Background: Immune checkpoint inhibitors (ICI) represent the backbone treatment of advanced non-small cell lung cancer (aNSCLC) patients. Emerging evidence suggests increased gut microbiome (GM) diversity is associated with favorable response. Conversely, antibiotic-induced dysbiosis may be associated with deleterious outcomes in patients receiving ICI in multiple retrospective studies and one prospective study. ¹⁸F-FDG physiologic colonic uptake on PET/CT increases following treatment with antibiotics and could be a surrogate marker for GM diversity and therefore clinical response. The aim of this study was to determine if ¹⁸F-FDG physiologic colonic uptake prior to ICI initiation correlates with outcomes and GM metagenomics in patients with advanced NSCLC. Methods: 71 patients with aNSCLC who underwent PET/CT prior to ICI were identified. For each patient, the colon was manually contoured, SUVmax was measured in each segment of the colon by a nuclear medicine specialist and average SUVmax was calculated for the whole colon. Patients were stratified in two groups according to median colon SUVmax (low vs high uptake). ¹⁸F-FDG physiologic colonic uptake was then compared to overall survival (OS), objective response (ORR), and progression-free survival (PFS). For patients with available stool samples (n = 10), GM composition was defined using metagenomics sequencing. Results: 71 patients (54% men, median age: 68 years) with aNSCLC were included in the study and ICI was the first line of therapy in 38% of those patients. The mean colon SUV for the low and high uptake groups were 1.41 (CI 95% 1.35-1.47) and 2.18 (CI 95% 1.90-2.46) respectively. The high uptake group had a higher proportion of non-responders (p = 0.033) and significant shorter PFS (4.1 months vs 11.3 months, p = 0.005). In the caecum, high uptake also correlated with numerically shorter OS (10.82 vs 27.56 months, p = 0.058) compared to low uptake group. Despite the low number of samples, metagenomics sequencing revealed that PLS-DA (Partial Least Squares Discriminant Analysis) for diversity was lower in the high SUV group (p = 0.008). Conclusions: Higher colon SUVmax on pre-ICI FDG PET/CT is associated with worse clinical outcomes and lower baseline GM diversity in patients with advanced NSCLC. Here, we propose that ¹⁸F-FDG physiologic colonic uptake on PET/CT could serve as a surrogate marker of GM diversity and predicts clinical outcomes.