CtDNA based molecular residual disease outcompetes carcinoembryonic antigen in predicting postoperative recurrence of non-small cell lung cancer

Introduction

Background

Serum carcinoembryonic antigen (CEA) is a well-known tumor biomarker in patients with non-small cell lung cancer (NSCLC) (1-3). It is also a predictor of disease recurrence and progression (4,5). Thus, CEA is routinely used by clinicians in some regions for postoperative surveillance. Emergingly, circulating tumor DNA (ctDNA)-based molecular residual disease (MRD) detection has been proposed to predict disease recurrence for NSCLC patients after surgery (6,7).

Rationale and knowledge gap

In our previous study (8), we highlighted the predictive value of longitudinal MRD in the postoperative setting, with a positive predictive value (PPV) of 89.1% and negative predictive value (NPV) of 96.8%. However, other studies have shown limited value in consecutive CEA monitoring during follow-up (9). For predicting recurrence in patients after radical resection, the accuracy of serum CEA level was uncertain, considering that the sensitivity was only 46% (10). The direct comparison of CEA and MRD has not been reported.

Objective

In this particular context, we evaluated the advantage of ctDNA-MRD could be an alternative to CEA. Thus, in this study, we compared CEA with ctDNA-MRD as a predictor for recurrent NSCLC disease. We present this article in accordance with the STARD reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-23-507/rc).

Methods

Study design and patients

This study was performed with two independent cohorts (Figure 1). To investigate the prognostic and predictive value of CEA, we retrospectively collected the data of NSCLC patients who underwent radical resection at Guangdong Provincial People’s Hospital from 2018 to 2019. Patients who met the following criteria were included: (I) pathological staging of IA2–IIIA [8th tumor-node-metastasis (TNM) staging system], (II) pathological diagnosis of NSCLC, and (III) at least two serum CEA follow-up tests within 2 years after surgery. Patients without available clinical and radiological record was excluded.

Figure 1 Flow chart of this study. CEA, carcinoembryonic antigen; MRD, molecular residual disease.

The paired comparison of longitudinal CEA and MRD was investigated in the prospective cohort (8). In this cohort, we screened patients with at least two serum CEA follow-up tests after surgery. The data of 204 patients with matched CEA and MRD data were analyzed. Recurrence was confirmed based on radiology.

The primary endpoint of this study was the sensitivity, specificity, PPV and NPV.

The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). This study was approved by the ethics and scientific committees of Guangdong Provincial People’s Hospital (No. KY-Q-2021-202-01). Informed consent for the publication of details relating to an individual person has been obtained.

CEA and ctDNA-MRD detection

Serum CEA levels were measured at the clinical laboratory of Guangdong Provincial People’s Hospital using chemiluminescence immunoassays. The cutoff value of CEA was set at 5 mg/L. The methods used for ctDNA-MRD detection have been previously described (8). The landmark CEA level was defined as the CEA level within 3 months after surgery. Longitudinal CEA was defined as the dynamic follow-up CEA record since landmark detection. Results of those examination were assessed independently. Analysis of the PPV and NPV was completed for patients with at least 180 days of follow-up since the correspondent CEA or MRD status.

Statistics

Statistical analyses were performed using R software (version 4.1.2) and GraphPad Prism 9. Survival data were analyzed using the Kaplan-Meier method, and comparisons were performed using the log-rank test. Sensitivity, specificity, PPV and NPV were calculated using the R package “epiR”. McNemar’s test was used to analyze the paired data. Statistical differences were set at a P value <0.05.

Results

Unsatisfactory monitoring value of CEA

The clinical characteristics of the cohorts are listed in Table 1. In the retrospective cohort, 346 patients were enrolled. The median follow-up time was 40.6 months, and 80 patients had disease relapse by June 11, 2022. Among 268 patients with landmark CEA records, a significant superior disease-free survival (DFS) was observed for patients with normal CEA levels as opposed to that for patients with abnormal CEA levels [not reached vs. 37.6 months, P<0.001, hazard ratio (HR) =0.16, 95% confidence interval (CI): 0.14–0.44, Figure 2A]. However, in terms of recurrence predictive efficiency, the sensitivity was only 0.27 (95% CI: 0.16–0.39), and the PPV was 0.55 (95% CI: 0.36–0.73, Table 2). Moreover, in the context of longitudinal CEA, 24 additional patients had abnormal CEA levels in the subsequent follow-up. As expected, the longitudinal CEA also had a significant prognostic value (not reached vs. 37.6 months, P<0.001, HR =0.20, 95% CI: 0.11–0.37, Figure 2B). However, the sensitivity was only close to one-half (0.49, 95% CI: 0.37–0.60, Table 2), and the PPV of longitudinal CEA was not significantly high (0.57, 95% CI: 0.44–0.68, Table 2). We Further investigated CEA in adenocarcinoma and the other pathological type. CEA elevation mainly occurred in adenocarcinoma and was a significant prognostic factor (HR =0.075, 95% CI: 0.039–0.146, P<0.001, Figure 2C) while not a prognostic factor for other pathological type (P=0.345).

