e20617 Background: Pembrolizumab (P) administered every 3 weeks ± chemotherapy is a standard treatment option for advanced NSCLC. However, there have been no post-approval studies to determine the optimal frequency of administration or if longer intervals between administrations are effective. Methods: We conducted a retrospective review of patients with advanced NSCLC treated with P for at least 4 cycles at a single academic center (02/2016-12/2018). Patients received P-based regimens administered at a standard frequency (3 weeks ± 3 days, group A), ≥ 25% of total treatment cycles outside the standard frequency (group B), and < 25% of total treatment cycles outside the standard frequency (group C). We extracted demographic, tumor characteristic, treatment detail and outcomes data. Results: Of 84 P-treated patients, 43 (51%) were identified as receiving at least 4 cycles (groups A: 17, B: 10, C: 16). There were no significant differences in sex, stage at diagnosis, smoking status, tumor histology, driver oncogene mutations, PD-L1 expression, tumor mutation burden, line of therapy, performance status, or immune-related adverse events (irAEs). Patients in group A were more likely to receive P+chemotherapy (groups A: 41.2%, B: 10%, C: 6.3%; p = 0.03). The reasons for non-standard cycles were: patient-physician preference (65.4% patients), irAEs (15.4%), and non-irAE medical issues (42.3%). Time to treatment discontinuation was significantly longer in groups B & C receiving P-based therapy at longer non-standard intervals and with no statistically significant differences in overall survival. Conclusions: Our data, though limited by sample size and single institution design, shows that a significant proportion of patients receive P at extended intervals in routine clinical practice and with comparable outcomes as would be expected for those with advanced NSCLC receiving P at label-specified 3-week intervals. Given the durability of benefit seen in such patients, this requires confirmation in larger datasets and prospective trials so as to maximize patient experience and clinical outcomes while minimizing financial toxicity.