Unveiling the clinical impact of plasma-only mutations in non-small cell lung cancer: a Korean multicenter experience

Introduction

Lung cancer remains one of the most prevalent and fatal cancer worldwide, accounting for approximately 1.8 million cancer-related deaths annually (1). A significant proportion of non-small cell lung cancer (NSCLC) cases are associated with oncogenic driver mutations. The emergence of precision oncology and targeted therapies has contributed to improving treatment outcomes and quality of life for NSCLC patients (2). In addition to well-established oncogenic drivers such as EGFR, ALK, and ROS1, newly emerging biomarkers including HER2 mutations and NRG1 fusions have recently been identified as actionable targets (3,4). Beyond traditional tyrosine kinase inhibitors (TKIs), novel therapeutic modalities such as antibody-drug conjugates (ADCs) and bispecific antibodies are entering clinical practice, reflecting the rapid evolution of precision medicine.

Comprehensive genomic profiling underpins precision medicine by enabling detailed characterization of oncogenic drivers and resistance mechanisms, thereby guiding personalized therapeutic strategies (5). However, acquiring sufficient tissue for genomic profiling is often challenging, particularly in cases of disease progression or poor patient condition. Furthermore, identifying secondary resistance mutations following treatment failure is crucial, yet invasive rebiopsy procedures are frequently hindered by patient frailty and logistical limitations. Tissue biopsy has traditionally been the gold standard for molecular diagnostics, enabling the detection of oncogenic driver mutations and resistant mechanisms. Nevertheless, its invasive nature poses risks of discomfort, complications, and procedural infeasibility in certain clinical scenarios. Furthermore, tissue genetic profiling often requires prolonged turnaround times, potentially delaying the initiation of appropriate therapies. Additionally, conventional biopsies sample only limited tumor regions, failing to fully capture intratumoral heterogeneity. These limitations highlight the need for less invasive, more comprehensive approaches to genomic profiling that can better reflect tumor heterogeneity and overcome tissue-related constraints.

Liquid biopsy has emerged as an innovative alternative in precision oncology, offering a minimally invasive, repeatable, and rapid method for genomic profiling (6). Compared to tissue genotyping, liquid biopsy typically have shorter turnaround times (3–10 vs. 10–20 days) (7), facilitating prompt treatment decisions. Liquid biopsy involves the analysis of analytes such as RNA, DNA, or circulating tumor cells in blood or body fluids, with DNA being the most widely adopted (8-10). Circulating tumor DNA (ctDNA) shed into the bloodstream from tumor cells, reflects the mutational landscape of both primary and metastatic lesions, thereby offering a comprehensive snapshot of tumor heterogeneity (11). Although the amount of DNA circulating in the blood is minimal, advances in sequencing technologies have significantly enhanced detection sensitivity and specificity. Liquid biopsy holds great potential for diverse clinical applications, including early cancer detection, disease monitoring, minimal residual disease (MRD) assessment, and the identification of resistance mechanisms, while also serving as a versatile platform for comprehensive genomic profiling in precision oncology. In 2016, the U.S. Food and Drug Administration (FDA) approved the first ctDNA-based liquid biopsy test (cobas EGFR Mutation Test v2) for plasma samples, targeting EGFR exon 19 deletions and exon 21 L858R mutations (12). Since then, several next-generation sequencing (NGS)-based ctDNA assays, including Guardant360 CDx and FoundationOne Liquid CDx, have also received FDA approval as companion diagnostics (13).

Despite its promise, several aspects of ctDNA biology and clinical utility remain to be elucidated. In this study, we performed plasma-based ctDNA analysis in Korean patients with recurrent or metastatic NSCLC to characterize the observed mutation profiles and evaluate its clinical usefulness, particularly when tissue testing was infeasible. We also explored potential reasons for discordance between tissue and plasma results and identified cases in which ctDNA-only actionable mutations provided additional treatment opportunities. We present this article in accordance with the STROBE reporting checklist (available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2025-715/rc).

