Biomarker analysis of neoadjuvant atezolizumab in upper-tract urothelial carcinoma: Results from the ABACUS-2 trial
No Thumbnail Available
Authors
Szabados, Bernadett Emma
Rodriguez-Vida, Alejo
Noguerón, Esther
Gross-Goupil, Marine
Hussain, Syed A.
Jia, Shidong
Jones, Robert
Zhang, Frank
Du, Pan
Jiang, Guanglong
Contact
Check for full-text access
Issue Date
2026
Type
Article
Language
Keywords
Alternative Title
Abstract
Background: High-risk upper-tract urothelial carcinoma (UTUC) is associated with poor clinical outcomes and there are no standard neoadjuvant therapy options currently. ABACUS-2 evaluates two cycles of neoadjuvant atezolizumab prior to radical nephroureterectomy (RNU). Here, we report circulating (ctDNA) and urinary tumour DNA (utDNA) findings. Methods: ABACUS-2 (NCT04624399) is a single-arm phase II study of atezolizumab (1200 mg IV Q3W × 2) prior to RNU in patients with high-grade or high-risk (hydronephrosis, tumour > 2 cm, multifocality, variant histology, or prior RC for UC) UTUC (cT1-4a N0-1 M0). Co-primary endpoints were pathological complete response (pCR; pT0N0) and predefined biomarker analyses on serial liquid biopsies. For these analyses, baseline genomic profiling was performed using the PredicineWES+ assay to identify somatic variants, followed by longitudinal minimal residual disease (MRD) monitoring with the PredicineBEACON platform. Correlations were explored between pCR, relapse-free survival (RFS) and overall survival (OS) with ctDNA/utDNA dynamics. Results: 27 pts received at least 1 cycle of atezolizumab. 4 pts did not undergo RNU (3 due to disease progression, 1 unfit for RNU). pCR occurred in 7% (2/27). After a median follow-up of 18.9 months (95% CI 13.9 –25.6), disease recurrence was observed in 26% (6/27) and 15% (4/27) had died. Treatment was well tolerated, with no surgery delays reported. 24 pts had successful MRD tests in at least one of the time points from screening to end of follow-up and were included in this biomarker evaluable subset. At baseline 25% (5/20) of pts were ctDNA(+) and all pts (100%, 14/14) were utDNA(+). After neoadjuvant atezolizumab and RNU, 80% (4/5) pts had ctDNA clearance. Only 1 pt demonstrated utDNA clearance following neoadjuvant atezolizumab, which correlated with pCR on RNU. Post-RNU, 3 pts were ctDNA(+) and subsequently developed metastatic disease. Among 17 pts with post-RNU utDNA monitoring, 29% (5/17) showed repeated utDNA positivity at multiple follow-up timepoints, correlating with bladder tumour recurrence. ctDNA(-) status post neoadjuvant atezolizumab correlated with significantly longer RFS (p = 0.024) and OS (p = 0.048). Conclusions: Neoadjuvant atezolizumab was feasible and well tolerated in patients with high-risk UTUC, but pathological complete responses were infrequent. Post-operative liquid biopsy analyses revealed that ctDNA and utDNA provide complementary information on systemic and local disease, respectively. Persistent or recurrent ctDNA positivity post-RNU was strongly associated with metastatic relapse and inferior survival, while utDNA positivity during follow-up correlated with intravesical recurrence. These findings support the utility of plasma and urine MRD monitoring for risk stratification and early relapse detection in UTUC. Clinical trial information: NCT04624399.
Description
Citation
Publisher
License
Journal
Journal of Clinical Oncology
Volume
44