Seoyoung Lee; Chaeyeon Kim; Chang Gon Kim; Min Hee Hong; Mina Han; Wonrak Son; Gamin Kim; Hyeong Jung Woo; Hyun Young Shin; Jungmin Lee; Minseok S Kim; Hye Ryun Kim

doi:10.4143/crt.2025.672

Articles

Page Path: HOME > J Korean Cancer Assoc > Ahead-of print articles > Article

Original Article Circulating Tumor Cell–Based Molecular Responses Stratify EGFR-TKI Efficacy in Patients with EGFR-Mutant Lung Cancer: Seoyoung Lee¹, Chaeyeon Kim², Chang Gon Kim³, Min Hee Hong³, Mina Han², Wonrak Son², Gamin Kim⁴, Hyeong Jung Woo⁵, Hyun Young Shin⁶, Jungmin Lee⁶, Minseok S Kim^5,6, Hye Ryun Kim^3,7; DOI: https://doi.org/10.4143/crt.2025.672
Published online: January 27, 2026

¹Division of Medical Oncology, Department of Internal Medicine, Gangnam Severance Hospital, Yonsei University College of Medicine, Seoul, Korea

²Department of Internal Medicine, Yonsei University College of Medicine, Seoul, Korea

³Division of Medical Oncology, Department of Internal Medicine, Yonsei Cancer Center, Yonsei University College of Medicine, Seoul, Korea

⁴Department of Oncology, Yonsei University College of Medicine, Seoul, Korea

⁵Department of New Biology, Daegu Gyeongbuk Institute of Science and Technology, Daegu, Korea

⁶CTCELLS Inc., Seoul, Korea

⁷Department of Internal Medicine, Graduate School of Medical Science, Brain Korea 21 Project, Yonsei University College of Medicine, Seoul, Korea

Correspondence: Minseok S Kim, Department of New Biology, Daegu Gyeongbuk Institute of Science and Technology, Daegu 42988, Korea
Tel: 82-70-4422-2909 E-mail: kms@dgist.ac.kr

Co-correspondence: Hye Ryun Kim, Division of Medical Oncology, Department of Internal Medicine, Yonsei Cancer Center, Yonsei University College of Medicine, Seoul 03722, Korea
Tel: 82-2-2228-8130 E-mail: nobelg@yuhs.ac

^*Seoyoung Lee and Chaeyeon Kim equally contributed to this work.

• Received: June 29, 2025 • Accepted: January 16, 2026

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

1,331 Views
118 Download

Full Article

Download PDF

Abstract

Purpose
Circulating tumor cell (CTC) is a promising minimally invasive biomarker for epidermal growth factor receptor (EGFR)–mutant non–small cell lung cancer (NSCLC). However, the rarity of CTCs and limitations in their isolation and molecular characterization hinder their clinical utility, particularly in predicting treatment outcomes. This study evaluates the potential of CTC molecular response to predict treatment efficacy and guide therapy in patients with EGFR-mutant NSCLC undergoing EGFR–tyrosine kinase inhibitors (TKI) therapy.
Materials and Methods
Seventy-seven patients with EGFR-mutant NSCLC treated with EGFR-TKIs were enrolled. CTCs were isolated using continuous centrifugal microfluidic technology (Continuous Centrifugal Microfluidics–Circulating Tumor Cell Disc [CCM-CTCD]) and compared with circulating tumor DNA (ctDNA) and tissue biopsy for EGFR mutation analysis. Patients were categorized as CTC molecular responders or non-responders based on a ≥ 44.4% reduction in CTC count from baseline. Progression-free survival (PFS) and tumor burden changes were evaluated.
Results
CTC responders had significantly longer PFS (46.3 vs. 13.6 months, p=0.007) and greater tumor burden reduction (–37.7% vs. –35.2%, p=0.218) compared to non-responders. The CCM-CTCD demonstrated concordance with the cobas test while exhibiting higher sensitivity for EGFR mutation detection among 46 patients who underwent both tests simultaneously. Mutational discordance among tissue, ctDNA, and CTCs highlighted tumor heterogeneity. CTC profiling complemented traditional methods for identifying genomic alterations and predicting early progression.
Conclusion
CTC analysis using CCM-CTCD shows potential as a biomarker for predicting treatment response and prognosis in EGFR-mutant NSCLC. Stratification by CTC molecular response may inform risk-adapted treatment; however, its clinical utility remains to be established. Prospective studies are warranted to validate these findings and determine the role of CTC-guided decision-making.
Key words: Circulating tumor cells, EGFR-mutant lung cancer, EGFR-tyrosine kinase inhibitors, Treatment response, Non-small cell lung cancer, Liquid biopsy

Introduction

Lung cancer is one of the most prevalent solid tumors worldwide and remains the leading cause of cancer-related death [1]. Its development is associated with multiple driver oncogenic alterations, among which mutations in the epidermal growth factor receptor (EGFR) gene represent a key actionable target. These mutations are detected in approximately 10%-15% of Western and over 40% of Asian patients with non–small cell lung cancer (NSCLC) [2]. EGFR tyrosine kinase inhibitors (TKIs) have become the standard of care for treating advanced EGFR-mutant NSCLC [3,4]. Third-generation EGFR-TKI, including osimertinib and lazertinib, are highly recommended as first-line therapies due to their demonstrated ability to improve survival, superior penetration of the central nervous system, and efficacy against the T790M resistance mutation [5,6]. Additionally, combining EGFR-TKIs with cytotoxic chemotherapy or other targeted agents has been extensively studied in clinical trials and has recently been approved as an alternative treatment option for patients [7,8].

Understanding resistance mechanisms in oncogene-addicted NSCLC is crucial for optimizing subsequent treatment strategies and improving patient outcomes [9]. Liquid biopsy offers a minimally invasive alternative to tissue biopsy, enabling therapeutic monitoring, precision medicine-based treatment decisions, and detection of tumor heterogeneity by analyzing tumor-derived components in body fluids [10]. U.S. Food and Drug Administration (FDA)–approved liquid biopsy tests, including the cobas EGFR mutation test and plasma-based next-generation sequencing assays such as Guardant360 and FoundationOne Liquid CDx, are widely utilized in clinical practice for circulating tumor DNA (ctDNA) detection [11-15].

Circulating tumor cells (CTCs) have the potential to provide a more comprehensive genomic and transcriptomic profile of cancer than ctDNA [16]. Quantitative analysis of CTC has shown their potential as predictive markers for prognosis and treatment response in various cancer types, including breast, colon, and prostate cancer [17-19]. Molecular characterization of CTCs can reveal resistant clones, serve as a surrogate marker for tumor heterogeneity, and guide subsequent treatment decisions [20,21]. Analyzing CTCs avoids the challenges associated with clonal hematopoiesis of indeterminate potential and germline mutation, which can arise when detecting ctDNA [22].

Despite their potential, detecting CTCs in peripheral blood is challenging due to their rarity and the limitations of marker-dependent and marker-independent isolation techniques, which may compromise sensitivity and specificity [23-30]. To address these challenges, advanced, automated, and unbiased methods are needed to accurately capture the heterogeneity of CTCs and enhance their utility as prognostic biomarkers.

The Continuous Centrifugal Microfluidics–Circulating Tumor Cell Disc (CCM-CTCD) enriches heterogeneous CTCs through automated microfluidic extraction and leukocyte depletion without interrupting the centrifugation process. The CTCeptor enables the detection of various CTCs, irrespective of their size, protein expression, or cancer type. Previous studies have explored the clinical validity of CTCeptor in patients with NSCLC, which suggested a correlation between CTC enumeration and treatment response [31].

Building on this foundation, this study explores CTC molecular response in patients with EGFR-mutant lung cancer, particularly in relation to treatment outcomes. By evaluating different methods for isolating and analyzing CTCs or ctDNAs, the research seeks to understand how CTC-based biomarkers can predict treatment efficacy, guide therapy decisions, and support strategies for overcoming resistance to third-generation EGFR-TKIs.

