Discovery of tumour specific peptides (neoantigens) generating target specific personalised T cell therapies is difficult because of the complex relationship between genetic mutations and neoantigen presentation. Studies predicting neoantigens have a 5% success rate, and direct observations of neoantigens are even rarer. We have developed a mass spectrometry proteomics guided approach to identify neoantigens in non-small cell lung cancer (NSCLC). We hypothesised that by characterising the observable peptides presented by HLA molecules (the immunopeptidome) we might better predict which mutations lead to neoantigens. From a cohort of 24 NSCLC patients, we performed a detailed analysis of 3 adenocarcinoma and 3 squamous cell carcinoma samples. Our workflow identified neoantigens generating functional T cell responses in 5 out of 6 patients, with one directly observed neoantigen, and an overall response rate of 20% of predicted neoantigens.