The role of tumour metabolic and proliferative indices in predicting non-small-cell lung cancer (NSCLC) patients' prognosis is unclear. We correlated fluorine 18 ((18) F)-fluorodeoxyglucose (FDG)-positron emission tomography (PET) value and Ki67 index to patients' survival, taking into account tumour heterogeneity, disease characteristics and genetic aberrations. A series of 383 NSCLCs was arranged into tissue microarrays and Ki67 staining was analysed by immunohistochemistry. The maximum standardized uptake (SUV(MAX) ) value detected by (18) F-FDG-PET analysis was calculated over a region of interest. Large-cell and squamous cell carcinomas had higher proliferative and metabolic activities than adenocarcinomas, and the two measures were correlated significantly. The hot-spot Ki67 value was correlated with patients' survival and the cut-off to discriminate patients in the survival risk groups was 20%. Ki67 hot-spot values were greater in anaplastic lymphoma kinase (ALK) rearranged tumours. Adenocarcinomas showed the highest intratumour heterogeneity in proliferative activity and the hot-spot Ki67 value predicted only the prognosis of patients in this group. Although tumour metabolic activity was not associated with patients' prognosis, a SUV(MAX) > 2 was related to nodal metastases, tumour size and grade. Our results highlight how tumour heterogeneity influences evaluation of prognostic biomarkers. Our data support Ki67 evaluation to estimate NSCLC patients' prognosis, particularly for adenocarcinoma.
Read full abstract