Determining Molecular and Genomic Characteristics of TROP2-high Non-Small Cell Lung Cancer (NSCLC)

IASLC WCLC 2026

Aug 19, 2026
Oncology
Presentation

K. Kennedy, Y. Yu, S. Fragkogianni, K.V. Layng, N. Sethakorn

Introduction
TROP2 is the target for several antibody-drug conjugates (ADCs), and its expression is correlated with poor prognosis and immunosuppressive features in NSCLC. Understanding TROP2 biology and RNA expression patterns may inform development of combination therapies and potential biomarkers of response to TROP2 ADCs. We examined patterns of TROP2 expression using a large database of NSCLC genomic and transcriptomic data.

 

Methods
We used the Tempus Lens Platform (Tempus AI, Inc., Chicago, IL) to query the de-identified, multimodal Tempus database for patients (pts) with an NSCLC diagnosis and no additional cancer diagnosis within 5 years who had undergone Tempus xT (DNA) and xR (RNA) testing (N = 3,974). RNA-seq data were normalized, quantified as transcripts per million (TPM), and reported as log2(TPM+1). Pts were categorized as TROP2-high or TROP2-low based on median TACSTD2 gene expression. Pt demographics, clinical characteristics, and actionable genomic alterations (AGAs) were compared using Fisher’s exact test, Pearson’s Chi-squared test, or a Wilcoxon rank sum test, as appropriate. The tumor immune microenvironment was estimated using quanTIseq. Real-world overall survival (rwOS) and progression-free survival (rwPFS) were defined as the time from treatment initiation to death or loss to follow up (FU) and as the time from treatment initiation to first progression, death, or loss to FU, respectively. Median OS and PFS were estimated by Kaplan-Meier and compared using log-rank tests.

 

Results
The demographics of TROP2-high vs. TROP2-low pts with NSCLC were similar. The TROP2-low group was slightly enriched in metastatic biopsy sites (45% vs. 42%, p < 0.001). Squamous cell carcinoma histology was enriched in the TROP2-high group (31% vs. 18%, p<0.001). TROP2-low NSCLC was more enriched in PD-L1-negative tumors (42% vs. 35%, p = 0.002). Pts harboring any AGA were more likely to be TROP2-low (49% vs. 42%, p < 0.001), and KRAS alterations were enriched in TROP2-low pts (26% vs. 20%, p <0.001). TROP2 expression did not correlate with rwPFS or rwOS in pts treated with any therapy or in subgroups treated with immunotherapies or ADCs. Although TROP2-high NSCLC had numerically higher rwOS in pts treated with ADCs, this cohort was small, the difference was not statistically significant, and the cohort was not limited to TROP2-targeted ADCs.

 

Conclusions
TROP2 gene expression was broadly detected in multiple genomic and histologic subtypes of NSCLC, and across multiple metastatic sites. Each subtype analyzed in this study showed a broad range of TROP2 expression. The genomic landscapes of TROP2-low vs. TROP2-high disease differed, with a higher frequency of KRAS alterations and overall AGAs in TROP2-low pts. Future work should investigate how TROP2 levels change after treatment with systemic therapies and elucidate the biological function(s) of TROP2 in distinct genomic subtypes of NSCLC.