Materials and Methods

1. Patient selection and enrollment

This study prospectively enrolled patients with advanced NSCLC harboring EGFR mutations who were undergoing treatment with EGFR-TKI. The EGFR mutation status of patients had been confirmed before enrollment through the analysis of fresh tissue biopsy or archival tissue as a part of clinical practice. Demographic and clinical data, including age, sex, smoking history, treatment regimens, prior treatments, and treatment outcomes, were collected prospectively and recorded in an electronic medical record system.

2. Sample collection

Peripheral blood samples (10 mL each) were collected from 77 patients at the Yonsei Cancer Center before and after the initiation of EGFR-TKI treatment. Baseline blood samples (T0) were obtained within 14 days prior to the first EGFR-TKI dose. Follow-up samples (T1) were collected at the time of response evaluation (the first or a subsequent response evaluation after TKI initiation; median, 4 weeks; interquartile range [IQR], 3 to 11 weeks).

3. CTC isolation and analysis

The CTC enrichment process using the CCM-CTCD system (CTCeptor) followed the methodology described previously [31]. The CTCD-DISC and CTCD-REAGENT–density gradient media (DGM) were manufactured and supplied by CTCELLS Corporation. Briefly, the DGM corresponding to the hematocrit level was injected into the BLOOD chamber inlet, while 0.875 mL of DGM was injected into the DEPLETION chamber inlet. The spin phase was then executed according to the CCM-CTCD system protocol. Next, 3 mL of blood was injected into the initial chamber inlet, and the RBC removal phase was performed following the CCM-CTCD protocol. Subsequently, 50 μL of Dynabeads CD45 (Invitrogen) and 50 μL of Dulbecco’s phosphate-buffered saline (DPBS) (–/–) (Welgene) were injected into the MIXING chamber, followed by the execution of the white blood cell removal phase following the CCM-CTCD system. After placing a magnet at the bottom of the CTC chamber and incubating for 10 minutes, the fluid from the CTC chamber was recovered into a 15 mL conical tube. The recovered fluid containing CTCs was washed with DPBS and centrifuged at 380 ×g for 5 minutes to obtain a pellet. CTC samples obtained using the CCM-CTCD method were processed in two parallel assays: droplet digital polymerase chain reaction (ddPCR) for EGFR mutation testing and immunocytochemistry (ICC) for CTC enumeration. Each blood sample was evenly divided into two aliquots, one for ICC-based enumeration and the other for ddPCR-based molecular analysis. Accordingly, a CTC count of zero indicates that no morphologically intact CTCs were identified in the enumeration aliquot, whereas EGFR mutations could still be detected in the separate molecular aliquot.

After isolation of CTC, ICC was performed in hydrogel pellicle ImmunoCytoChemistry (HypICC) system (supplied by CTCELLS Corporation) as previously described [32]. Cells were fixed in 16% paraformaldehyde (Pierce 16% formaldehyde (w/v), methanol-free, Thermo Fisher Scientific) for 10 minutes at room temperature and stored at 4°C. The ICC chip was cleaned with Isopropyl alcohol. The DPBS, poly(ethylene glycol) diacrylate (PEGDA) (average Mn 700, Sigma-Aldrich), and 2-hydroxy-2-methylpropiophenone (Photoinitiator, Sigma-Aldrich) mixtures were filtered with a 0.22 μm filter. A 270 μL volume of filtered 20% PEGDA was applied to the ICC chips. The collected cells were centrifuged at 1,500 rpm for 5 minutes to obtain a pellet. The pellet was resuspended, and 870 μL of the cell suspension was loaded onto the ICC chips using a pipette coated with 2% bovine serum albumin (BSA) (Sigma-Aldrich). The ICC chip was centrifuged at 3,000 rpm for 3 minutes, then exposed to 75% UV for 30 seconds to promote hardening. The cells in the ICC chip were washed three times with DPBS, then permeabilized with 0.2% Triton X-100 (Sigma-Aldrich) for 10 minutes at room temperature. Afterwards, the cells were blocked with 2% BSA for 1 hour at room temperature. The cells were incubated overnight at 4°C with a mixture of antibodies: Pan Cytokeratin Monoclonal Antibody (pan reactive) (Alexa Fluor 488) (BioLegend), and anti-human CD45 antibody (Alexa Fluor 594) (BioLegend). After three washes with 0.05% Triton X-100, the cells were incubated with Hoechst 33342 (Thermo Fisher Scientific) or DAPI (Abcam) for 20 minutes.

Following ICC staining, CTCs were enumerated. Images were obtained using an LSM710 confocal microscope (Carl Zeiss), and sequential images were captured after fluorescent staining. Multiple frames were taken for subsequent analysis using image analysis software. CTCs were counted using the ZEN 3.5 software of the Zeiss microscope. CTCs were identified based on the staining profile: [PanCK (+)/CD45 (–)/DAPI/Hoechst (+)]. The results were reported as the number of CTCs per 3 mL of whole blood.

4. DNA extraction and EGFR mutation detection for CTC, plasma, and tissue

Blood samples were collected from patients with EGFR mutation, and plasma was isolated by centrifugation using the CCM-CTCD system. The supernatant was collected and stored in an Eppendorf tube at –80°C. ctDNA was extracted using the QIAmp Circulating Nucleic Acid Kit (Qiagen) according to the manufacturer’s protocol. The genomic DNA (gDNA) of CTC was isolated using the DNeasy Blood & Tissue Kit (Qiagen) following the manufacturer’s protocol. ddPCR was performed using the Bio-Rad QX200 ddPCR system to validate EGFR exon 19 deletion (E19del), T790M, or L858R mutations in plasma ctDNA and CTC gDNA. For ddPCR, droplets were generated in a DG8 Cartridge (Bio-Rad), which was preloaded with 20 μL of the sample and 70 μL of droplet generation oil. The droplets were then transferred to a 96-well plate and sealed with a PX1 PCR Plate Sealer (Bio-Rad). The thermal cycler was programmed as follows: 96°C for 10 minutes, followed by 40 cycles of 94°C for 30 seconds and 60°C for 60 seconds, then 98°C for 10 minutes. The droplets containing amplicons were quantified using the QX200 Droplet Reader and analyzed with the QuantaSoft software package (Bio-Rad).

The EGFR mutation status of tumor tissue was determined prior to patient enrollment and outside the scope of the clinical study using either the PNA Mutyper real-time PCR assay or ddPCR. The results of the cobas EGFR mutation test, obtained as part of routine clinical practice, served as a reference for comparison with CTC-based mutation detection. For concordance analyses, paired samples were required to be collected within ±7 days of each other.

5. Treatment process, response assessment, and endpoints

All patients were treated with EGFR-TKI until disease progression, death, or the occurrence of an unacceptable adverse event. Treatment response was evaluated using the Response Evaluation Criteria in Solid Tumors ver. 1.1, as assessed by the treating physician [33]. The primary endpoints were the objective response rate (ORR) and progression-free survival (PFS), stratified by CTC molecular response. ORR was defined as the proportion of patients with partial or complete response as their best response to treatment. PFS was defined as the time from treatment initiation to disease progression or death. The secondary endpoint was the detection rate of EGFR mutations in CTCs.

6. Statistical analysis

Descriptive statistics were used to analyze demographic data. Categorical variables were reported as the number (n) and proportions (%) of patients, while continuous variables were presented as median and range. The baseline characteristics of patients were compared by CTC molecular response using either a t-test or a chi-square test. The comparison of baseline and follow-up CTC count was assessed using the Wilcoxon signed-rank test. Correlation analysis was performed using Spearman’s rank-order correlation test. McNemar’s test was used to compare the sensitivity of EGFR mutation detection between CTC and cobas test. Maximally selected log-rank statistics were employed to determine the optimal cutoff of CTC molecular response. To mitigate potential inflation of type I error associated with evaluating multiple candidate cutpoints, p-values were adjusted using the Lau92 approximation for maximally selected log-rank statistics. To assess the robustness of the CTC cutoff derived using maximally selected log-rank statistics, we conducted internal validation using bootstrap resampling (1,000 iterations) and 5-fold cross-validation. In addition, alternative analyses were performed, including classification based on a fixed 30% reduction threshold consistent with Response Evaluation Criteria in Solid Tumors–based response criteria and Cox proportional hazards modeling treating ΔCTC% as a continuous variable. The comparison of treatment outcomes between CTC molecular responders and non-responders was conducted using the chi-square test, Fisher exact test, and Wilcoxon rank-sum test. PFS was analyzed using the Kaplan-Meier method and presented as the median with a two-sided 95% confidence interval (CI). The log-rank test was used to compare PFS, stratified by CTC molecular response. We employed a Cox proportional hazards regression model to evaluate PFS, including CTC molecular response, treatment regimen, treatment line, smoking status, and other relevant covariates. Model comparisons were performed using the likelihood ratio test, along with assessments of model fit via the Akaike information criterion and Bayesian information criterion. p-values were two-sided, with p < 0.05 considered statistically significant. All statistical analyses were performed using R software ver. 4.2.2 (http://www.R-project.org).