Figure 2 Postoperative CEA and MRD monitoring for NSCLC patients after radical resection. (A) Kaplan-Meier curve of DFS stratified by landmark normal and abnormal CEA levels in the retrospective cohort. (B) Kaplan-Meier curve of DFS stratified by longitudinal normal and abnormal CEA levels in the retrospective cohort. (C) The proportion of recurrence in the longitudinal normal and abnormal CEA groups, and the outcome for patients with progressively elevated CEA levels in the retrospective cohort. (D) The matched 2×2 tables for longitudinal CEA and longitudinal MRD in patients with and without disease relapse. (E) The Kaplan-Meier curve of DFS stratified by longitudinal CEA and longitudinal MRD. (F) Representative case from the prospective cohort. CEA, carcinoembryonic antigen; DFS, disease-free survival; CEA+, abnormal CEA level; CEA−, normal CEA level; MRD−, undetectable longitudinal MRD; MRD, molecular residual disease; MRD+, detectable longitudinal MRD; NSCLC, non-small cell lung cancer.

Furthermore, we focused on patients with progressively elevated CEA levels. Of the 69 patients with abnormal postoperative CEA levels, 31 showed progressively elevated levels. However, 32.3% (10/31) of the patients remained disease-free, with a PPV of 0.68 (95% CI: 0.49–0.83) (Figure 2D, Table 2). Overall, these data suggested that postoperative CEA did not provide satisfactory value for monitoring disease progression in NSCLC patients.

MRD outperforms CEA in predicting postoperative recurrence

In the prospective cohort, there were 204 patients with matched longitudinal CEA data. The sensitivity, specificity, PPV, and NPV of longitudinal MRD and CEA in this cohort are shown in Table 2. The sensitivity and specificity of MRD were significantly better than those of CEA (P<0.001) (Figure 2E). As shown in Figure 2F, no significant difference was observed between abnormal and normal CEA levels in the MRD detectable or undetectable groups (P=0.17 in detectable MRD group, P=0.82 in undetectable MRD group).

Next, we focused on one representative case in this cohort. This patient was a 56-year-old female diagnosed with lung adenocarcinoma (pT2aN0M0, stage IB), harboring an EGFR L858R mutation. After complete resection, this patient underwent adequate imaging and follow-up with evaluation of CEA and MRD. Interestingly, this patient exhibited progressive elevation of CEA levels after resection, but multiple imaging scans, including positron emission tomography/computed tomography and MRD, did not detect any signs of recurrence. The endoscopic and ultrasound examination also excluded second primary tumors from the digestive and reproductive systems.

Discussion

Key findings

In this study, we analyzed the monitoring value of CEA in the postoperative context. CEA did not effectively predict disease recurrence for NSCLC patients after surgery, even for those with progressively elevated CEA levels. Moreover, MRD outperformed CEA in the paired analysis.

Strengths and limitations

Our study reported the comparison of longitudinal CEA and MRD data in a prospective cohort, which provided high quality evidence for MRD monitoring after radical resection. The present study has several limitations. The sample size was relatively small and the follow-up duration was not long enough.

Comparison with similar researches

Several studies have indicated that patients with abnormal CEA levels had a poorer prognosis (10,11). Buccheri et al. (9) showed that the specificity and sensitivity of CEA were 90% and 46%, respectively, in 118 patients who underwent resection, which is consistent with our results observed in the retrospective cohort. Although the specificity of CEA was 0.89, only 50% of the patients with recurrence had abnormal CEA levels. Even progressively increasing CEA levels did not improve the PPV to a satisfactory level. Generally, CEA level is a prognostic factor for NSCLC after radical surgery, but it is not an effective indicator of recurrence.

