PO.CL01.08 · 临床研究
肿瘤知情的循环肿瘤 DNA 动态反映肾细胞癌的侵袭性组织学和影像学疾病状态
Tumor-informed circulating tumor DNA dynamics reflect aggressive histology and radiologic disease status in renal cell carcinoma
作者与单位 Authors & Affiliations
摘要 Abstract
中文摘要
引言：循环肿瘤 DNA（ctDNA）能够对肿瘤生物学和治疗反应进行微创监测。尽管 ctDNA 在各种恶性肿瘤中的应用日益增多，但其在肾细胞癌（RCC）中的效用尚未得到充分确立。我们试图了解 ctDNA、肿瘤组织学与临床结局在局部晚期（laRCC）和转移性 RCC（mRCC）中的关系。
方法：48 例 laRCC 和 mRCC 患者接受了使用商用肿瘤知情检测（Signatera）的系列血浆 ctDNA 检测。对每例患者的肿瘤组织进行测序以识别个体化变异，随后利用这些变异构建个性化的肿瘤知情 ctDNA 检测。系列血浆检测对这些患者特异性变异进行定量，以结合间隔影像学评估 ctDNA 动态。分析了基线 ctDNA 状态、以每毫升平均肿瘤分子数（MTM/mL）表示的 ctDNA 峰值水平、组织学亚型、侵袭性形态学特征（肉瘤样、横纹肌样、坏死）与影像学发现之间的关联。定量变量采用 Kruskal-Wallis 检验和 Mann-Whitney U 检验进行比较。生存期采用 Kaplan-Meier 和 log-rank 检验进行评估。
结果：总体上 39.5%（17/48）的患者在基线时检测到 ctDNA，其中 laRCC 为 5.2%（1/19），mRCC 为 55.2%（16/29）。在透明细胞（cc）RCC 中，具有侵袭性形态学特征（肉瘤样、横纹肌样或坏死）的病例基线 ctDNA 检出率为 66%，而无这些特征的透明细胞肿瘤为 15%。非透明细胞 RCC 的基线检出率为 42%（3/7）。在 ccRCC 中，具有侵袭性特征的肿瘤中位 ctDNA 峰值水平显著高于无此类特征者（0.82 对 0.07 MTM/mL，p < 0.05）。ctDNA 峰值水平与影像学疾病状态密切相关：疾病进展（PD）患者的中位 ctDNA 水平显著更高（8.86 MTM/mL），而表现为疾病控制的患者，包括完全缓解（CR，0.00 MTM/mL）、部分缓解（PR，0.41 MTM/mL）或疾病稳定（SD，0.27 MTM/mL；H = 9.16，p = 0.027）水平较低。当按疾病控制（CR/PR/SD）与疾病进展（PD）分层时，疾病控制的患者峰值 ctDNA 水平显著低于进展患者（中位 0.18 MTM/mL，n = 12 对中位 8.86 MTM/mL，n = 8；p = 0.007）。在疾病病程中从 ctDNA 阳性转为 ctDNA 阴性与总生存期改善相关（p < 0.05）。然而，单独的基线 ctDNA 检出与生存期无显著相关性。
结论：ctDNA 可作为 RCC 侵袭性组织病理学特征的预测指标。定量负荷和 ctDNA 动态似乎与疾病状态和总生存期相关。这些发现凸显了肿瘤知情 ctDNA 作为 RCC 中一种有前景的生物标志物。
查看英文原文 English abstract
Introduction: Circulating tumor DNA (ctDNA) enables minimally invasive monitoring of tumor biology and treatment response. Although ctDNA is increasingly used across malignancies, its utility in renal cell carcinoma (RCC) has not been well established. We sought to understand the relationship between ctDNA, tumor histology, and clinical outcomes in locally advanced (laRCC) and metastatic RCC (mRCC).
Methods: Forty-eight patients with laRCC and mRCC underwent serial plasma ctDNA testing with a commercially available, tumor-informed assay (Signatera). Tissue from each patient's tumor was sequenced to identify individualized variants, which were then used to construct a personalized tumor-informed ctDNA assay. Serial plasma testing quantified these patient-specific variants to assess ctDNA dynamics alongside interval imaging. Associations between baseline ctDNA status, peak level of ctDNA in mean tumor molecules per milliliter (MTM/mL), histologic subtype, aggressive morphologic features (sarcomatoid, rhabdoid, necrosis), and radiologic findings were analyzed. Quantitative variables were compared using Kruskal-Wallis and Mann-Whitney U tests. Survival was evaluated using Kaplan-Meier and log-rank testing.
Results: ctDNA was detected at baseline in 39.5% (17/48) of patients overall, including 5.2% (1/19) with laRCC and 55.2% (16/29) with mRCC. Among clear cell (cc) RCC, baseline ctDNA was detected in 66% of cases with aggressive morphologic features (sarcomatoid, rhabdoid, or necrosis) versus 15% of clear cell tumors without these features. Baseline detection in non-clear cell RCC was 42% (3/7). Among ccRCC, the median peak ctDNA level was significantly higher in tumors with aggressive features compared to those without (0.82 vs 0.07 MTM/mL, p < 0.05). Peak ctDNA levels were strongly associated with radiologic disease status: patients with progressive disease (PD) had markedly higher median ctDNA levels (8.86 MTM/mL) compared with those demonstrating disease control with complete response (CR, 0.00 MTM/mL), partial response (PR, 0.41 MTM/mL), or stable disease (SD, 0.27 MTM/mL; H = 9.16, p = 0.027). When stratified by disease control (CR/PR/SD) versus disease progression (PD), patients with disease control exhibited significantly lower peak ctDNA levels (median 0.18 MTM/mL, n = 12) than those with progression (median 8.86 MTM/mL, n = 8; p = 0.007). Conversion from ctDNA-positive to ctDNA-negative during the disease course was associated with improved overall survival (p < 0.05). However, baseline ctDNA detection alone did not correlate significantly with survival.
Conclusions: ctDNA may serve as a predictor of aggressive histopathologic features in RCC. Quantitative burden and ctDNA dynamics appear associated with disease status and overall survival. These findings highlight tumor-informed ctDNA as a promising biomarker in RCC.
利益披露 Disclosure
E. Martin, None..
E. Yu, None..
A. Murthi, None..
H. Wang, None..
L. Linville, None..
J. B. Aragon-Ching, None.