Total-body dynamic PET/CT of micro-metastatic lymph node in a patient with lung cancer

Abstract

Lymph nodemicro-metastasis (LNMM)was previously reported to be significantly correlated with recurrence of lung cancer [1]. Conventional whole-body PET/CT has limited specificity and sensitivity in the qualitative diagnosis of small lesions and micro-lymph nodes but total-body PET/CT, in which the entire body is scanned in a single shot using a 194-cm axial PET scanner, has opened new avenues for lesion detection [2]. It has been reported that quantitative Ki parametric map is more sensitive than SUV in normal size tumors [3]. Therefore, it comes as no surprise that total-body PET affords improved dynamic reconstruction and provides improved image quality as compared to conventional PET [4]. In this work, we report on the use of dynamic total-body PET and computed Ki parametric map to improve early detection of micro-metastatic lymph nodes as compared to conventional static whole-body PET. A 49-year-old man with a 1.8-cm nodule in the left lower lobe nodule detected on lung CT was admitted for further evaluation. Thin-section CT scan was performed (Fig. 1a, b) along with total-body dynamic [F]FDG PET to investigate the presence of metastases. A micro-lymph node in the left lower hilar (diameter about 4 mm) with higher metabolism was considered marginally detectable when considering conventional static metrics of standard uptake values (SUVmax = 3.5) (Fig. 2a, b). In contrast, the Ki parametric map (Kimean = 0.0145 ml/g/min, Kimax = 0.0156 ml/g/min) (Fig. 3a, b) provided much greater confidence in the presence of metastatic disease in the lower left hilum associated with a greater contrast-to-noise ratio (CNR) detectability as compared to SUV: The CNR computed over multiple noise realizations generated from the listmode acquired data was 13.06 ± 3.47 in the Ki map (Fig. 5b) as compared to 4.58 ± 0.87 in the SUVmax map (Fig. 5a). This is a significant and substantial increase in contrast to noise ratio as compared to conventional PET. The micro-lymph node was subsequently resected and referred to the pathology department which confirmed the presence of abnormal epithelial cells with cord-like or patchy distribution consistent with metastatic tumor (Fig. 4a (HE × 100), 4b (HE × 200)). This case shows that dynamic total-body PET/CT yielded substantial and significant increase in sensitivity that not only makes dynamic whole-body parametric imaging feasible, but also makes marginally detectable lesions far easier to detect even at the early stage of micro-metastatic lymph nodes as compared to traditional SUV.