Methods

Study design

This was a retrospective, multicenter observational study conducted as part of the K-MASTER project in South Korea. The K-MASTER project is a nationwide precision oncology initiative supported by the Korean Ministry of Health and Welfare, which performs comprehensive genomic profiling of solid tumors to enable molecularly guided therapy and facilitate clinical trial enrollment. Between June 2017 and December 2021, patients with recurrent or metastatic NSCLC were referred from 54 participating centers; among these, 132 patients from ten sites underwent plasma-based ctDNA testing and were included in this analysis. As this was a retrospective study, no formal sample size calculation was performed, and all eligible patients who underwent plasma ctDNA testing during the study period were included.

Patients

Patients with pathologically confirmed recurrent or metastatic NSCLC were eligible. Among 822 patients referred to the K-MASTER project, those who underwent plasma-based ctDNA analysis were selected for this study. The decision to perform ctDNA analysis was based on the unavailability of tissue specimens or as part of clinical decision-making. Clinical data, including tissue genotyping results, follow-up treatment, and survival status, were obtained from the K-MASTER database. To minimize selection bias, we included all patients who underwent plasma-based ctDNA analysis during the study period, regardless of clinical indication. However, as patients were selected based on clinical decision-making or lack of tissue availability, selection bias cannot be entirely excluded.

Targeted sequencing

Targeted sequencing was performed using the Axen Cancer Panel 1 which includes the exons of 88 cancer-related genes and the intronic regions of 3 genes (14). Actionable mutations were defined as mutations which can impact therapy selection according to the NCCN guideline version (10.2023) (15). Detected DNA alterations were annotated as pathogenic mutation in the case of “likely oncogenic” or “oncogenic” for solid tumors in the OncoKB database. Maximum variant allele frequency (maxVAF) is defined as the highest percentage of VAF among actionable mutations or pathogenic mutations. Details of the targeted sequencing are described in supplementary file (Appendix 1).

Statistical analysis

Descriptive statistics were used to report patient characteristics as well as mutation types and frequencies. Quantitative data are represented as median and range. Survival curves were generated using the Kaplan-Meier method and assessed with log-rank test. Overall survival (OS) was defined as the period from the first diagnosis of recurrent or metastatic NSCLC until death. Cox proportional hazards models were used to define the association of genomic profiles with OS, after adjusting for tumor histology and known prognostic factors, including smoking status, metastasis status, age, sex and previous treatment. Missing data were not imputed and were excluded from analyses where applicable. All analyses were performed using R studio (version 4.2.0).

Ethical statement

The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Institutional Review Board of Korea University Anam Hospital, Seoul, Korea (2017AN0401) as an institution of a principal investigator and all participant sites were also approved by the Institutional Review Board of their institutions. All patients provided written informed consent for this study.

Results

Patient characteristics

Between March 2018 and December 2021, a total of 132 patients with recurrent or metastatic NSCLC underwent ctDNA analysis. All patients who received ctDNA testing during the study period were included. Baseline characteristics are summarized in Table 1. The cohort included were 77 men (58.3%) and 55 women (41.7%), with a median age of 66 years (range, 37–101 years). Smoking status was documented for 75 patients, with 32 identified as current or former smokers and 43 as non-smokers. Adenocarcinoma accounted for the majority of cases (75.8%), followed by squamous cell carcinoma (15.2%), and other histologic subtypes (9.0%). At the time of ctDNA testing, 64 patients (48.5%) were treatment-naive, while 68 (51.5%) had received prior treatment. Intrathoracic metastases were observed in 41 patients, while 88 patients exhibited extrathoracic dissemination.

Mutational profile

The mutational landscape of plasma ctDNA is illustrated in Figure 1. The most frequently altered gene was TP53 (56%), followed by EGFR (30%), ALK (16%), RET (12%), ERBB2 (11%) and KRAS (7%). Notably, 9 patients (6.8%) exhibited no detectable nonsynonymous mutations (Table 2). Pathogenic mutations were identified in 68 patients (51.5%). The MaxVAF of these pathogenic mutations ranged from 0 to 59.5 %, with a median of 0.1%. The highest MaxVAF was observed in a patient with an EGFR L858R mutation accompanied by EGFR amplification (copy number =5). Among the 33 EGFR mutations identified, 29 were classified as actionable. Four EGFR mutations (3 EGFR G824D and 1 R252H) had unknown oncogenic significance. One patient exhibited EGFR copy number gain. Further detailed information regarding EGFR mutations is shown in Table S1.

