Circulating tumor cells in patients with recurrent or metastatic head and neck carcinoma: prognostic and predictive significance

Abstract

Introduction: We investigated the frequency of detection and the prognostic and predictive significance of circulating tumor cells (CTCs) in patients with recurrent/metastatic (R/M) head and neck carcinoma (HNC) before starting systemic therapy.

Patients and methods: Using the CellSearch technology, CTCs were assessed prospectively in peripheral blood of 53 R/M-HNC patients. We performed spiking experiments to test the diagnostic performance of the CellSearch platform in identifying squamous carcinoma cells.

Results: CTCs were identified in 14 (26%) and 22 (41%) patients at baseline and at any time point, respectively. In univariate analysis ≥2 CTCs had a poorer prognostic role than 0-1 CTC. In multivariate analysis, the presence of one CTC or more was associated with a poor prognosis both in terms of progression-free survival (PFS) [Hazard Ratio (HR): 3.068, 95% confidence interval (CI): 1.53-6.13, p 0.002] and overall survival (OS) [HR: 3.0, 95% CI: 1.48-6.0, p 0.002]. A disease control after systemic therapy was obtained in 8% of CTC-positive patients as opposed to 45% in CTC-negative ones (p 0.03). The epidermal growth factor receptor (EGFR) expression was identified in 45% of CTC-positive patients.

Discussion: In conclusion, CTCs are detected in one out of three patients with RM-HNC. CTC detection is a strong prognostic parameter and may be predictive of treatment efficacy. The frequency of EGFR expression in CTCs seems to be lower than that expected in the primary tumor.

Figure 1. Recovery efficiency of known numbers of spiked A-431 cells from 7.5 mL of blood.

The number of spiked cells is plotted against cells recovered by the CellSearch. The Pearson correlation value (R²) indicates strong correlation between spiked and recovered cells.

Figure 2. Kaplan-Meier curves of (A) progression-free survival and (B) overall survival in patients with 0, 1 and ≥2 CTCs.

Figure 3. Kaplan-Meier estimates of overall survival according to a combined risk factors model with Argiris factors and CTCs.

Continuous line indicates absence of both risk factors; small dotted line indicates the presence of only one of the two risk factors; large dotted line indicates the presence of both risk factors.

Figure 4. Association between the presence of CTCs before starting a new line of chemotherapy and response to treatment.

Higher response rate is observed in CTC-negative patients at baseline (A). Dynamic variation of CTCs numbers before and after treatment in patients (n = 10) with at least two determinations and at least one CTC at any time point. CTCs changes did not correlate with tumor response (B).

Figure 5. Example of CTCs analysis in a patient with mediastinal and axillary nodal metastases from an oropharyngeal squamous cell carcinoma.

(A) the CellSearch output of baseline CTC analysis showing two CTCs with heterogeneous EGFR expression. (B) Timeline of CTC analysis and treatments. (C) Correlative imaging analysis by CT/PET at baseline and after chemotherapy. In this patient 3 CTCs were detected at baseline. After 4 cycles of a chemotherapy, CTC number rised to 9 suggesting progressive disease then confirmed by CT/PET imaging.