Abstract
TMI has been shown to be a prognostic factor in early stage NSCLC (Muley et al., Lung Cancer, 60: 408-415, 2008).
The objective of the study was to analyse the value of TMI to predict tumour relapse in completely resected p-stage I NSCLC patients.
112 patients entered the study (71 male/41 female; 37 SCC, 59 AC, 16 other NSCLC; relapse: n=33; death: n=20; median follow-up: 39.3 months). Preoperative markers were measured with immunoassays (Roche, Mannheim, Germany). TMI is defined as the geometric mean of normalized CEA and CYFRA 21-1 values. Statistical analysis was done with SPSS 18.0 (Chicago, USA).
CEA (HR: 2.3, p=0.015) and CYFRA 21-1 (HR: 3.0, p=0.002) differed significantly between patients with low and high risk of tumour relapse. The differentiation was improved (HR: 3.6) by using TMI (cut-off 0.58). 3-year disease-free survival was 83.8% and 54.0%, respectively (p=0.001). 8/53 patients in the low risk group and 25/59 patients in the high risk group had tumour recurrence. The overall survival was 95.4% in low risk and 77.5% in high risk patients (p=0.001). TMI was found to be a significant prognostic factor (DFS, OS) in multivariate analyses. A small effect of adjuvant chemotherapy (n=20) could be seen in the high risk group. The relapse rate was 2/8 patients with adjuvant chemotherapy compared to 23/51 patients without adjuvant chemotherapy. In contrast, patients receiving adjuvant chemotherapy in the low risk group had a higher rate (5/12) compared to patients without adjuvant chemotherapy (3/41).
Patients with elevated TMI levels were shown to be at an increased risk of relapse and might therefore be appropriate candidates for adjuvant therapy.
- © 2011 ERS