Figure 1 Landscape of somatic mutations detected in plasma ctDNA. Mutational landscape of the most frequently altered genes detected by plasma ctDNA analysis. The top panel shows number of mutations for individual samples. Mutation types are color-coded as follows: missense mutation (blue), nonsense mutation (red), in-frame insertion/deletion (yellow), frameshift insertion/deletion (orange), and multiple hits (purple). Frequencies of gene alterations are shown on the right side of the plot. ctDNA, circulating tumor DNA.

Figure 2 compares the distribution of actionable genomic alterations identified by tissue-based diagnostics and plasma ctDNA sequencing. Across the cohort, actionable alterations were detected in 42 of 143 patients (31.8%) by plasma ctDNA alone and in 57 of 132 patients (43.2%) when integrating both tissue and plasma results. Figure 3 summarizes the concordance between tissue-based genotyping and plasma ctDNA analysis for actionable mutations, illustrating cases with concordant, discordant, or unavailable tissue results. Among them, 27 actionable mutations were concordantly detected in both tissue and plasma, while 15 were detected exclusively in tissue (Table S2), and 15 exclusively in plasma (Table S3).

Figure 2 Comparison of genomic alterations detected by tissue-based conventional diagnostics and plasma ctDNA sequencing. Distribution of actionable genomic alterations identified by tissue biopsy, plasma ctDNA analysis, or plasma-only detection among patients with recurrent or metastatic NSCLC. Green bars represent actionable mutations detected in tissue samples, red bars represent actionable mutations detected in ctDNA, and purple bars represent actionable mutations detected exclusively in ctDNA. EGFR mutations were the most frequently observed, followed by ALK fusion, BRAF V600E, and ROS1 fusion events. The category “Others” included KRAS G12C, RET fusion, and MET amplification. ctDNA, circulating tumor DNA; NSCLC, non-small-cell lung cancer.

Figure 3 Concordance of actionable mutation detection between tissue and plasma analyses. Concordance rates of actionable mutation detection in tissue and plasma ctDNA among patients with NSCLC. “No actionable mutations” refers to patients without detectable actionable alterations in either tissue or plasma. “Tissue only” indicates actionable mutations detected exclusively in tissue samples, “Both tissue and plasma” represents concordant detection, and “Plasma only” denotes actionable mutations detected exclusively in plasma ctDNA. ctDNA, circulating tumor DNA; NSCLC, non-small-cell lung cancer.

Notable co-occurring alterations

Three patients harbored co-occurring actionable alterations

Case 1 (S0113)

Concurrent KRAS G12C (VAF 2.0%) and EGFR exon 19 deletion (VAF 0.4%) were identified in ctDNA. The patient progressed on gefitinib after only 139 days. Given the typically mutual exclusivity of KRAS and EGFR mutations in NSCLC, and considering the low VAF of the EGFR variant and short duration of response, several possibilities can be considered—including a second primary lung cancer, intratumoral heterogeneity, or a potential false-positive ctDNA finding.

Case 2 (S0096)

An ALK fusion was detected by immunohistochemistry (D5F3 CDx assay), alongside an EGFR exon 20 insertion (VAF 29.4%) identified in ctDNA. The patient, who initially presented with brain, adrenal, liver, and bone metastases, received alectinib but unfortunately succumbed after 26 days of therapy. Considering that ALK-positive patients generally exhibit robust responses to alectinib, the rapid fatal course may have been driven by the EGFR exon 20 insertion-positive tumor. In this sense, ctDNA analysis was instrumental in identifying the critical alteration underlying the patient’s poor outcome.

