Purpose: Somatostatin receptor (SSTR) PET has become a mainstay in the diagnosis of neuroendocrine tumors (NETs) and for selecting patients for SSTR-based therapy; however, no consensus has yet been reached in terms of prognosis. A systematic review and meta-analysis was performed on the prognostic value of the maximum standardized uptake value (SUVmax) for Ga-SSTR PET in patients with NETs.
Patients and methods: We performed a systematic search using the following keywords: PET, SSTR, NET, and prognosis. The inclusion criteria were the use of Ga-SSTR PET as an imaging tool, studies limited to NETs, studies that reported progression-free survival (PFS) and/or overall survival (OS), and studies that included SUVmax as a prognostic parameter. The effect of SUVmax on PFS and OS was measured in terms of the hazard ratio (HR).
Results: Eight eligible studies with 474 patients were finally included and analyzed. The combined HR of SUVmax on PFS was 2.31 with significance (95% confidence interval [CI], 1.34-4.00; P = 0.003). The trim and fill adjusted analysis for SUVmax on PFS demonstrated the combined HR as 1.81 with significance (95% CI, 1.11-2.95; P = 0.017), as the publication bias was found (Egger P = 0.004). The combined HR of SUVmax on OS was 2.97 with significance (95% CI, 1.71-5.15; P = 0.0001), without publication bias (Egger P = 0.929). The subgroup analysis revealed that well-differentiated NETs (grade 1 or 2) on PFS showed significance (P = 0.03); however, all grades of NETs (including grade 3) on PFS did not reach significance (P = 0.11). Tumor site and type of radiotracer did not affect the prognostic value of SUVmax.
Conclusions: Low SUVmax of Ga-SSTR PET was associated with a worse prognosis for PFS and OS in patients with NETs. Well-differentiated NETs had more prognostic value compared with all grades of NETs. The SUVmax of Ga-SSTR PET could be used as an objective prognosis predictor.