Explanation of findings

Considering the advantage of MRD, CEA levels were not a satisfactory predictor of recurrence in patients with NSCLC after radical resection. Since CEA could not further stratify the risk of recurrence in the longitudinal detectable MRD and undetectable populations, there was little benefit from CEA monitoring in MRD context.

Implications and actions needed

MRD monitoring in an emerging and strong predictor for NSCLC recurrence with high NPV and PPV, which might further stratify patients and even provide instruction for adjuvant therapy. Larger study for further confirmation is required.

Conclusions

In conclusion, although CEA is a well-established tumor biomarker in NSCLC, there is insufficient evidence to prove the monitoring value of CEA in the postoperative context. On the other hand, MRD had obvious advantages in terms of sensitivity, specificity, PPV, and NPV.

Acknowledgments

Funding: This work was supported by the Guangdong Provincial Key Laboratory of Lung Cancer Translational Medicine (Grant No. 2017B030314120 to W.Z.Z.), the National Natural Science Foundation of China Major Joint Project on Key scientific issues of Lung Cancer (Grant No. 82241235 to W.Z.Z.), the Guangdong Basic and Applied Basic Research Foundation (Grant No.2019B1515130002 to W.Z.Z.) and the High-level Hospital Construction Project (Grant No. DFJH201801). The funding sources had no role in the preparation of this manuscript.

Footnote

Reporting Checklist: The authors have completed the STARD reporting checklist. Available at https://jtd.amegroups.com/article/view/10.21037/jtd-23-507/rc

Data Sharing Statement: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-23-507/dss

Peer Review File: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-23-507/prf

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jtd.amegroups.com/article/view/10.21037/jtd-23-507/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 (as revised in 2013). This study was approved by the ethics and scientific committees of Guangdong Provincial People’s Hospital (No. KY-Q-2021-202-01). Informed consent for the publication of details relating to an individual person has been obtained.

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. Grunnet M, Sorensen JB. Carcinoembryonic antigen (CEA) as tumor marker in lung cancer. Lung Cancer 2012;76:138-43. [Crossref] [PubMed]
2. Okamura K, Takayama K, Izumi M, et al. Diagnostic value of CEA and CYFRA 21-1 tumor markers in primary lung cancer. Lung Cancer 2013;80:45-9. [Crossref] [PubMed]
3. Yuan J, Sun Y, Wang K, et al. Development and validation of reassigned CEA, CYFRA21-1 and NSE-based models for lung cancer diagnosis and prognosis prediction. BMC Cancer 2022;22:686. [Crossref] [PubMed]
4. Kozu Y, Maniwa T, Takahashi S, et al. Risk factors for both recurrence and survival in patients with pathological stage I non-small-cell lung cancer. Eur J Cardiothorac Surg 2013;44:e53-8. [Crossref] [PubMed]
5. Yamazaki M, Ishikawa H, Kunii R, et al. A combination of preoperative CT findings and postoperative serum CEA levels improves recurrence prediction for stage I lung adenocarcinoma. Eur J Radiol 2015;84:178-84. [Crossref] [PubMed]
6. Chaudhuri AA, Chabon JJ, Lovejoy AF, et al. Early Detection of Molecular Residual Disease in Localized Lung Cancer by Circulating Tumor DNA Profiling. Cancer Discov 2017;7:1394-403. [Crossref] [PubMed]
7. Gale D, Heider K, Ruiz-Valdepenas A, et al. Residual ctDNA after treatment predicts early relapse in patients with early-stage non-small cell lung cancer. Ann Oncol 2022;33:500-10. [Crossref] [PubMed]
8. Zhang JT, Liu SY, Gao W, et al. Longitudinal Undetectable Molecular Residual Disease Defines Potentially Cured Population in Localized Non-Small Cell Lung Cancer. Cancer Discov 2022;12:1690-701. [Crossref] [PubMed]
9. Buccheri G, Ferrigno D. Identifying patients at risk of early postoperative recurrence of lung cancer: a new use of the old CEA test. Ann Thorac Surg 2003;75:973-80. [Crossref] [PubMed]
10. Numata T, Endo T, Yanai H, et al. Serum CEA and CYFRA Levels in ALK-rearranged NSCLC Patients: Correlation With Distant Metastasis. In Vivo 2020;34:2095-100. [Crossref] [PubMed]
11. Shintani T, Matsuo Y, Iizuka Y, et al. Prognostic Significance of Serum CEA for Non-small Cell Lung Cancer Patients Receiving Stereotactic Body Radiotherapy. Anticancer Res 2017;37:5161-7. [Crossref] [PubMed]