Case 3 (S0056)

EGFR T790M was detected in tissue, while BRAF V600E (VAF 0.3%) was identified in ctDNA. The patient had previously received erlotinib and subsequently received osimertinib, achieving a partial response but progressing after a progression-free survival (PFS) of 5.4 months. Although concurrent EGFR T790M and BRAF V600E detection raises the possibility of multiple resistance mechanisms, this seems less likely. Given the very low allele frequency of BRAF V600E and the partial response to osimertinib, the ctDNA finding is more consistent with a false-positive result or subclonal event than a true oncogenic driver.

Mutational discordance between plasma and tissue

A total of 15 cases exhibited actionable mutations that were detected exclusively by plasma ctDNA. Among these, 12 lacked corresponding tissue genotyping data, and 3 showed discordant results—positive in plasma but negative in tissue. Of these 15 patients, seven patients subsequently received targeted therapy based on ctDNA results. Notably, in the three discordant cases, tissue biopsies had been obtained from bone [1], liver [1], and lymph node [1]. In one bone-derived sample (S0091), EGFR PCR quality control failed, whereas in another (S0113), QC passed but EGFR mutation was not detected in tissue despite being identified in plasma. These findings underscore the challenges of molecular testing in bone-derived specimens, where decalcification and sample degradation frequently compromise DNA quality and sequencing performance. Beyond these technical limitations, biological factors such as spatial or temporal tumor heterogeneity—particularly when tissue is obtained from metastatic rather than primary lesions—may also contribute to these discrepancies. Although false-positive ctDNA results cannot be entirely excluded, these findings emphasize that plasma-based testing can provide complementary genomic information when tissue analysis is limited by sample quality or site of biopsy.

Among the 15 tissue-only mutation cases, ten patients harbored EGFR mutations, while four had gene fusions (ALK, n=2; ROS1, n=2) and one had a BRAF V600E mutation. Of the ten patients with EGFR mutations, four had previously received EGFR TKI therapy, and six were treatment-naïve. In the four patients who had received prior EGFR TKI therapy, the absence of EGFR mutations in ctDNA may be partly explained by molecular clearance following treatment. Moreover, the discordance may have been influenced by the temporal gap between tissue biopsy and ctDNA testing. The remaining six treatment- naïve cases likely represent true false-negatives in ctDNA analysis. Notably, gene fusion events—specifically the two ALK fusions and two ROS1 fusions—are known to be less reliably detected by ctDNA platforms, which may explain their absence in plasma. Among these 15 tissue-only cases, seven patients had intrathoracic disease and eight exhibited extrathoracic metastases. Interestingly, even among patients with disseminated disease involving sites such as bone, brain, and liver, some cases demonstrated true false-negative results in ctDNA analysis, with actionable mutations identified in tissue but not detected in plasma. This suggests that disease burden or metastatic location alone may not be sufficient to predict ctDNA detectability.

Effect of ctDNA-guided therapy on survival outcomes

For survival analysis, patients were categorized into three groups based on the receipt of targeted therapy: no targeted therapy, tissue-based targeted therapy, and ctDNA-based targeted therapy. Kaplan-Meier survival analysis revealed significant difference in OS among these groups (P=0.03) (Figure 4). Patients who received tissue-based targeted therapy showed the most favorable outcomes, followed by those treated according to ctDNA-based results. These findings suggest that actionable mutation-guided targeted therapy, whether derived from tissue or plasma genomic profiling, may confer a survival advantage in patients with recurrent or metastatic NSCLC.

Figure 4 Kaplan-Meier analysis of overall survival by treatment group: tissue-based targeted therapy, ctDNA-based targeted therapy, and no targeted therapy. Patients receiving tissue-based targeted therapy exhibited the best survival outcomes, followed by the ctDNA-based targeted therapy group and the no targeted therapy group (P=0.03). ctDNA, circulating tumor DNA.

Discussion

This study aimed to explore the practical applicability of plasma-based ctDNA analysis in Korean patients with recurrent or metastatic NSCLC, particularly in scenarios where tissue biopsy was unavailable or inadequate. Our findings indicate that ctDNA testing can identify actionable genomic alterations that are often undetectable or inaccessible through conventional tissue analysis, owing to sampling limitations, inadequate specimens, or tumor heterogeneity. Notably, actionable mutations were identified in 43.2% of patients when combining tissue and plasma results, including 11.4% detected exclusively by ctDNA. Several of these plasma-only findings directly informed targeted therapy decisions, underscoring the clinical value of ctDNA in expanding treatment opportunities for patients otherwise ineligible for molecularly guided therapy (4,5).