Results

1. Study scheme

A total of 77 patients with NSCLC harboring EGFR mutation were included in the study. Blood samples were collected serially before and after EGFR-TKI treatment to analyze CTC and EGFR ctDNA (Fig. 1A, S1 Table). Most follow-up samples were collected within 6 months from the start of treatment, except for two patients whose follow-up samples were collected at 7 to 8 months from baseline (S2 Table). We defined two analysis subsets according to the timing of follow-up sampling after TKI initiation: one including patients with samples collected within 24 weeks and another including those sampled within 12 weeks. Patients with delayed sampling beyond these cutoffs were excluded from the respective analyses (S2 Table). Clinical response was assessed through imaging studies every 2 to 3 months, as per standard clinical practice (Fig. 1B). EGFR mutation status was evaluated in tumor tissue, plasma, and CTCs from each patient (Fig. 1C). Tissue biopsy was performed at the start of treatment in 69 of the 77 patients, while eight patients underwent only liquid biopsy using the cobas EGFR mutation test (Fig. 1C).

2. Patient characteristics

The baseline characteristics of the study population are summarized in Table 1. The median age was 65 years (range, 41 to 84 years). Approximately two-thirds of the patients were women (68.8%). At enrollment, all patients presented with stage IV disease, of whom 34 (44.2%) were stage IVA and 43 (55.8%) were stage IVB. While non-smoking is more common in EGFR-mutant NSCLC, 28.5% of patients in this cohort were current or former smokers. Among the patients, 49 (63.6%) had the EGFR E19del mutation, 28 (36.3%) had the L858R mutation, and 23 (29.8%) had the T790M mutation acquired after prior therapy. Fifty-three patients (68.8%) were treatment-naïve, while 24 (31.1%) had received at least one prior line of palliative therapy. Among these, 18 (23.4%) had undergone one line, five (6.5%) had received two lines, and one patient (1.3%) had been treated with five regimens prior to enrollment. Regarding treatment regimens, 26 patients (33.7%) received first- or second-generation EGFR-TKIs, while 51 (66.2%) were treated with third-generation EGFR-TKIs.

3. Enumeration of CTCs and detection of ctDNA

CTCs were enumerated for each patient before and after treatment. Cytokeratin-positive, CD45-negative CTCs were isolated and visualized using immunofluorescence staining (Fig. 2A, S3 Fig.). The baseline CTC count was higher than the follow-up count (median, 5/mL vs. 3/mL, p=0.005) (Fig. 2B, S4 and S5 Tables). ctDNA of the original EGFR mutation was detected in the plasma of each patient at baseline (S6 Fig.). The median concentration of EGFR ctDNA did not significantly differ between mutation types (L858R 1.36 copies/μL vs. E19del 1.39 copies/μL, p=0.212) (Fig. 2C, S5 Table). The baseline CTC count and ctDNA concentration showed a positive correlation, reflecting tumor burden in each patient (Spearman’s rank correlation ρ=0.66; p=0.0003 for L858R, and ρ=0.39; p=0.004 for E19del) (Fig. 2D, S5 Table).

4. Association between changes in CTC count and clinical response

Patients were categorized by percent change in CTC counts from baseline (ΔCTC%=100×[baseline−post]/baseline): CTC molecular responders were defined as ΔCTC% ≥ 44.4% (i.e., post-treatment CTC ≤ 55.6% of baseline) (Fig. 3A), and non-responders as ΔCTC% < 44.4% (i.e., post-treatment CTC > 55.6% of baseline). The optimal ΔCTC% threshold determined by the maximally selected log-rank statistic was 44.4%. Internal validation showed consistent results, with bootstrap resampling yielding a median cutoff of 40.0% and 5-fold cross-validation producing cutoff estimates centered around the original threshold (median 44.4%). Five patients with zero CTC counts at both baseline and follow-up in the enumeration aliquot (0→0) could not be classified as responders or non-responders by definition and were therefore excluded from ΔCTC%-based survival analyses. Among the remaining 72 patients, 25 (34.7%) were classified as responders, and 47 (65.3%) were non-responders (Table 1). No significant differences in baseline characteristics were observed between responders and non-responders except for age (Table 1, S7 Table).

Treatment efficacy based on radiologic assessments was compared between CTC molecular responders and non-responders. The median follow-up duration for PFS in all patients was 23.6 months (range, 0.8 to 48.7 months). The median PFS for the entire cohort was 15.9 months (95% CI, 13.3 to not reached). CTC responders showed a significantly longer median PFS of 46.3 months (95% CI, 15.9 to not reached), compared with 13.6 months (95% CI, 8.1 to 22.4) in CTC non-responders (p=0.007) (Fig. 3B). To assess whether the observed prognostic association was dependent on the selected cutoff, alternative analyses were performed. Using a fixed ≥ 30% reduction threshold, CTC molecular responders also showed significantly longer PFS than non-responders (46.3 vs. 12.2 months, p=0.020). Consistently, when ΔCTC% was analyzed as a continuous variable in a Cox proportional hazards model, greater CTC reduction remained significantly associated with improved PFS (p < 0.001), supporting the robustness of the prognostic association beyond a single cutoff definition.

The ORR was 71.4%, consistent with response rates observed in clinical trials of EGFR-TKIs (Table 2). The ORR was not statistically different between CTC responders and non-responders (Table 2, S8 Table). Maximal tumor burden reduction after treatment was slightly greater in CTC responders, with a median change of –37.7% compared to –35.2% in non-responders, although the difference was not statistically significant (Fig. 3C and D, S8 Table). At the time of data cutoff, 17 of 25 CTC responders (68.0%) were receiving ongoing treatment, compared to 15 of 47 CTC non-responders (32.0%) (Fig. 3E).

Univariate analyses were conducted to assess the associations of individual variables with PFS (S9 Table). For multivariable analysis, two approaches were applied: one including all available covariates (S10 Fig.), and another restricted to clinically meaningful variables to minimize overfitting. Consistently, both approaches demonstrated that CTC molecular response was an independent predictor of improved PFS (hazard ratio [HR], 0.3; p=0.005) (Fig. 3F, S10 Fig.). A non-significant trend toward shorter PFS was observed in patients receiving later-line treatments (HR, 1.6; p=0.274), first- or second-generation EGFR-TKIs (HR 1.8, p=0.11), and those with a history of smoking (HR, 1.5; p=0.267). Sensitivity analyses using the predefined sampling windows (≤ 24 and ≤ 12 weeks) yielded findings that were broadly consistent with the main analyses. Median PFS, the extent of maximal tumor reduction, and the direction of hazard ratios in the multivariable models appeared largely unchanged after excluding patients with delayed sampling (S11 Fig.). Across these analyses, CTC responders showed a tendency toward more favorable outcomes, although this pattern should be interpreted cautiously given the smaller number of patients included in the restricted subsets. Taken together, these results provide supportive evidence that our main findings are reasonably robust to variation in follow-up sampling timing.

Additional sensitivity analyses further supported the robustness of the findings. Multivariable Cox models were re-estimated including patients with zero CTC counts at both baseline and follow-up (0→0), who were non-evaluable for ΔCTC%-based response classification. In extended multivariable models, including both a model adjusted for four clinically meaningful covariates and a more comprehensive model incorporating all available covariates with complete data, the hazard ratios for ΔCTC% response remained highly consistent with those of the primary analysis, with no change in direction or statistical significance (S12 Fig.).

