Background: There were scarcely germline variants of familial lung cancer (LC) identified. We conducted an study with whole-exome sequencing of pedigrees with familial lung cancer to analyze the potential genetic susceptibility. Methods: Probands with the highest hereditary background were identified by our large-scale epidemiological study and five ones were enrolled as a learning set. The germline SNPs (single-nucleotide polymorphisms) of other five similar probands, four healthy individuals in the formerly pedigrees and three patients with sporadic LC were used as a validation set, controlled by three healthy individuals without family history of any cancer. The network of mutated genes was generated using STRING-DB and visualized using Cytoscape. Results: Specific and shared somatic mutations and germline SNPs were not the shared cause of familial lung cancer. However, individual germline SNPs showed distinct protein-protein interaction network patterns in probands versus healthy individuals and patients with sporadic lung cancer. SNP-containing genes were enriched in the PI3K/AKT pathway. These results were validated in the validation set. Furthermore, patients with familial lung cancer were distinguished by many germline variations in the PI3K/AKT pathway by a simple SVM classification method. It is worth emphasizing that one person with many germline variations in the PI3K/AKT pathway developed lung cancer during follow-up. Conclusions: The phenomenon that the enrichments of germline SNPs in the PI3K/AKT pathway might be a major predictor of familial susceptibility to lung cancer. © 2020, The Author(s).