Background: Interest has surged for KRAS mutated (KRAS^mut) non-small-cell lung cancer (NSCLC) after approval of the covalent KRAS^G12C inhibitors sotorasib and adagrasib. It remains unclear, how prognosis of these tumors is influenced by co-mutations and which cases could benefit most from novel drugs alongside immunotherapy (IO) or chemo-immunotherapy (CHT-IO). Methods: This retrospective study included all NSCLC patients with KRAS^mut NSCLC treated in the Thoraxklinik Heidelberg from 01/2014 until 01/2021. For molecular profiling, PCR-based next-generation sequencing (NGS) was performed with a 40-gene panel, including TP53, KEAP1 and STK11. Date of progression was verified through reevaluation of radiologic images by the investigators according to RECIST v1.1. Stratification was performed according to the type of KRAS mutation (G12C vs. other), presence of TP53, STK11 and KEAP1 co-mutations (vs. wild-type [WT] status), and the PD-L1 tumor proportion score (TPS < 1 aka PD-L1^neg, vs. TPS 1-49, vs. TPS 50+ aka PD-L1^high). Results: Among 370 identified patients, no differences (all p-values > 0.08) were observed between KRAS^G12C (n = 163) and KRAS^non-G12C(n = 207) regarding clinicopathological features, like age (median 65 years), sex, smoking status, initial ECOG performance status, PD-L1 TPS, type of treatment in any line, number of treatment lines, and overall survival (OS, 19.4 months in median; details in the poster). In contrast, there was a strong association of PD-L1 TPS with OS (20.2 vs. 9.7 months for PD-L1^high vs. PD-L1^neg, p = 0.011) and progression-free survival (PFS, 5.7 vs. 1.9 months, p = 0.004) under IO. In addition, specifically within the PD-L1^neg subset, OS was longer for KRAS^G12C compared to KRAS^other patients (22.6 months vs. 12.1 months, p = 0.032). This was driven by the longer PFS under CHT-IO (9.7 vs. 4.5 months for KRAS^G12Cvs.KRAS^other, p = 0.005), particularly in the absence of TP53 mutations (8.5 vs. 4.0 months with p = 0.004 for TP53^wt patients; p > 0.50 for TP53^mut) and KEAP1 mutations (11.3 vs. 5.1 months with p = 0.01 for KEAP1^wt patients; p > 0.70 for KEAP1^mut). Also, across all patients, KEAP1^mut cases showed a trend for shorter PFS under CHT-IO (2.7 vs. 8.0 months, p = 0.07) and IO (1.9 vs. 3.9 months, p = 0.081), as well as shorter OS (15.9 vs. 19.5 months, p = 0.068) compared to KEAP1^wt, in contrast to STK11^mut cases (p = 0.42-0.92 for the same comparisons). Overall, KRAS^mut NSCLC after chemotherapy and immunotherapy showed poor prognosis under standard therapies with a median PFS of 3 months, and a median OS of 6.8 months. Conclusions: In PD-L1^neg NSCLC, KRAS^G12Cis associated with better outcome under CHT-IO and longer OS compared to KRAS^other, especially in the absence of TP53 and KEAP1 co-mutations. KEAP1, but not STK11 mutations are associated with impaired benefit from (CHT-)IO in KRAS^mut NSCLC. Treatment of KRAS^mut NSCLC after chemo-immunotherapy represents an unmet medical need.