In subgroup analyses stratified by treatment line, ΔCTC% responders exhibited longer PFS than non-responders in both treatment-naïve and previously treated patients, although statistical significance was observed only in the previously treated subgroup (S13 Fig.). An interaction analysis revealed no statistically significant interaction between ΔCTC% response and treatment line (interaction p=0.15).

5. Comparison of EGFR mutation status across tumor tissue, plasma, and CTCs

The EGFR mutation status determined at the time of CTC collection was compared with results from tissue biopsy or liquid biopsy obtained from clinical practice at a similar time point (Fig. 4A, S14 Table). Liquid biopsy results were included only when collected within a ±7-day window of the corresponding CTC sampling. Among the included pairs, the median interval between the two sample types was 4 days (IQR, 0 to 6 days). EGFR mutation detection by ddPCR was performed using CTC-derived gDNA from the baseline molecular aliquot. Cases with detectable EGFR mutations were included in comparative analyses across tissue, CTC, and cobas platforms, irrespective of CTC enumeration results. Notably, among five patients with zero CTC counts at both baseline and follow-up, four showed detectable EGFR mutations in the molecular aliquot despite zero counts in the enumeration aliquot and were therefore included in the comparative analyses (S15 Table). In contrast, three patients were labeled as “test not done” and excluded from the comparative analyses because of insufficient material for mutation testing in the molecular aliquot; however, these patients had a baseline CTC count of zero with increased CTC counts at follow-up and were thus classified as CTC non-responders in the survival analysis. While gDNA copy numbers did not differ significantly between patients with zero versus detectable CTC counts, patients with higher CTC burden than cohort median (CTC ≥ 5/mL) exhibited significantly greater CTC-derived gDNA copy numbers compared with those with lower counts (Wilcoxon p < 0.001) (S16 Table).

Of the 77 patients, 41 underwent both tissue and liquid biopsies simultaneously. Excluding three patients with insufficient CTC samples for EGFR mutation analysis, the primary E19del or L858R mutation was concordant across tumor tissue, plasma, and CTCs in 21 of 38 patients (55.3%) (S17A Fig.). EGFR mutations were identified in 33 of 38 patients (86.8%) through CTC analysis, confirming the authenticity of the isolated cells as CTCs (S17A Fig.). In this study, the EGFR mutation detection sensitivity of the CCM-CTCD method was 87.8% (S17B Fig.). Among 46 patients who underwent both cobas liquid biopsy and CTC analysis simultaneously, the sensitivity was 69.6% for cobas and 87.0% for CCM-CTCD (S17C Fig.). The concordance rate between CCM-CTCD and cobas liquid biopsy was 65.2%. In this selected cohort, CCM-CTCD demonstrated a higher observed sensitivity compared with the cobas test (87.0% vs. 69.6%, p=0.029) (S18 Table). Of these 46 patients, 19 (41.3%) had both samples collected on the same day. Within this subgroup, the sensitivity was 57.9% for cobas and 84.2% for CCM-CTCD, reflecting a consistent pattern favoring CCM-CTCD (S18 Table). Among five patients with de novo T790M mutations confirmed by tissue biopsy prior to any EGFR-TKI treatment, four were identified through CTC analysis (data not shown). EGFR T790M mutation status was compared across CTCs, cobas liquid biopsy, and tissue samples in patients previously treated with first- or second-generation EGFR-TKIs and considered susceptible to resistance-associated T790M mutations (Fig. 4B). Discrepancies in T790M mutation detection among tissue, plasma, and CTCs were observed, highlighting the heterogeneity of cancer cells following TKI treatment. In this cohort, CCM-CTCD demonstrated higher detection rate of T790M mutation compared to the cobas test.

Discussion

This study highlights the clinical utility of CTCs in predicting treatment efficacy and stratifying survival outcomes in patients with EGFR-mutant NSCLC treated with EGFR-TKI therapy. The high performance of the CCM-CTCD method in isolating genuine CTCs was validated by demonstrating the concordance of EGFR mutation status across tumor tissue, plasma, and CTCs. These findings provide compelling evidence supporting the role of CTC analysis in risk stratification and guiding treatment decisions in patients with EGFR-mutant lung cancer.

One key finding is the significant association between CTC molecular response and PFS in patients with EGFR-mutant lung cancer undergoing EGFR-TKI therapy. Patients classified as CTC molecular responders exhibited significantly longer PFS compared to non-responders, reinforcing the predictive role of CTC dynamics in treatment outcomes. The greater tumor burden reduction observed in responders further supports using CTC molecular response as a reliable surrogate marker for evaluating treatment efficacy. These findings are consistent with prior studies that have reported the prognostic and predictive significance of CTC enumeration and molecular profiling in various cancer types, including NSCLC [34,35].

In this study, we enrolled patients with EGFR-mutant lung cancer to validate the isolation of genuine CTCs. EGFR mutations identified in CTCs were compared with those detected in ctDNA and tissue biopsies. CCM-CTCD showed concordance with cobas assay and higher observed sensitivity for EGFR mutation detection in this selected cohort. However, this reflects patient-level detection performance rather than analytical sensitivity and may have been influenced by prior treatment, pre-analytical factors, and platform-specific differences. Previously reported clinical validation studies have shown cobas sensitivities of approximately 60%-70%, consistent with our findings [36,37]. While our results align with literature suggesting a potential sensitivity advantage of CTC-based approaches in certain settings, they do not establish generalizable superiority, and external validation in larger cohorts is warranted. Mutational discordance among tissue, plasma ctDNA, and CTCs highlighted tumor heterogeneity and underscored the complementary roles of each method in identifying genomic alterations. The observation of detectable EGFR mutations in cases with zero enumerated CTCs is attributable to the aliquot-based assay design. Because molecular analysis and CTC enumeration were performed on separate aliquots, rare CTCs or CTC-derived gDNA may be present in the molecular aliquot even when no morphologically intact CTCs are identified by ICC. Consistent with this explanation, gDNA copy numbers did not differ significantly between patients with zero versus detectable CTC counts, indicating that molecular signals can be present even in the absence of enumerated CTCs. In contrast, a clear burden-dependent pattern emerged when CTC counts were stratified by the cohort median, with significantly higher gDNA copy numbers observed in patients with higher CTC burden, and CTC counts showed a moderate positive correlation with CTC-derived gDNA copies. Taken together, these findings support an overall quantitative concordance between CTC enumeration and molecular analysis across the spectrum of CTC burden. CTC counts were compared with CT imaging to monitor patient status, revealing discrepancies in some cases. These findings suggest that CTC analysis could serve as a parameter for predicting treatment response and detecting early progression. While baseline CTC counts reflect tumor burden and have demonstrated prognostic value, their distinct role compared to imaging and tumor staging remains unclear.

Our study also suggests the potential of CTC molecular profiling to guide treatment intensification in poor responders. Stratifying patients based on CTC molecular response could enable more tailored therapeutic decisions, advancing precision oncology. Recent clinical trials have demonstrated an association between ctDNA clearance and improved survival outcomes [38,39]. Similarly, CTC kinetics have been evaluated in breast and prostate cancers, where reductions in CTC counts by 25% to 50% from baseline have correlated with treatment response [40-43]. Consistent with these findings, our study showed that CTC molecular responders exhibited prolonged PFS and greater tumor burden reduction compared to CTC non-responders. These results underscore the predictive value of CTC dynamics and the importance of close surveillance to detect early disease progression in non-responders. However, the impact of early treatment intensification for CTC non-responders on survival outcomes remains uncertain. For example, in metastatic breast cancer, early modification of chemotherapy regimens for patients with persistent CTC detection after one treatment cycle did not improve survival outcomes [44]. This underscores the need for further research to determine the clinical utility of treatment intensification strategies in this subgroup. In EGFR-mutated NSCLC, combining EGFR-TKI with other targeted agents or chemotherapy, either upfront or subsequent therapy, has emerged as an effective strategy to address heterogeneous acquired resistance and prolong survival [7,8,45,46]. Ongoing clinical trials are evaluating add-on strategies in patients with persistent ctDNA following treatment with third-generation EGFR-TKI (NCT04410796 and NCT06020989) [47]. Similarly, incorporating chemotherapy, antibody-drug conjugates, or other targeted therapies for CTC molecular non-responders may provide a rationale for treatment intensification in patients harboring EGFR mutation. Currently, all patients harboring EGFR mutation receive uniform treatment approaches. However, developing robust risk stratification methods could enable the implementation of risk-adapted treatment strategies, optimizing outcomes for this patient population.

