Abstract 3653: Sequencing of plasma cfDNA from patients with locally advanced bladder cancer for surveillance and therapeutic efficacy monitoring

Abstract

Abstract Minimally invasive methods for assessment of tumor burden, early detection of disease relapse and for monitoring therapeutic efficacy are needed to improve individualized follow-up and treatment for patients diagnosed with bladder cancer. The ability to predict pathologic complete response after neoadjuvant chemotherapy (NAC) through detection of ctDNA in plasma may enable strategies for bladder preservation. The aim of the study was to use patient-specific mutations to identify residual disease, metastatic relapse and to monitor treatment response in longitudinally collected plasma samples. For this, 50 patients diagnosed with locally advanced muscle-invasive bladder cancer (MIBC) and scheduled for chemotherapy were prospectively recruited between 2013 and 2017. In total, 42 patients received four cycles of cisplatin-based NAC prior to cystectomy and 82% showed response (pathologic downstaging). Eight patients received 2-6 cycles of cisplatin-based first-line chemotherapy due to diagnosis of T4b or lymph node metastasis prior to cystectomy (3CR, 2PR, 2PD, 1 ongoing). So far, 8/50 patients (16%) experienced disease relapse and three patients had metastatic progression. The mean follow-up time after radical cystectomy (RC) was 320 days (65-973); the mean follow-up time for disease-free patients was 397 days (119-778). Whole-exome sequencing (106x mean target coverage) of tumor and germline DNA was performed from a time point before systemic treatment. MUTECT2 identified a mean of 33 high (6-121), 340 (67-2838) moderate and 223 (29-2955) low-impact SNVs or insertion-deletions (InDels) per tumor. High-impact mutations in known cancer genes such as TP53 (60%), KDM6A (34%), ARID1A (32%), RB1 (28%), BRCA2 (26%), FGFR3 (22%) and ERCC2 (20%) were identified with no significant difference between responders vs nonresponders. Signatures 2, 13 (APOBEC), and 3 (BRCA) were frequently identified, with signature 13 being more prevalent in nonresponders (p=0.05). Personalized, multiplex-PCR assay-panels were individually designed, targeting each patient's tumor-specific mutations in plasma. Targeted sequencing was performed using cfDNA from 4-16 longitudinally collected plasma from each patient taken pre- and post-systemic therapy and at scheduled control visits after RC. In a blinded fashion, ctDNA was analyzed and results were compared to pathologic response and radiographic imaging data for each patient. Conclusion: Preliminary results indicated that analysis of plasma cfDNA is an appropriate method to detect metastatic relapse after RC and to monitor treatment efficacy. Here we perform a larger ctDNA study, applying direct sequencing of plasma cfDNA using highly sensitive NGS methods to determine clinical applicability of the method. Data collection is ongoing and all details will be presented at the AACR 2018 meeting. Citation Format: Karin Birkenkamp-Demtröder, Emil Christensen, Shruti Sharma, Ann Taber, Mads Agerbæk, Himanshu Sethi, Hsin-Ta Wu, Svetlana Shchegrova, Philippe Lamy, Iver Nordentoft, Ryan Swenerton, Samantha Navarro, Tony Tin, Ramya Srinivasan, John So, Spenser Alexander, Solomon Moshkevich, Bernhard Zimmermann, Cheng-Ho Jimmy Lin, Jørgen Bjerggaard Jensen, Lars Dyrskjøt. Sequencing of plasma cfDNA from patients with locally advanced bladder cancer for surveillance and therapeutic efficacy monitoring [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 3653.