Brain metastasis (BM) genetically diverges from the primary tumor in non‐small‐cell lung cancer (NSCLC). Hence, accurately capturing clinically relevant alterations is pivotal for the delivery of targeted therapies. Circulating tumor DNA (ctDNA) sequencing has emerged as a promising liquid biopsy in the biomarker‐based clinical management of recurrent and extracranial metastatic NSCLC. However, the absence of simultaneous sequencing data from brain metastatic sites prevents the definitive evaluation of the efficacy of ctDNA in representing genetic profiles in BM. Here, we performed parallel genomic comparisons between matched BM and primary tumor DNA, plasma ctDNA, and cerebrospinal fluid (CSF) ctDNA. The results indicated that CSF ctDNA had a greater ability than plasma ctDNA to comprehensively represent the mutational landscape of BM, with CSF ctDNA detecting all BM mutations in 83.33% of patients, while plasma ctDNA was only 27.78%. Mutant allele frequency (MAF) in CSF ctDNA was highly correlated with the tumor size of BM (r = 0.95), and the mean MAF in CSF ctDNA was higher than that in plasma ctDNA (38.05% vs. 4.57%, respectively). MAF and tumor mutational burden in CSF ctDNA were strongly associated with those in BM (r = 0.96 and 0.97, respectively). Of note, CSF ctDNA had significantly higher concordance with BM than plasma ctDNA (99.33% vs. 67.44%), facilitating the identification of clinically relevant mutations. Moreover, we found that plasma ctDNA has stronger profiling performance, with a concordance of 93.01% in multiple brain metastases, equivalent to CSF ctDNA. Collectively, our study indicates that CSF ctDNA is superior to plasma ctDNA in accurately representing the profiling of single BM. Plasma ctDNA could be an alternative liquid biopsy material to be applied in multiple brain metastatic NSCLC.

Homozygous deletion (HD) of CDKN2A and CDKN2B (CDKN2A/B^HD) is the most frequent copy‐number variation (CNV) in lung adenocarcinoma (LUAD). CDKN2A/B^HD has been associated with poor outcomes in LUAD; however, the mechanisms of its prognostic effect remain unknown. We analyzed genome, transcriptome, and clinical data from 517 patients with LUAD from the Cancer Genome Atlas (TCGA) and from 788 primary LUAD tumor and matched control samples from the MSK‐IMPACT clinical cohort. CDKN2A/B^HD was present in 19.1% of the TCGA‐LUAD cohort and in 5.7% of the MSK‐IMPACT cohort. CDKN2A/B^HD patients had shorter disease‐free survival and overall survival compared with CDKN2A/B^WT individuals in both cohorts. Differences in clinical features did not influence the outcomes in the CDKN2A/B^HD population. Mutation analyses showed that overall tumor mutational burden and mutations in classical drivers such as EGFR and RB1 were not associated with CDKN2A/B^HD. In contrast, homozygous deletion of type I interferons (IFN‐I^HD) frequently co‐occurred with CDKN2A/B^HD. CDKN2A/B and IFN‐I are co‐located in the same p21.3 region of chromosome 9. The co‐occurrence of CDKN2A/B^HD and IFN‐I^HD was not related to whole‐genome doubling, chromosome instability, or aneuploidy. Patients with co‐occurring CDKN2A/B^HD and IFN‐I^HD had shorter disease‐free survival and overall survival compared with CDKN2A/B^WT patients. CDKN2A/B^HDIFN‐I^HD had downregulated several key immune response pathways, suggesting that poor prognosis in CDKN2A/B^HD LUAD could potentially be attributed to an immunosuppressive tumor microenvironment as a result of IFN‐I depletion.

Objectives: Circulating tumor DNA (ctDNA) has emerged as a reliable prognostic biomarker in both early- and late-stage non-small cell lung cancer (NSCLC) patients. However, its role in NSCLC with pleural dissemination (M1a), a subset of disease with indolent biology, remains to be elucidated. Methods: We collected 41 M1a patients with serial ctDNA and CEA monitoring. Progression-free survival (PFS) was assessed between patients with different levels of ctDNA and CEA. An independent cohort of 61 M1a patients was included for validation. Results: At the diagnostic landmark, the detection rates for ctDNA and CEA were 22% and 55%, respectively. Among patients who experienced disease progression with pleural metastases, only ten had detectable ctDNA in longitudinal timepoints, resulting in a sensitivity of 50%. Moreover, there was no significant difference in PFS between patients with longitudinally detectable and undetectable ctDNA (HR: 0.86, 95% CI 0.33–2.23, p = 0.76). In contrast, patients with a decreasing CEA trend within 3 months after diagnosis were associated with an improved PFS (HR: 0.22; 95% CI, 0.03–1.48, p = 0.004). This finding is confirmed in an independent M1a patient cohort. Conclusions: Together, our findings suggest that M1a NSCLC with isolated pleural dissemination may represent a “non-shedding” tumor type, where ctDNA shows limited diagnostic and prognostic value. Monitoring early changes in CEA could be a more cost-effective predictor of disease progression.

Duplex sequencing technology has been widely used in the detection of low-frequency mutations in circulating tumor deoxyribonucleic acid (DNA), but how to determine the sequencing depth and other experimental parameters to ensure the stable detection of low-frequency mutations is still an urgent problem to be solved. The mutation detection rules of duplex sequencing constrain not only the number of mutated templates but also the number of mutation-supportive reads corresponding to each forward and reverse strand of the mutated templates. To tackle this problem, we proposed a Depth Estimation model for stable detection of Low-Frequency MUTations in duplex sequencing (DELFMUT), which models the identity correspondence and quantitative relationships between templates and reads using the zero-truncated negative binomial distribution without considering the sequences composed of bases. The results of DELFMUT were verified by real duplex sequencing data. In the case of known mutation frequency and mutation detection rule, DELFMUT can recommend the combinations of DNA input and sequencing depth to guarantee the stable detection of mutations, and it has a great application value in guiding the experimental parameter setting of duplex sequencing technology.