Molecular diagnostics are essential for enabling targeted therapies and predicting treatment response. Liquid biopsy, particularly ctDNA detection, provides a non-invasive alternative when tissue biopsy is not feasible. CTC detection provides comprehensive insights into various cancer cell components. However, current technologies, including Cell-Search, the only FDA-approved CTC detection method, face significant limitations in sensitivity, specificity, and standardization, especially in capturing heterogeneous CTCs. Due to the reduced expression of epithelial cell adhesion molecule (EpCAM) on CTCs in metastatic NSCLC, alternative platforms are essential. Recent advancements in CTC detection technologies have introduced various innovative approaches. However, they have their own limitations in detecting heterogeneous CTCs. Telomerase-based assays [48-50] lack an enrichment step and require extensive imaging and scanning, which can be time-consuming and computationally intensive. Nanostructured platforms have improved CTC recovery [51-53], but have inherent limitations in that they rely on positive selection methods or size-dependent methods [54,55]. As previously evaluated, CCM-CTCD demonstrated the ability to detect diverse CTC subtypes through its marker-agnostic approach, including those undergoing epithelial-mesenchymal transition and stem-like CTCs, which are associated with aggressive disease and poor prognosis [31]. Our findings provide clinical evidence supporting CCM-CTCD as an effective tool for predicting EGFR-TKI treatment response in patients with NSCLC, underscoring its value in advancing liquid biopsy technologies [56].

Despite its strengths, this study has some limitations that should be acknowledged. First, the relatively small sample size necessitates validation in larger cohorts. Because the ΔCTC% cutoff was derived using maximally selected log-rank statistics, concerns regarding potential inflation of type I error may arise. To mitigate this risk, we applied an established adjustment for multiple testing within the maximally selected log-rank framework. In addition, as the cutoff was identified within a single cohort, the possibility of overfitting and limited generalizability cannot be excluded. However, internal validation using bootstrap resampling and cross-validation demonstrated that the cutoff consistently centered around the originally selected threshold, supporting its internal robustness. Importantly, the prognostic relevance of CTC dynamics was also confirmed using alternative analytic approaches, including a fixed 30% reduction threshold and continuous-variable modeling, suggesting that the observed associations were not dependent on a single cutoff definition. Nevertheless, external validation in independent cohorts will be required to confirm the generalizability and clinical applicability of these findings. Second, the heterogeneous clinical setting of our patients may have influenced findings. In particular, the timing of post-treatment sampling was not standardized across patients, and repeated follow-up sampling could not be performed. Nevertheless, the intervals from tissue biopsy to baseline sampling (median, 15 days; IQR, 11 to 16 days) and from treatment initiation to follow-up sampling (median, 4 weeks; IQR, 3 to 11 weeks) were reasonably long, which may help limit the impact of transient fluctuations related to recent procedures or early treatment effects. In addition, compound EGFR T790M mutations were detected at a higher frequency than previously reported, likely reflecting the inclusion of patients with prior EGFR-TKI exposure. Although only a small subset of our cohort had undergone extensive pretreatment, heterogeneity in treatment history may still have affected CTC dynamics and survival outcomes. To address this, we incorporated prior treatment status into our multivariable analyses. In addition, we performed subgroup analyses stratified by treatment line and formally tested for interaction between ΔCTC% response and treatment line. While ΔCTC% responders showed longer PFS than non-responders in both treatment-naïve and previously treated patients, no statistically significant interaction was observed, indicating that the prognostic impact of ΔCTC% response was not materially modified by treatment line. These findings suggest that the observed association between ΔCTC% response and PFS is broadly consistent across treatment-line strata and that differences in subgroup-level statistical significance likely reflect sampling variability rather than true biological heterogeneity. In interpreting these results, ΔCTC% response could not be defined in patients with 0→0 CTC counts; however, sensitivity analyses confirmed that exclusion of these cases did not affect the observed associations. Finally, the observational design limits the ability to draw definitive conclusions about causality between CTC dynamics and clinical outcomes.

In conclusion, our findings suggest that CTC analysis may serve as a minimally invasive biomarker for monitoring treatment response and stratifying prognosis in patients with EGFR-mutant NSCLC. The observed associations indicate that CTC dynamics could potentially inform treatment decisions, including consideration of treatment intensification in selected high-risk patients; however, clinical benefit from such an approach remains unproven. Advanced technologies like CCM-CTCD provide deeper insights into tumor biology, but prospective interventional trials will be required to determine whether CTC-guided treatment modification improves clinical outcomes. Further well-designed, larger prospective studies are needed to validate risk stratification based on CTC dynamics and to define their optimal clinical application.

Electronic Supplementary Material

Supplementary materials are available at Cancer Research and Treatment website (https://www.e-crt.org).

crt-2025-672_S1_Table.pdf

crt-2025-672_S2_Table.pdf

crt-2025-672_S3_Fig.pdf

crt-2025-672_S4_Table.pdf

crt-2025-672_S5_Table.pdf

crt-2025-672_S6_Fig.pdf

crt-2025-672_S7_Table.pdf

crt-2025-672_S8_Table.pdf

crt-2025-672_S9_Table.pdf

crt-2025-672_S10_Fig.pdf

crt-2025-672_S11_Fig.pdf

crt-2025-672_S12_Fig.pdf

crt-2025-672_S13_Fig.pdf

crt-2025-672_S14_Table.pdf

crt-2025-672_S15_Table.pdf

crt-2025-672_S16_Table.pdf

crt-2025-672_S17_Fig.pdf

crt-2025-672_S18_Table.pdf

NOTES

Ethical Statements

This study was approved by the Institutional Review Board (IRB) of Severance Hospital (IRB No. 4-2013-0059). It was conducted in accordance with the Declaration of Helsinki. All patients provided written informed consent for study participation.

Author Contributions

Conceived and designed the analysis: Lee S, Kim MS, Kim HR.

Collected the data: Lee S, Kim C.

Contributed data or analysis tools: Lee S, Kim C, Kim CG, Hong MH, Han M, Son W, Kim G, Woo HJ, Shin HY, Lee J.

Performed the analysis: Lee S, Kim C, Woo HJ.

Wrote the paper: Lee S, Kim C, Woo HJ, Kim MS, Kim HR.

Analysis of preprocessing outcomes from CTCeptor: Shin HY, Lee J.

Methodology for CTC analysis: Kim MS.

Hyeong Jung Woo, Hyun Young Shin, Jungmin Lee, and Minseok S Kim did not have access to clinical outcome data of patients and were not involved in the correlative analyses, which were conducted by authors affiliated with Yonsei Cancer Center (Academia). Accordingly, their affiliations did not influence the study design, data interpretation, results, or conclusions.

Conflicts of Interest

Conflict of interest relevant to this article was not reported.

Funding

This work was supported by the ‘Supporting Project for the evaluation of Domestic Medical Devices in Hospitals’ (funded by MOHW and KHIDI), the Brain Korea 21 FOUR program, the Technology Innovation Program (20022947, Ministry of Trade Industry and Energy, Korea), the National Research Foundation (2021R1A2C2094629, Ministry of Science and ICT, and the Yonsei Fellow Program (Lee Youn Jae).

Fig. 1.

Study scheme. (A) Experimental design of liquid biopsy and tissue sampling. (B) Swimmer plot showing the treatment response (by Response Evaluation Criteria in Solid Tumors ver. 1.1) of participants, annotated with circulating tumor cell (CTC) sampling. (C) CONSORT diagram of the study. CCM-CTCD, Continuous Centrifugal Microfluidics–Circulating Tumor Cell Disc; ctDNA, circulating tumor DNA; ddPCR, droplet digital polymerase chain reaction; EGFR, epidermal growth factor receptor; NSCLC, non–small cell lung cancer; OS, overall survival; PFS, progression-free survival; PR, partial response; SD, stable disease.

