PO.CL01.22 · 临床研究
晚期实体瘤组织与液体活检CGP的比较:来自社区癌症中心的见解
Comparison of tissue and liquid biopsy CGP in advanced solid tumors: Insights from a community cancer center
作者与单位 Authors & Affiliations
摘要 Abstract
中文摘要
引言:二代测序(NGS)肿瘤分型是指导肿瘤学治疗的支柱。循环肿瘤DNA(ctDNA)检测提供了一种额外的、微创的检测可干预变异的方法。我们评估了大平原地区一个以农村为主的队列中,实体组织与液体活检综合基因组分析(CGP)结果之间的一致性。
方法:参与者纳入Avera癌症测序与分析方案(ASAP;NCT05142033)。纳入标准要求为晚期不可切除或转移性实体瘤,并在90天内同时采集肿瘤及液体活检CGP。一致性基于报告的变异计算,排除了两个检测组均未覆盖的变异(液体活检组:Tempus xF+;实体组织组:Labcorp OmniSeq INSIGHT®;分析的变异:388个SNV、7个CNV、10个融合)及意义未明的变异(VUS)。
结果:86例患者符合标准(平均年龄±标准差:65±10岁;52%为女性)。共代表17种肿瘤类型;肺癌(26%)及头颈癌(16%)最为常见,20%有≥2种癌症诊断。患者水平的完全一致性(定义为组织与液体活检中鉴定出的生物学相关变异集合完全相同)为59%。所有变异类型的变异水平一致性为31%(CNV[n=23]:22%,融合[n=6]:50%,SNV[n=365]:30%)。在独特变异中,大多数SNV(54%)及CNV(89%)仅在组织中检测到。从液体活检中过滤17个与克隆性造血(CH)相关的基因后,所有变异类型的变异水平一致性为35%(CNV[n=23]:22%,融合[n=6]:50%,SNV[n=296]:38%)。将OncoKB治疗证据级别(1-4及R1-2)应用于所鉴定的变异及肿瘤类型。在数据集中112个具有注释证据级别的变异中,42个为一致变异,45个为组织独有,25个为液体活检独有。MSI状态100%一致(2例MSI-high/84例MSI稳定)。高肿瘤突变负荷(TMB-H)显示90%一致性(72/80一致,8/80不一致,6例不可评估)。后续分析将按肿瘤类型、采样组织、转移部位、治疗、液体与组织样本采集间隔时间及从诊断到样本采集时间评估一致性。
结论:组织与液体活检CGP各自为治疗选择提供互补的、可干预的信息。40%的患者在液体或组织检测中有独特发现。这些结果支持在晚期实体瘤中采用综合CGP策略的价值。
查看英文原文 English abstract
Introduction: Next-generation sequencing (NGS) tumor profiling is a mainstay for guiding therapy in oncology. Circulating tumor DNA (ctDNA) testing offers an additional, minimally invasive approach of detecting actionable alterations. We evaluated the concordance between solid tissue and liquid biopsy comprehensive genomic profiling (CGP) results in a largely rural cohort within the Great Plains.
Methods: Participants were enrolled in the Avera Cancer Sequencing and Analytics Protocol (ASAP; NCT05142033). Inclusion required an advanced unresectable or metastatic solid tumor and concurrent tumor and liquid biopsy CGP collected within 90 days. Concordance was calculated based on reported alterations after excluding those not covered by both panels (liquid biopsy panel: Tempus xF+; solid tissue panel: Labcorp OmniSeq INSIGHT®; analyzed alterations: 388 SNV, 7 CNV, 10 Fusions) and variants of unknown significance (VUS).
Results: Eighty-six patients met criteria (mean age +/- SD: 65 +/-10y; 52% female). Seventeen tumor types were represented; lung cancer (26%) and head-and-neck cancer (16%) were most frequent, and 20% had ≥ 2 cancer diagnoses. Patient-level full concordance, defined as identical sets of biologically relevant identified in both tissue and liquid biopsy, was 59%. Variant-level concordance across all alteration types was 31% (CNV [n=23]: 22%, Fusion [n=6]: 50%, SNV [n=365]: 30%). Among unique variants, most SNVs (54%) and CNVs (89%) were detected only in tissue. After filtering 17 genes related to clonal hematopoiesis (CH) from liquid biopsy, variant-level concordance across all alteration types was 35% (CNV [n=23]: 22%, fusion [n=6]: 50%, SNV [n=296]: 38%). OncoKB levels of therapeutic evidence (1-4 and R1-2) were applied to the variants and tumor types in which they were identified. Of the 112 variants with an annotated level of evidence in the data set, 42 were concordant variants, 45 unique to tissue, and 25 to liquid biopsy. MSI status was 100% concordant (2 MSI-high / 84 MSI-stable). High tumor mutational burden (TMB-H) showed 90% concordance (72/80 concordant, 8/80 discordant, 6 non-evaluable). Additional analyses will assess concordance by tumor type, tissue sampled, metastatic sites, treatment, time between liquid and tissue sample collection, and time from diagnosis to sample collection.
Conclusions: Tissue and liquid biopsy CGP each provide complementary, actionable information for therapy selection. Forty percent of patients had unique findings in liquid or tissue-based testing. These results support the value of a comprehensive CGP strategy in advanced solid tumors.
利益披露 Disclosure
P. Swaminathan, None..
M. Perrin, None..
C. Hattum, None.
S. B. Hastings,
Labcorp Employment.
S. Mir,
Tempus AI Employment.
Z. Wallen,
Labcorp Employment.
L. Speroni,
Tempus AI Employment.
M. F. Green,
Labcorp Employment.
E. Teslow,
Tempus AI Employment.
H. Ko,
Labcorp Employment.
R. A. Previs,
Labcorp Employment.
K. Patel,
Labcorp Employment.
S. Ramkissoon,
Labcorp Employment.
D. Starks, None..
B. Solomon, None..
W. Spanos, None.
T. Meissner,
Tempus AI ).
LabCorp ).
Cellworks ).