To elucidate the potential causes of plasma-only mutation detection, we further reviewed this subset of patients. First, of the three plasma-positive, tissue-negative cases, biopsy sites included bone (n=1), lymph node (n=1), and liver (n=1). Bone biopsies are particularly susceptible to limitations, including reduced DNA yield and quality degradation during decalcification processing, which can compromise the sensitivity of NGS (16,17). Furthermore, liver metastases—another frequent biopsy site in our cohort—are known to exhibit distinct clonal architectures compared to primary tumors, further contributing to sampling bias (18,19). Although the primary tumor generally provides the most representative molecular profile of the dominant clone, obtaining tissue from the primary site is often unfeasible in advanced-stage disease due to anatomical inaccessibility or procedural risk. Consequently, biopsies are frequently performed from metastatic lesions, as in our cases, which may harbor distinct subclonal architectures shaped by site-specific selective pressures. This highlights the inherent limitation of relying solely on tissue biopsies for comprehensive genomic profiling and underscores the complementary value of ctDNA analysis, which can integrate genomic signals from multiple tumor sites and more accurately reflect overall tumor heterogeneity. Second, nearly half of the plasma-only mutation group (43.7%) had received prior systemic therapy before undergoing ctDNA testing. This raises the possibility that clonal evolution under therapeutic pressure may have led to the emergence of resistant subclones, which were subsequently detected in plasma but not in archival tissue specimens (20,21). These observations illustrate the dynamic nature of tumor genomics and highlight the role of ctDNA as a practical tool for monitoring molecular evolution when repeat biopsies are infeasible.

The integration of plasma ctDNA analysis into routine clinical practice faces several challenges. Technically, assay sensitivity, false-positive rates, and variant interpretation require ongoing refinement (7,8). In our study, several gene fusion events (e.g., ALK and ROS1) were not detected in plasma, consistent with prior observations that structural variants are less reliably captured by ctDNA platforms (22). Emerging approaches, including circulating tumor RNA (ctRNA) analysis, hybrid-capture-based enrichment, and long-read sequencing technologies, may improve the sensitivity for fusions and copy number variants (CNVs) (23,24). Additionally, EGFR mutations were undetectable in several treatment-naïve patients, including those with extrathoracic metastases, suggesting that biological and technical limitations may contribute to false-negative ctDNA results even in the presence of substantial disease burden. Despite stringent quality control measures, the possibility of detecting clonal hematopoiesis of indeterminate potential (CHIP)-related mutations in plasma samples remains a confounding factor that warrants careful consideration (8). Importantly, histologic/cellular transformations in lung cancer (e.g., small-cell or squamous transformation) are fundamentally phenotypic shifts that ctDNA cannot capture, highlighting that tissue biopsy remains indispensable for confirming such transformations and guiding subsequent therapy. Beyond technical and biological constraints, successful clinical implementation also requires standardized workflows to ensure timely sample handling and reporting, as well as evaluation of cost-effectiveness within diverse healthcare systems (6,7,9).

Several limitations of this study should be acknowledged. The relatively modest sample size and heterogeneity in clinical and treatment characteristics may limit the generalizability of our findings. The decision to perform ctDNA testing was primarily driven by the unavailability or inadequacy of tissue samples and physician discretion in clinical decision-making, rather than as part of routine molecular profiling. Consequently, patients with known EGFR mutations confirmed by tissue testing were less likely to undergo ctDNA analysis, which may partially explain the relatively lower frequency of EGFR alterations compared with other real-world cohorts. In addition, because this study exclusively enrolled Korean patients, ethnic and epidemiologic differences in driver mutation prevalence may have contributed to the observed disproportion between EGFR/ALK and KRAS alterations. Collectively, these factors—and the retrospective nature of data collection—indicate an inherent selection bias that should be considered when interpreting our results. Furthermore, because the study was conducted during a period when certain targeted therapies were not yet approved or widely available in Korea, the full therapeutic potential of identified mutations may not have been fully realized (10). In addition, the survival analysis was exploratory and included a limited number of patients in the ctDNA-guided therapy group; therefore, the results should be interpreted with caution and are not intended to suggest any prognostic implications. Despite these limitations, our study offers practical insights into the use of plasma ctDNA analysis for the management of advanced NSCLC. Particularly in settings where tissue biopsy is challenging or infeasible, ctDNA offers a valuable alternative for identifying actionable mutations and guiding treatment decisions (5,7,10). Moreover, as additional targeted therapies and molecularly guided treatment strategies continue to emerge, the clinical utility of plasma ctDNA testing is likely to expand.

