Clinical application of individualized tumor-informed circulating tumor DNA for therapeutic response and relapse prediction in patients with neuroblastoma.

10053 Background: Patients with high-risk neuroblastoma (NB) generally present with widely metastatic disease and often relapse despite intensive therapy. At present, circulating tumor DNA (ctDNA)-based minimal residual disease (MRD) has been studied in various tumors and proven to be valuable in guiding treatment and predicting recurrence, but its application in NB has rarely been reported. This study aimed to assess the effectiveness of tumor-informed ctDNA monitoring for evaluating the response to induction therapy in NB patients, as well as its role in predicting the risk of relapse. Methods: We conducted a prospective cohort study of NB patients. Primary tumor samples were collected at baseline and blood samples were collected sequentially from baseline until disease progression or the last follow-up. Tumor tissue samples were subjected to whole exome sequencing (WES), and clonal mutations were selected for the individualized MRD panel. Sequencing was conducted at OrigiMed, a College of American Pathologists accredited and Clinical Laboratory Improvement Amendments certified laboratory. Results: Our cohort included 43 individuals diagnosed with NB, with a median age of 3.5 years (0.42-40 years). WES revealed that 75% (2054/2753) of the mutated genes were unique to each patient, and 98.9% (704/712) of the selected tumor-informed single nucleotide variants were variants of unknown significance, suggesting that individualized tumor-informed MRD (n = 23) is superior to fixed tumor-agnostic panel (n = 20) MRD in NB. A successful individualized MRD panel was constructed for 53.5% (23/43) of patients, with a remarkable 92.3% success rate for those older than 5 years (p < 0.01). The initial positivity rate for ctDNA detection was 68%, with the initial positivity rate being associated with the stage (p < 0.01). During treatment and follow-up, 0/6 of the patients whose ctDNA levels remained negative experienced a relapse, whereas 2/5 patients who had a negative conversion or an increase in positivity experienced a relapse. In addition, ctDNA changes matched radiological evaluations in 94% of patients. Notably, ctDNA was detected prior to radiological relapse, with a lead time of 6 months in one patient. Patients can be divided into four groups according to the mutation landscape: ALK mutation, MYCN amplification, chromosome amplification, and other groups. Among them, the rate of ctDNA clearance was the highest in the chromosome amplification group, indicating that patients in this group were the most sensitive to chemotherapy. Conclusions: Individualized tumor-informed ctDNA monitoring shows promise for evaluating therapeutic response and forecasting relapse in NB patients. This approach proves more effective than fixed tumor-agnostic panel-based MRD due to high genetic mutation uniqueness. Clinical trial information: NCT05076071 .