 0.026), stage (p = 0.001), STAS (p = 0.040), and LVI (p = 0.002). To further validate our findings and exclude confounding factors, we performed multivariate Cox regression analysis separately for OS and PFS. The multivariable Cox regression model revealed that none of the analyzed factors were significantly associated with OS. However, STAS and stage (stage III vs stage I) emerged as independent prognostic factors for RFS, with hazard ratios of 3.748 (p = 0.036) and 10.631 (p = 0.007), respectively. Based on the significant association between STAS and RFS, we proposed a novel 3-tier grading system for PLEC that based on tumor border and the presence of STAS. According to this grading system, tumors were classified into three grades according to their tumor border patterns (TBP): TBP1 tumor (15, 26.8%) with well-defined tumor borders, TBP2 tumor (23, 41.1%) with spiculated borders without STAS, and TBP3 tumor (18, 32.1%) with spiculated borders and STAS. This grading system demonstrated significant prognostic value, as evidenced by the RFS rates observed in each tier: 92.9% for TBP1 tumors, 65.7% for TBP2 tumors, and 45.0% for TBP3 tumors. In conclusion, our study identified STAS and stage (stage III vs stage I) were independent prognostic factors for RFS in PLEC. Additionally, the newly proposed 3-tier TBP grading system showed significant prognostic value, providing valuable information for clinical decision-making in PLEC.