Fig. 2.

Immunostaining images of circulating tumor cells (CTCs) (A) and CTC counts per mL before and after targeted therapy (B) (baseline: median, 5 [range, 0 to 20]; follow-up: median, 3 [range, 0 to 14]). (C) Average circulating tumor DNA (ctDNA) copy number across all patients. (D) Correlation between CTC counts and the copy number of each ctDNA mutation.

Fig. 3.

Circulating tumor cell (CTC) change and clinical response. (A) Classification of patients as CTC molecular responders and non-responders. (B) Kaplan-Meier plot of progression-free survival (PFS) for all patients, CTC molecular responders, and non-responders. (C) Waterfall plot illustrating best tumor shrinkage in individual patients, (D) Maximal tumor burden reduction. (E) Swimmer plot stratified by CTC molecular response. (F) Forest plot of PFS. AIC, Akaike information criterion; TKI, tyrosine kinase inhibitor. ^**p < 0.01.

Fig. 4.

Epidermal growth factor receptor (EGFR) mutation profile of individual patients. (A) Mutation profile of all patients detected from tissue, circulating tumor DNA (ctDNA), and circulating tumor cell (CTC). CTC and ctDNA samples were collected at closely matched time points (≤ 7 days apart). (B) Original and T790M mutation profile in patients previously treated with EGFR–tyrosine kinase inhibitor. CCM-CTCD, Continuous Centrifugal Microfluidics–Circulating Tumor Cell Disc.

Table 1.

Baseline characteristics

Characteristic	Total (n=77)	CTC molecular responder (n=25)	CTC molecular non-responder (n=47)
Age (yr)	65 (41-84)	68 (50-84)^a)	63 (41-82)^a)
Female sex	53 (68.8)	18 (72.0)	31 (66.0)
Smoking
Never smoker	55 (71.4)	18 (72.0)	33 (70.2)
Former/Current smoker	22 (28.6)	7 (28.0)	14 (29.8)
EGFR mutation type
Exon 19 deletion	49 (63.6)	16 (64.0)	30 (63.8)
L858R	28 (36.4)	9 (36.0)	17 (36.2)
T790M	23 (29.8)	9 (36.0)	12 (25.5)
Clinical stage
Stage IVA	34 (44.2)	12 (48.0)	19 (40.4)
Stage IVB	43 (55.8)	13 (52.0)	28 (59.6)
Line of treatment
1st line	53 (68.8)	15 (60.0)	34 (72.3)
2nd or more advanced	24 (31.2)	10 (40.0)	13 (27.7)
Treatment regimen
Gefitinib/Erlotinib/Afatinib	26 (33.7)	9 (36.0)	17 (36.2)
Osimertinib/Lazertinib	51 (66.2)	16 (64.0)	30 (63.8)
Palliative radiotherapy	17 (22.1)	4 (16.0)	12 (25.5)

Values are presented as median (range) or number (%). CTC, circulating tumor cell; EGFR, epidermal growth factor receptor.

^a) The median age of molecular responders was older than that of non-responders (p=0.007).

Table 2.

Summary of treatment outcome

	Total (n=77)^a)	CTC molecular responder (n=25)	CTC molecular non-responder (n=47)^a)
Best response, n (%)
Complete response	0	0	0
Partial response	55 (71.4)	18 (72.0)	33 (71.7)
Stable disease	18 (23.3)	7 (28.0)	10 (21.7)
Progressive disease	3 (3.8)	0	3 (6.5)
Objective response rate (%)	71.4	72.0	71.7
Disease control rate (%)	96.2	100	93.5

Values are presented as number (%). CTC, circulating tumor cell.

^a) Before the first response assessment, one CTC non-responder patient expired due to septic shock, and was excluded from this analysis of best response.

