The Feasibility of Using Biomarkers Derived from Circulating Tumor DNA Sequencing as Predictive Classifiers in Patients with Small-Cell Lung Cancer

Background: Exploring predictive biomarkers to individualize patients with small cell lung cancer (SCLC) remains challenging in clinical practice, and current imaging techniques cannot fully meet the clinical needs due to their inherent limitations. Methods: A total of 127 longitudinal biological DNA samples from 35 SCLC patients treated with first-line standard therapy were analyzed by capture-based targeted (average coverage for tissue and circulating tumor DNA (ctDNA) samples were 841× and 2670.5×, respectively) genome next-generation sequencing, which was designed to cover coding sequencing or hot exons of 1,021 genes frequently mutated in solid tumors. PyClone was used to infer the molecular tumor burden index (mTBI). Pre-treatment tumor tissues [T1] and serial plasma samples were collected (pre-treatment [B1], after two [B2], six [B3] cycles of chemotherapy and at progression [B4]). Findings: In total, 38 SCLC patients were enrolled with 35 patients included in the analysis set. The overall concordance rate between tissue and plasma sequencing in 30 patients with matched T1 and B1 samples was 66.5%. The median progression-free survival (PFS) was 9.8 months (95% confidence interval [CI]: 6.3-13.3). Patients with persisted RB1 and/or TP53 mutations at B2 demonstrated inferior PFS compared with others (5.2 months [95%CI: 3.8-6.6] vs. 13.1 months [95%CI: 9.8-16.4], p=0.06). The median PFS for patients with persisted RB1 mutations at B2 was significantly worse than the other two groups (4.2 months [95%CI: 2.8-5.6] vs. not reached, p<0.0001 and vs. 10.2 months, [95%CI: 6.3-14.1], p=0.001). The performance of mTBI monitoring was highly consistent with tumor imaging, and earlier mTBI elevation was observed in 64.7% of patients with B3 samples. In addition, the dynamic changes of mTBI from B4 to B1 were sensitive to predict new metastases, with sensitivity and specificity of 100% and 85.7% (95%CI: 50.8-120.7), respectively. Interpretation: Monitoring ctDNA based RB1 mutations and mTBI provided a feasible tool to predict prognosis of first-line therapy in SCLC, which guaranteed the development of prospective stratification in future trials. Trial Registration: This study was registered at the Chinese Clinical Trial Registry as ChiCTR1900023956. Funding: This study was supported by the National Key Research and Development Project (2019YFC1315704, to ZW), CAMS Innovation Fund for Medical Sciences (2017-I2M-1-005, to ZW), the National Natural Sciences Foundation (81871889, 82072586 to ZW) and Beijing Natural Science Foundation (7212084 to ZW). Declaration of Interest: The authors declare no potential competing interests. Ethical Approval: All procedures were conducted in accordance with the Declaration of Helsinki. This study was approved by the ethics committee of National Cancer Center/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College (Beijing, China; approved No. 18- 151/1729).