Despite these limitations, our multi-center Korean cohort provides real-world evidence supporting the clinical feasibility of ctDNA analysis for identifying actionable mutations when tissue biopsy is unavailable or inadequate. Our findings highlight plasma–tissue discordance and its clinical implications, reinforcing the complementary role of ctDNA testing in precision oncology.

Conclusions

In conclusion, the integration of plasma-based ctDNA analysis into the diagnostic and therapeutic continuum of NSCLC patients enhances the precision of genomic profiling, broadens access to targeted therapies, and holds the potential to improve clinical outcomes. Continued research efforts to refine ctDNA technologies, standardize workflows, and validate clinical utility will be essential to fully harness the transformative potential of liquid biopsy in the next generation of oncology care (5,6,9). Future prospective studies incorporating serial plasma sampling, orthogonal validation, and functional characterization of detected alterations will be crucial to further elucidate the true clinical relevance of plasma-only findings and to guide evidence-based implementation into routine practice.

Acknowledgments

We extend our sincere gratitude to the patients nationwide who willingly participated in the K-MASTER program, as well as to our dedicated colleagues, researchers, nurses, and medical professionals whose invaluable support and cooperation were essential to the success of this study.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2025-715/rc

Data Sharing Statement: Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2025-715/dss

Peer Review File: Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2025-715/prf

