Background: KRAS is the most frequently mutated oncogene in NSCLC, and such tumors lack effective targeted therapy. Unlike other activating oncogenes associated with never/light-smokers (NS), KRAS mutations are found in both never/light-smokers and smokers. However, the relationship between smoking status and a KRASm+ tumor’s genomic complexity is unclear. Methods: Targeted next generation sequencing (NGS) data at our institution from 7/13-1/16 was reviewed to identify KRASm+ NSCLC tumors. All patients with a < 10 pack-year (py) smoking history (NS) and a subset of heavy smokers (HS) ( > 40 py) were identified, with clinical and genomic analysis. Fisher’s exact test was used to compare frequency of gene mutations and the Mann-Whitney test was used to compare the distribution of total variants between cohorts. Results: 22 NS and 44 HS patients were evaluated. In the NS cohort, the median age was 64.5y with 17/22 women (77%), 21/22 adenocarcinomas (95%), and limited KRAS variants (G12A = 3, G12C = 6, G12D = 11, G12V = 2). In the HS cohort, the median age was 62.5y with 25/44 women (57%), 43/44 adenocarcinomas (98%), and a wide range of KRAS variants (G12A = 2, G12C = 20, G12D = 4, G12R = 1, G12S = 1, Q61L = 1, G12V = 7, Q61H = 3, G13D = 5) (variant distribution between cohorts p = .005). HS patients were more likely to be diagnosed with Stage IV disease (28/44, 64%) v NS patients (11/22, 50%). Compared to NS patients, tumors from HS patients were also genomically more complex, with increased total nucleotide variants (median = 10 v 5, p < .001). HS tumors were more likely to have: a) concurrent deleterious mutation or 2-copy deletion in TP53 or STK11 (25/44, 57% v NS tumors (6/22, 27%) (.04) and b) concurrent mutations/2-copy deletions in > 1 tumor suppressor (TS)/tumor (panel of TP53, STK11, APC, CDKN2A/B, RB)(15/44, 34% v 1/22, 5%) (p = .01). Finally, median OS for HS patients with Stage IV disease was 9.0m v 14.9m in NS patients (HR = 0.86). Conclusions: The genomic landscape of KRASm+ NSCLC from HS patients is distinct from NS patients and includes increased total mutations and frequency of TS loss. These differences suggest NS KRASm+ tumors may be a genetically distinct cohort within the broader context of KRASm+ NSCLC.