REFERENCES

1. Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2021;71:209–49. Article PubMed PDF
2. Melosky B, Kambartel K, Hantschel M, Bennetts M, Nickens DJ, Brinkmann J, et al. Worldwide prevalence of epidermal growth factor receptor mutations in non-small cell lung cancer: a meta-analysis. Mol Diagn Ther. 2022;26:7–18. Article PubMed PDF
3. Hendriks LE, Kerr KM, Menis J, Mok TS, Nestle U, Passaro A, et al. Oncogene-addicted metastatic non-small-cell lung cancer: ESMO clinical practice guideline for diagnosis, treatment and follow-up. Ann Oncol. 2023;34:339–57. Article PubMed
4. Jaiyesimi IA, Leighl NB, Ismaila N, Alluri K, Florez N, Gadgeel S, et al. Therapy for stage IV non-small cell lung cancer with driver alterations: ASCO living guideline, version 2023.3. J Clin Oncol. 2024;42:e1–22. PubMed
5. Soria JC, Ohe Y, Vansteenkiste J, Reungwetwattana T, Chewaskulyong B, Lee KH, et al. Osimertinib in untreated EGFR-mutated advanced non-small-cell lung cancer. N Engl J Med. 2018;378:113–25. PubMed
6. Cho BC, Ahn MJ, Kang JH, Soo RA, Reungwetwattana T, Yang JC, et al. Lazertinib versus gefitinib as first-line treatment in patients with EGFR-mutated advanced non-small-cell lung cancer: results from LASER301. J Clin Oncol. 2023;41:4208–17. Article PubMed
7. Planchard D, Janne PA, Cheng Y, Yang JC, Yanagitani N, Kim SW, et al. Osimertinib with or without chemotherapy in EGFR-mutated advanced NSCLC. N Engl J Med. 2023;389:1935–48. Article PubMed
8. Cho BC, Lu S, Felip E, Spira AI, Girard N, Lee JS, et al. Amivantamab plus lazertinib in previously untreated EGFR-mutated advanced NSCLC. N Engl J Med. 2024;391:1486–98. PubMed
9. Scheffler M, Wiesweg M, Michels S, Nogova L, Kron A, Herold T, et al. Rebiopsy in advanced non-small cell lung cancer, clinical relevance and prognostic implications. Lung Cancer. 2022;168:10–20. Article PubMed
10. Lone SN, Nisar S, Masoodi T, Singh M, Rizwan A, Hashem S, et al. Liquid biopsy: a step closer to transform diagnosis, prognosis and future of cancer treatments. Mol Cancer. 2022;21:79.Article PubMed PMC PDF
11. Wu YL, Zhou C, Liam CK, Wu G, Liu X, Zhong Z, et al. First-line erlotinib versus gemcitabine/cisplatin in patients with advanced EGFR mutation-positive non-small-cell lung cancer: analyses from the phase III, randomized, open-label, ENSURE study. Ann Oncol. 2015;26:1883–9. Article PubMed
12. U.S. Food and Drug Administration. cobas EGFR Mutation Test v2 [Internet]. U.S. Food and Drug Administration; 2016. [cited 2025 Oct 1]. Available from: https://www.fda.gov/drugs/resources-information-approved-drugs/cobas-egfr-mutation-test-v2
13. Leighl NB, Page RD, Raymond VM, Daniel DB, Divers SG, Reckamp KL, et al. Clinical utility of comprehensive cell-free DNA analysis to identify genomic biomarkers in patients with newly diagnosed metastatic non-small cell lung cancer. Clin Cancer Res. 2019;25:4691–700. Article PubMed PDF
14. Woodhouse R, Li M, Hughes J, Delfosse D, Skoletsky J, Ma P, et al. Clinical and analytical validation of FoundationOne Liquid CDx, a novel 324-Gene cfDNA-based comprehensive genomic profiling assay for cancers of solid tumor origin. PLoS One. 2020;15:e0237802Article PubMed PMC
15. Odegaard JI, Vincent JJ, Mortimer S, Vowles JV, Ulrich BC, Banks KC, et al. Validation of a plasma-based comprehensive cancer genotyping assay utilizing orthogonal tissue- and plasma-based methodologies. Clin Cancer Res. 2018;24:3539–49. Article PubMed PDF
16. Vidlarova M, Rehulkova A, Stejskal P, Prokopova A, Slavik H, Hajduch M, et al. Recent advances in methods for circulating tumor cell detection. Int J Mol Sci. 2023;24:3902.Article PubMed PMC
17. Cabel L, Proudhon C, Gortais H, Loirat D, Coussy F, Pierga JY, et al. Circulating tumor cells: clinical validity and utility. Int J Clin Oncol. 2017;22:421–30. Article PubMed PDF
18. Ju S, Chen C, Zhang J, Xu L, Zhang X, Li Z, et al. Detection of circulating tumor cells: opportunities and challenges. Biomark Res. 2022;10:58.Article PubMed PMC PDF
19. Wang X, Wang L, Lin H, Zhu Y, Huang D, Lai M, et al. Research progress of CTC, ctDNA, and EVs in cancer liquid biopsy. Front Oncol. 2024;14:1303335.Article PubMed PMC
20. Wang C, Mu Z, Ye Z, Zhang Z, Abu-Khalaf MM, Silver DP, et al. Prognostic value of HER2 status on circulating tumor cells in advanced-stage breast cancer patients with HER2-negative tumors. Breast Cancer Res Treat. 2020;181:679–89. Article PubMed PMC PDF
21. Boral D, Vishnoi M, Liu HN, Yin W, Sprouse ML, Scamardo A, et al. Molecular characterization of breast cancer CTCs associated with brain metastasis. Nat Commun. 2017;8:196.Article PubMed PMC PDF
22. Antonarakis ES, Kuang Z, Tukachinsky H, Parachoniak C, Kelly AD, Gjoerup O, et al. Prevalence of inferred clonal hematopoiesis (CH) detected on comprehensive genomic profiling (CGP) of solid tumor tissue or circulating tumor DNA (ctDNA). J Clin Oncol. 2021;39(15 Suppl):3009.Article
23. Paterlini-Brechot P, Benali NL. Circulating tumor cells (CTC) detection: clinical impact and future directions. Cancer Lett. 2007;253:180–204. Article PubMed
24. Lin D, Shen L, Luo M, Zhang K, Li J, Yang Q, et al. Circulating tumor cells: biology and clinical significance. Signal Transduct Target Ther. 2021;6:404.Article PubMed PMC PDF
25. Deng Z, Wu S, Wang Y, Shi D. Circulating tumor cell isolation for cancer diagnosis and prognosis. EBioMedicine. 2022;83:104237.Article PubMed PMC
26. Ferreira MM, Ramani VC, Jeffrey SS. Circulating tumor cell technologies. Mol Oncol. 2016;10:374–94. Article PubMed PMC PDF
27. Keller L, Werner S, Pantel K. Biology and clinical relevance of EpCAM. Cell Stress. 2019;3:165–80. Article PubMed PMC
28. Grover PK, Cummins AG, Price TJ, Roberts-Thomson IC, Hardingham JE. Circulating tumour cells: the evolving concept and the inadequacy of their enrichment by EpCAM-based methodology for basic and clinical cancer research. Ann Oncol. 2014;25:1506–16. Article PubMed
29. Banko P, Lee SY, Nagygyorgy V, Zrinyi M, Chae CH, Cho DH, et al. Technologies for circulating tumor cell separation from whole blood. J Hematol Oncol. 2019;12:48.PubMed PMC
30. Rushton AJ, Nteliopoulos G, Shaw JA, Coombes RC. A review of circulating tumour cell enrichment technologies. Cancers (Basel). 2021;13:970.Article PubMed PMC
31. Woo HJ, Kim SH, Kang HJ, Lee SH, Lee SJ, Kim JM, et al. Continuous centrifugal microfluidics (CCM) isolates heterogeneous circulating tumor cells via full automation. Theranostics. 2022;12:3676–89. Article PubMed PMC
32. Woo HJ, Kim SH, Kang HG, Kim T, Kim S, Kim JM, et al. Lossless immunocytochemistry based on large-scale porous hydrogel pellicle for accurate rare cell analysis. ACS Appl Mater Interfaces. 2023;15:15059–70. Article PubMed PDF
33. Eisenhauer EA, Therasse P, Bogaerts J, Schwartz LH, Sargent D, Ford R, et al. New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer. 2009;45:228–47. Article PubMed
34. Krebs MG, Sloane R, Priest L, Lancashire L, Hou JM, Greystoke A, et al. Evaluation and prognostic significance of circulating tumor cells in patients with non-small-cell lung cancer. J Clin Oncol. 2011;29:1556–63. Article PubMed
35. Lindsay CR, Faugeroux V, Michiels S, Pailler E, Facchinetti F, Ou D, et al. A prospective examination of circulating tumor cell profiles in non-small-cell lung cancer molecular subgroups. Ann Oncol. 2017;28:1523–31. Article PubMed
36. Itani H, Iwamoto K, Ito Y, Sasano H, Kondo S, Tanigawa M. P3.01-064 Detection of EGFR mutations in circulating tumor DNA using plasma samples: clinical validation of Cobas EGFR mutation test V2. J Thorac Oncol. 2017;12(11 Suppl 2):S2226.Article
37. Passiglia F, Rizzo S, Di Maio M, Galvano A, Badalamenti G, Listi A, et al. The diagnostic accuracy of circulating tumor DNA for the detection of EGFR-T790M mutation in NSCLC: a systematic review and meta-analysis. Sci Rep. 2018;8:13379.Article PubMed PMC PDF
38. Soo RA, Martini JF, van der Wekken AJ, Teraoka S, Ferrara R, Shaw AT, et al. Early circulating tumor DNA dynamics and efficacy of lorlatinib in patients with treatment-naive, advanced, ALK-positive NSCLC. J Thorac Oncol. 2023;18:1568–80. Article PubMed
39. Gray JE, Ahn MJ, Oxnard GR, Shepherd FA, Imamura F, Cheng Y, et al. Early clearance of plasma epidermal growth factor receptor mutations as a predictor of outcome on osimertinib in advanced non-small cell lung cancer: exploratory analysis from AURA3 and FLAURA. Clin Cancer Res. 2023;29:3340–51. Article PubMed PDF
40. Deutsch TM, Stefanovic S, Feisst M, Fischer C, Riedel F, Fremd C, et al. Cut-off analysis of CTC change under systemic therapy for defining early therapy response in metastatic breast cancer. Cancers (Basel). 2020;12:1055.Article PubMed PMC
41. Lorente D, Olmos D, Mateo J, Bianchini D, Seed G, Fleisher M, et al. Decline in circulating tumor cell count and treatment outcome in advanced prostate cancer. Eur Urol. 2016;70:985–92. Article PubMed PMC
42. Deutsch TM, Riethdorf S, Nees J, Hartkopf AD, Schonfisch B, Domschke C, et al. Impact of apoptotic circulating tumor cells (aCTC) in metastatic breast cancer. Breast Cancer Res Treat. 2016;160:277–90. Article PubMed PDF
43. Scher HI, Jia X, de Bono JS, Fleisher M, Pienta KJ, Raghavan D, et al. Circulating tumour cells as prognostic markers in progressive, castration-resistant prostate cancer: a reanalysis of IMMC38 trial data. Lancet Oncol. 2009;10:233–9. Article PubMed PMC
44. Smerage JB, Barlow WE, Hortobagyi GN, Winer EP, Leyland-Jones B, Srkalovic G, et al. Circulating tumor cells and response to chemotherapy in metastatic breast cancer: SWOG S0500. J Clin Oncol. 2014;32:3483–9. Article PubMed PMC
45. Passaro A, Wang J, Wang Y, Lee SH, Melosky B, Shih JY, et al. Amivantamab plus chemotherapy with and without lazertinib in EGFR-mutant advanced NSCLC after disease progression on osimertinib: primary results from the phase III MARIPOSA-2 study. Ann Oncol. 2024;35:77–90. PubMed
46. Sequist LV, Peled N, Tufman A, Servidio L, Li J, Taylor R, et al. P47.11 COMPEL: chemotherapy with/without osimertinib in patients with EGFRm advanced NSCLC and progression on first-line osimertinib. J Thorac Oncol. 2021;16(10 Suppl):S1101.Article
47. Desai A, Vazquez TA, Arce KM, Corassa M, Mack PC, Gray JE, et al. ctDNA for the evaluation and management of EGFR-mutant non-small cell lung cancer. Cancers (Basel). 2024;16:940.Article PubMed PMC
48. Yang Z, Bai H, Hu L, Kong D, Li G, Zhao C, et al. Improving the diagnosis of prostate cancer by telomerase-positive circulating tumor cells: a prospective pilot study. EClinicalMedicine. 2022;43:101161.Article PubMed PMC
49. Yang Z, Bai H, Wu Q, Hu L, Li G, Kong D, et al. Telomerase positive CTCs with PSMA high expression associated with prostate cancer metastasis. Transl Androl Urol. 2022;11:803–13. Article PubMed PMC
50. Zhang W, Qin T, Yang Z, Yin L, Zhao C, Feng L, et al. Telomerase-positive circulating tumor cells are associated with poor prognosis via a neutrophil-mediated inflammatory immune environment in glioma. BMC Med. 2021;19:277.Article PubMed PMC PDF
51. Guo L, Liu C, Qi M, Cheng L, Wang L, Li C, et al. Recent progress of nanostructure-based enrichment of circulating tumor cells and downstream analysis. Lab Chip. 2023;23:1493–523. Article PubMed
52. Zhan L, Edd J, Mishra A, Toner M. Label-free microfluidic apheresis of circulating tumor cell clusters. Adv Sci (Weinh). 2024;11:e2405853Article PubMed PMC
53. Lu L, Wang Y, Ding Y, Wang Y, Zhu Z, Lu J, et al. Profiling phenotypic heterogeneity of circulating tumor cells through spatially resolved immunocapture on nanoporous micropillar arrays. ACS Nano. 2024;18:31135–47. PubMed
54. Mansor MA, Yang C, Chong KL, Jamrus MA, Liu K, Yu M, et al. Label-free and rapid microfluidic design rules for circulating tumor cell enrichment and isolation: a review and simulation analysis. ACS Omega. 2025;10:6306–22. Article PubMed PMC PDF
55. Wei YJ, Wei X, Zhang X, Wu CX, Cai JY, Chen ML, et al. A hydrodynamic-based dual-function microfluidic chip for high throughput discriminating tumor cells. Talanta. 2024;273:125884.Article PubMed
56. Wang J, Lu W, Tang C, Liu Y, Sun J, Mu X, et al. Label-free isolation and mRNA detection of circulating tumor cells from patients with metastatic lung cancer for disease diagnosis and monitoring therapeutic efficacy. Anal Chem. 2015;87:11893–900. Article PubMed

