Raine Medical Research Foundation

Abstract Introduction: Circulating tumor DNA (ctDNA) is the most frequently utilized liquid biopsy. High sensitivity is critical for minimal residual disease detection to inform treatment response, risk stratification and molecular relapse. Whole-genome-based, tumor-informed strategy is increasingly being used in ctDNA detection in adult cancers but not in pediatric cancers. Methods: We designed a personalized ctDNA assay which utilized paired tumor-normal whole genome sequencing data of individual patient to design a panel of ∼400 patient-specific tumor biomarkers. 83 patients with high-risk solid tumors from the ZERO Childhood Cancer Precision Medicine Program in Australia who had 2 or more serial liquid biopsy samples collected were selected for analysis. Results: The cohort consisted of 54 sarcoma, 16 neuroblastoma and 13 other solid tumors. 303 liquid biopsy samples were analyzed for plasma ctDNA, ranging from 2 - 7 samples per patient. The detection limit was between 10^-4 and 10^-5 ctDNA fraction for 90% of the samples and between 10^-3 and 10^-4 for the remaining 10%. 73 (88%) patients had at least one blood sample with detectable ctDNA. ctDNA detection rate was 96% in newly diagnosed disease, with ctDNA fraction ranging from 0.01 to 99% (median 36%), 88% in off-treatment relapse (median 3.6%; range: 0.011 – 77%) and 85% in refractory/progressive disease (PD) (median 0.8%; range: 0.37 – 81%). For patients with complete response (CR), 89% had undetectable ctDNA. In the remaining 11% with CR and detectable ctDNA, majority (71%) experienced relapse. ctDNA levels were significantly associated with disease burden and treatment response. Patients with metastatic disease at enrolment had significantly higher ctDNA fraction (median 8.2%; range: 0 – 99%) than those with localized disease (median 0.06%; range: 0 – 42%) (P<0.05). The median ctDNA fraction corresponding to CR was 0% (range: 0 – 4.0%), 0.0095% (range: 0 – 4.7%) for partial response, 0.022% (range: 0 – 23%) for stable disease and 5.6% (range: 0 – 82%) for PD (P<0.0001). Most importantly, ctDNA predicted clinical outcomes. Patients with early clearance of ctDNA between 60 and 120 days from start of treatment, compared with those with detectable ctDNA, had significantly better PFS (5-yr PFS 47% vs 21%; P<0.01) and OS (5-yr OS 54% vs 37%; P<0.05). Patients with undetectable ctDNA at the end of treatment (+/-42 days from completing treatment), compared with those with persistence ctDNA, also had significantly better PFS (5-yr PFS 50% vs 0%; P<0.0001) and OS (5-yr OS 66% vs 17%; P<0.001). Conclusion: This ultrasensitive personalized ctDNA assay with a detection limit between 10^-4 and 10^-5 ctDNA fraction is highly correlative of disease burden and treatment response. Early clearance of ctDNA and persistence of ctDNA at end of the treatment both correlated with clinical outcome. The assay will be evaluated prospectively to determine its clinical utility in complementing treatment response evaluation, detecting early recurrence and predicting outcome in pediatric cancers. Citation Format: Loretta M.S. Lau, Rob Salomon, Wenhan Chen, Charles Shale, Mojgan Toumari, Wenyan Li, Jingwei Chan, Sajad Razavi-Bazaz,, Louise Cui, Sam El-Kamand, Marie Wong, Edwin Cuppen, Michelle Haber, Vanessa Tyrrell, Paul G Ekert, David S Ziegler, Peter Priestley, Mark J Cowley. Whole-genome-based, tumor-informed circulating tumor DNA detection correlates with treatment response and survival of pediatric solid tumor [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Discovery and Innovation in Pediatric Cancer— From Biology to Breakthrough Therapies; 2025 Sep 25-28; Boston, MA. Philadelphia (PA): AACR; Cancer Res 2025;85(18_Suppl_2):Abstract nr A001-PR009.