Immunotherapy has been widely used to treat non-small cell lung cancer (NSCLC) but is only effective in 20% of patients. Cyclin-dependent kinase inhibitor 2A (CDKN2A) is an important tumor suppressor gene, and its loss of function (LOF) is quite common in NSCLC. Pre-clinical studies suggest CDKN2A LOF promotes immune evasion; however, the results in relation to NSCLC are controversial, and debate continues as to the effect of CDKN2A LOF on immunotherapy. In this study, we collected the data of 49 CDKN2A LOF and 173 CDKN2A wild-type NSCLC consecutive patients treated by any line of immunotherapy. Through immunohistochemical (IHC) and immunofluorescent (IF) staining, we analyzed the CDKN2A predominant transcription protein p16INK4A in the CDKN2A LOF and CDKN2A wild-type NSCLC patients. Using Kaplan-Meier curves, we also examined the relationship between CDKN2A LOF and immunotherapy. The IHC and IF staining results showed that most CDKN2A LOF patients were p16INK4A negative, while most CDKN2A wild-type patients were p16INK4A positive. In the LOF group, five patients had partial responses, 35 had stable disease, and nine had progressive disease after the first evaluation of immunotherapy. The LOF group had a median progression-free survival (PFS) time of 4.67 months, while the wild-type group had a median PFS time of 8.63 months [hazard ratio (HR): 0.54; 95% confidence interval (CI): 0.38-0.77; P<0.001]. The LOF group had a median overall survival (OS) time of 9.07 months, while the wild-type group had a median OS time of 21.37 months (HR: 0.42; 95% CI: 0.29-0.61; P<0.001). Our study revealed that CDKN2A LOF NSCLC patients treated with immune checkpoint inhibitor (ICI) mono-therapy or combined therapy had a worse prognosis than those with CDKN2A wild-type NSCLC. However, our study also suggested that ICI could work quite effectively in selective CDKN2A LOF patients.