Folate receptor-positive circulating tumor cells in predicting ground-glass nodules malignancy

Am J Transl Res. 2023 Apr 15;15(4):2828-2835. eCollection 2023.

Abstract

Objective: To study the clinical significance of folate receptor-positive circulating tumor cells (FR+CTCs) in determining malignancy of ground-glass nodules (GGNs) and assess the added value of FR+CTC in the classic GGN evaluation model (Mayo Model).

Methods: Sixty-five patients with single indeterminate GGN were recruited. Twenty-two participants had benign/pre-malignant diseases, and forty-three had lung cancers, as confirmed by histopathology examination. FR+CTC was enumerated by CytoploRare® Kit. A CTC model was drawn based on the multivariate logistic analysis. The area under the receiver operating characteristic curve (AUC) was analyzed to evaluate the diagnostic performance of FR+CTC, CTC model and Mayo model.

Results: In the cohort, the mean age of 13 males and 9 females with benign/pre-malignant diseases was 57.7 ± 10.2 years. The mean age of 13 males and 30 females with lung cancers was 53.8 ± 11.7 years. There was no significant difference between the age and the smoking history (P=0.196 and P=0.847, respectively). FR+CTC can effectively differentiate lung cancer from benign/pre-malignant diseases [sensitivity: 88.4%, specificity: 81.8%, the AUCs was 0.8975, 95% confidence interval (CI): 0.8174-0.9775] in patients with GGN. Multivariate analysis revealed that FR+CTC level, tumor size, and tumor location were independent predictors of GGN malignancy (P<0.05). The prediction model based on these factors showed better diagnostic efficiency than the Mayo model (AUC: 0.9345 vs. 0.6823), yielding superior sensitivity (81.4% vs. 53.5%) and specificity (95.5% vs. 86.4%).

Conclusion: The FR+CTC exhibited a promising potential in determining the malignancy of indeterminate GGNs, and the CTC model's diagnostic efficiency was superior to the Mayo model.

Keywords: Circulating tumor cell; folate receptor; ground-glass nodule; malignancy; prediction model.