18F-fluorodeoxyglucose positron emission tomographic scan in solid-type p-stage-I pulmonary adenocarcinomas: What can produce false-negative results?

OBJECTIVE: False-negative (FN) uptake of 18F-fluorodeoxyglucose (FDG) can be divided into those cases related to technological limitations of positron emission tomography (PET) and those related to inherent properties of neoplasms. Our goal was to clarify possible factors causing FN PET results in patients with solid-type pulmonary adenocarcinomas (PAs). METHODS: From January 2007 to December 2014, of the 255 patients with p-stage-1 non-small-cell lung cancer observed and treated (surgically) in our institution, we retrospectively reviewed the PET/computed tomography (CT) records, the clinical information, the preoperative thin-section CT images, and the pathological features [classified by the International Association for the Study of Lung Cancer/ American Thoracic Society/European Respiratory Society (IASLC/ATS/ERS) subtyping criteria] of 94 consecutive solid-type p-stage-1 PAs. Univariable and multivariable logistic analyses were used to identify and weigh the independent predictors of the PET findings using the following variables: body weight, blood glucose level, tumour size, tumour location, and histological classification. RESULTS: There were 58 men and 36 women (mean age = 68.7 ± 8.9 years, range 42-85). Considering the maximum standardized uptake value (SUVmax) >2.5 as a 'PET-positive' result, 77 lesions (81.9%) proved PET positive and 17 lesions (18.1%), PET negative (with SUVmax < 2.5). Overall, the median SUVmax value was 5.7 [interquartile range (IQR) 2.8-10.3]. Higher SUVmax values (P < 0.001) were observed in those PAs larger than 2 cm in their major axis (median SUVmax = 9.0; IQR 4.6-14.6); in PAs < 2 cm, the median SUVmax was 4.1; IQR 2.2-5.9. When clustering the cohort in two histological classes (class A, colloid/mucinous/lepidic versus class B, micropapillary/ solid/acinar/papillary), the radiometabolic patterns were significantly different (median SUVmax = 2.8; IQR 1.7-4.9 in class A vs median = 7.4 IQR 4.5-13.9 in class B, P < 0.001). Significant PET FN rates were reported in (i) PAs measuring < 2 cm in their major axis (27.9%), (ii) lesions located in the lower zones of the lung (31.0%), and (iii) class A tumours (37.5%). In the multivariable logistic analysis, histological type (IASLC/ATS/ERS aggregated clusters) proved to be the only independent relevant factor for determining whether PET results were negative or positive (OR:7.23, 95% CI: 2.05-25.43, P = 0.002). CONCLUSIONS: The IASLC/ATS/ERS pattern significantly influences FDG uptake in solid-type p-stage-1 PAs. The fact that colloid/mucinous/ lepidic adenocarcinomas have a notable tendency to produce negative findings on PET scans warrants particular attention.