Funding: This study was supported by a grant from the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health & Welfare, Republic of Korea (grant No. HI17C2206), and by a grant from Korea University Anam Hospital, Seoul, Republic of Korea (grant Nos. O2310951, O2207671).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2025-715/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Institutional Review Board of Korea University Anam Hospital, Seoul, Republic of Korea (2017AN0401) as an institution of a principal investigator and all participant sites were also approved by the Institutional Review Board of their institutions. All patients provided written informed consent for this study.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. World Health Organization. Estimated number of new cases in 2020, worldwide, both sexes, all ages. 2023 (cited 2023 Oct 5). Available online: https://gco.iarc.fr/today/online-analysis-table?v=2020&mode=cancer&mode_population=continents&population=900&populations=900&key=asr&sex=0&cancer=39&type=0&statistic=5&prevalence=0&population_group=0&ages_group%5B%5D=0&ages_group%5B%5D=17&group_cancer=1&include_nmsc=0&include_nmsc_other=1
2. Howlader N, Forjaz G, Mooradian MJ, et al. The Effect of Advances in Lung-Cancer Treatment on Population Mortality. N Engl J Med 2020;383:640-9. [Crossref] [PubMed]
3. Schram AM, Goto K, Kim DW, et al. Efficacy of Zenocutuzumab in NRG1 Fusion-Positive Cancer. N Engl J Med 2025;392:566-76. [Crossref] [PubMed]
4. Li BT, Smit EF, Goto Y, et al. Trastuzumab Deruxtecan in HER2-Mutant Non-Small-Cell Lung Cancer. N Engl J Med 2022;386:241-51. [Crossref] [PubMed]
5. Mosele F, Remon J, Mateo J, et al. Recommendations for the use of next-generation sequencing (NGS) for patients with metastatic cancers: a report from the ESMO Precision Medicine Working Group. Ann Oncol 2020;31:1491-505. [Crossref] [PubMed]
6. Baldacchino S. Current Advances in Clinical Application of Liquid Biopsy. In: Strumfa I, Bahs G. editors. Pathology - From Classics to Innovations. IntechOpen: Rijeka; 2021.
7. Sacher AG, Paweletz C, Dahlberg SE, et al. Prospective Validation of Rapid Plasma Genotyping for the Detection of EGFR and KRAS Mutations in Advanced Lung Cancer. JAMA Oncol 2016;2:1014-22. [Crossref] [PubMed]
8. Corcoran RB, Chabner BA. Application of Cell-free DNA Analysis to Cancer Treatment. N Engl J Med 2018;379:1754-65. [Crossref] [PubMed]
9. Husain H, Velculescu VE. Cancer DNA in the Circulation: The Liquid Biopsy. JAMA 2017;318:1272-4. [Crossref] [PubMed]
10. Neumann MHD, Bender S, Krahn T, et al. ctDNA and CTCs in Liquid Biopsy - Current Status and Where We Need to Progress. Comput Struct Biotechnol J 2018;16:190-5. [Crossref] [PubMed]
11. Kustanovich A, Schwartz R, Peretz T, et al. Life and death of circulating cell-free DNA. Cancer Biol Ther 2019;20:1057-67. [Crossref] [PubMed]
12. Kwapisz D. The first liquid biopsy test approved. Is it a new era of mutation testing for non-small cell lung cancer? Ann Transl Med 2017;5:46. [Crossref] [PubMed]
13. Nikanjam M, Kato S, Kurzrock R. Liquid biopsy: current technology and clinical applications. J Hematol Oncol 2022;15:131. [Crossref] [PubMed]
14. Lee Y, Lee S, Sung JS, et al. Clinical Application of Targeted Deep Sequencing in Metastatic Colorectal Cancer Patients: Actionable Genomic Alteration in K-MASTER Project. Cancer Res Treat 2021;53:123-30. [Crossref] [PubMed]
15. National Comprehensive Cancer Network. Non-Small Cell Lung Cancer (Version 3.2023). 2023 (cited 2023 Oct 05). Available online: https://www.nccn.org/guidelines/guidelines-detail?category=1&id=1450
16. Sailer V, Schiffman MH, Kossai M, et al. Bone biopsy protocol for advanced prostate cancer in the era of precision medicine. Cancer 2018;124:1008-15. [Crossref] [PubMed]
17. Zheng G, Lin MT, Lokhandwala PM, et al. Clinical mutational profiling of bone metastases of lung and colon carcinoma and malignant melanoma using next-generation sequencing. Cancer Cytopathol 2016;124:744-53. [Crossref] [PubMed]
18. Burrell RA, McGranahan N, Bartek J, et al. The causes and consequences of genetic heterogeneity in cancer evolution. Nature 2013;501:338-45. [Crossref] [PubMed]
19. Dagogo-Jack I, Shaw AT. Tumour heterogeneity and resistance to cancer therapies. Nat Rev Clin Oncol 2018;15:81-94. [Crossref] [PubMed]
20. Lim SM, Yang SD, Lim S, et al. Molecular landscape of osimertinib resistance in patients and patient-derived preclinical models. Ther Adv Med Oncol 2022;14:17588359221079125. [Crossref] [PubMed]
21. Del Re M, Crucitta S, Gianfilippo G, et al. Understanding the Mechanisms of Resistance in EGFR-Positive NSCLC: From Tissue to Liquid Biopsy to Guide Treatment Strategy. Int J Mol Sci 2019;20:3951. [Crossref] [PubMed]
22. Esagian SM, Grigoriadou GI, Nikas IP, et al. Comparison of liquid-based to tissue-based biopsy analysis by targeted next generation sequencing in advanced non-small cell lung cancer: a comprehensive systematic review. J Cancer Res Clin Oncol 2020;146:2051-66. [Crossref] [PubMed]
23. Heyer EE, Deveson IW, Wooi D, et al. Diagnosis of fusion genes using targeted RNA sequencing. Nat Commun 2019;10:1388. [Crossref] [PubMed]
24. Dorney R, Dhungel BP, Rasko JEJ, et al. Recent advances in cancer fusion transcript detection. Brief Bioinform 2023;24:bbac519. [Crossref] [PubMed]