Figure & Data

REFERENCES

Citations

Citations to this article as recorded by

PubReader
ePub Link

Cite

CITE: export Copy Download Format; Close

Download Citation

Download a citation file in RIS format that can be imported by all major citation management software, including EndNote, ProCite, RefWorks, and Reference Manager.

Format:

RIS — For EndNote, ProCite, RefWorks, and most other reference management software
BibTeX — For JabRef, BibDesk, and other BibTeX-specific software

Include:

Citation for the content below
Citation and abstract for the content below

Circulating Tumor Cell–Based Molecular Responses Stratify EGFR-TKI Efficacy in Patients with EGFR-Mutant Lung Cancer

DOI: https://doi.org/10.4143/crt.2025.672

XML Download

Abstract
Introduction
Materials and Methods
Results
Discussion
Electronic Supplementary Material
NOTES
REFERENCES

Circulating Tumor Cell–Based Molecular Responses Stratify EGFR-TKI Efficacy in Patients with EGFR-Mutant Lung Cancer

Fig. 1. Study scheme. (A) Experimental design of liquid biopsy and tissue sampling. (B) Swimmer plot showing the treatment response (by Response Evaluation Criteria in Solid Tumors ver. 1.1) of participants, annotated with circulating tumor cell (CTC) sampling. (C) CONSORT diagram of the study. CCM-CTCD, Continuous Centrifugal Microfluidics–Circulating Tumor Cell Disc; ctDNA, circulating tumor DNA; ddPCR, droplet digital polymerase chain reaction; EGFR, epidermal growth factor receptor; NSCLC, non–small cell lung cancer; OS, overall survival; PFS, progression-free survival; PR, partial response; SD, stable disease.

Fig. 2. Immunostaining images of circulating tumor cells (CTCs) (A) and CTC counts per mL before and after targeted therapy (B) (baseline: median, 5 [range, 0 to 20]; follow-up: median, 3 [range, 0 to 14]). (C) Average circulating tumor DNA (ctDNA) copy number across all patients. (D) Correlation between CTC counts and the copy number of each ctDNA mutation.

Fig. 3. Circulating tumor cell (CTC) change and clinical response. (A) Classification of patients as CTC molecular responders and non-responders. (B) Kaplan-Meier plot of progression-free survival (PFS) for all patients, CTC molecular responders, and non-responders. (C) Waterfall plot illustrating best tumor shrinkage in individual patients, (D) Maximal tumor burden reduction. (E) Swimmer plot stratified by CTC molecular response. (F) Forest plot of PFS. AIC, Akaike information criterion; TKI, tyrosine kinase inhibitor. **p < 0.01.

Fig. 4. Epidermal growth factor receptor (EGFR) mutation profile of individual patients. (A) Mutation profile of all patients detected from tissue, circulating tumor DNA (ctDNA), and circulating tumor cell (CTC). CTC and ctDNA samples were collected at closely matched time points (≤ 7 days apart). (B) Original and T790M mutation profile in patients previously treated with EGFR–tyrosine kinase inhibitor. CCM-CTCD, Continuous Centrifugal Microfluidics–Circulating Tumor Cell Disc.

Fig. 1.

Fig. 2.

Fig. 3.

Fig. 4.

Circulating Tumor Cell–Based Molecular Responses Stratify EGFR-TKI Efficacy in Patients with EGFR-Mutant Lung Cancer

Characteristic	Total (n=77)	CTC molecular responder (n=25)	CTC molecular non-responder (n=47)
Age (yr)	65 (41-84)	68 (50-84)^a)	63 (41-82)^a)
Female sex	53 (68.8)	18 (72.0)	31 (66.0)
Smoking
Never smoker	55 (71.4)	18 (72.0)	33 (70.2)
Former/Current smoker	22 (28.6)	7 (28.0)	14 (29.8)
EGFR mutation type
Exon 19 deletion	49 (63.6)	16 (64.0)	30 (63.8)
L858R	28 (36.4)	9 (36.0)	17 (36.2)
T790M	23 (29.8)	9 (36.0)	12 (25.5)
Clinical stage
Stage IVA	34 (44.2)	12 (48.0)	19 (40.4)
Stage IVB	43 (55.8)	13 (52.0)	28 (59.6)
Line of treatment
1st line	53 (68.8)	15 (60.0)	34 (72.3)
2nd or more advanced	24 (31.2)	10 (40.0)	13 (27.7)
Treatment regimen
Gefitinib/Erlotinib/Afatinib	26 (33.7)	9 (36.0)	17 (36.2)
Osimertinib/Lazertinib	51 (66.2)	16 (64.0)	30 (63.8)
Palliative radiotherapy	17 (22.1)	4 (16.0)	12 (25.5)

	Total (n=77)^a)	CTC molecular responder (n=25)	CTC molecular non-responder (n=47)^a)
Best response, n (%)
Complete response	0	0	0
Partial response	55 (71.4)	18 (72.0)	33 (71.7)
Stable disease	18 (23.3)	7 (28.0)	10 (21.7)
Progressive disease	3 (3.8)	0	3 (6.5)
Objective response rate (%)	71.4	72.0	71.7
Disease control rate (%)	96.2	100	93.5

Table 1. Baseline characteristics

Values are presented as median (range) or number (%). CTC, circulating tumor cell; EGFR, epidermal growth factor receptor.

The median age of molecular responders was older than that of non-responders (p=0.007).

Table 2. Summary of treatment outcome

Values are presented as number (%). CTC, circulating tumor cell.

Before the first response assessment, one CTC non-responder patient expired due to septic shock, and was excluded from this analysis of best response.