Background: PIK3CA mutations are a rare oncogenic event of potential therapeutic relevance in NSCLC. Here we report frequency and characteristics of patients with PIK3CA mutated lung tumors. Methods: Patients with NSCLC and PIK3CA mutations were identified within our regional Network for Molecular Screening in Lung Cancer. We further analyzed the presence of BRAF, KRAS, EGFR mutations as well as ALK translocation, ERBB2 and FGFR1 amplifications in PIK3CA mutated samples. Clinical data on age, sex, TNM classification and tumor stage, histological type, grading, overall survival, smoking status, comorbidity, BMI and secondary malignancies were retrieved from clinical charts in accordance with the local ethics committee. Results: PIK3CA mutations were detected with a frequency of 3.7% (24% exon 20,76% exon 9) in 1000 patients. Histologically 32% were defined as squamous cell carcinoma, 48% as adenocarcinoma and 18% other histological subtypes or NSCLC-NOS. Exon 9 mutations were present in the acinar and lepidic subtype, whereas exon 20 mutations were seen in the papillary and solid subtype. Cooccuring genetic lesions were observed in 16% (mutations in KRAS=2, EGFR=1, BRAF=1; FGFR1 amplification=2). 14 were female, 23 male with a mean age of 69 years. 21 of these patients were further clinically annotated. 11 patients presented with stage IIIb/IV eligible for palliative treatment and 10 stage I – IIIa eligible for surgical therapy +/- adjuvant therapy. All but 1 patient were smokers with an average BMI of 26,2kg/m2 with a typical high load of comorbidity mainly of cardiovascular diseases, 8 of 21 patients showed prior malignancies in their medical history. The median overall survival within this population has not been reached yet. Conclusions: Screening for PIK3CA mutations is feasible. A high proportion (38%) of patients with PIK3CA mutated lung cancer have prior malignancies and show a high load of comorbidity. Furthermore PIK3CA mutations are not exclusive to KRAS, EGFR or BRAF mutations or FGFR1 amplifications. Successful identification of patients with oncogenic lesions in lung cancer in a screening network might allow future personalized treatment of these patients.