CtDNA based molecular residual disease outcompetes carcinoembryonic antigen in predicting postoperative recurrence of non-small cell lung cancer.
J Thorac Dis
; 16(1): 423-429, 2024 Jan 30.
Article
em En
| MEDLINE
| ID: mdl-38410594
ABSTRACT
Background:
Carcinoembryonic antigen (CEA) has been routinely used as a postoperative monitoring biomarker for non-small cell lung cancer (NSCLC). Emergingly, circulating tumor DNA (ctDNA)-molecular residual disease (MRD) detection is a well-established prognostic marker, with better positive predictive value (PPV) and negative predictive value (NPV). However, the actual clinical efficiency of CEA in MRD context remain unknown. Hence, we conducted this study for direct comparison of CEA and MRD.Methods:
Two cohorts were analyzed in this study. To investigate the prognostic and predictive value of CEA, we retrospective enrolled NSCLC patient stage IA2-IIIA (8th tumor-node-metastasis staging system) with longitudinal CEA between 2018 and 2019. We also performed a paired comparison of CEA and MRD in our previous published cohort. 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:
In the retrospective cohort, the sensitivity of longitudinal CEA was only 0.49 [95% confidence interval (CI) 0.37-0.60]. Even for patients with progressively elevated CEA levels, 32% of them still remained disease-free, with PPV of 0.68 (0.49-0.83) and NPV of 0.81 (0.77-0.85). Furthermore, we then compared CEA and MRD values in a previously described MRD cohort. As expected, CEA levels could not stratify the risk of recurrence in detectable versus undetectable MRD populations.Conclusions:
MRD is superior to CEA in postoperative monitoring. there is insufficient evidence to support its use as postoperative monitoring tumor marker.
Texto completo:
1
Base de dados:
MEDLINE
Idioma:
En
Ano de publicação:
2024
Tipo de documento:
Article