Oncogenic EGFR mutations, in particular those in tyrosine kinase domain, have been confirmed to associate with sensitivity to tyrosine kinase inhibitors (TKIs) and are indicators for prescribing TKIs in non-small-lung cancer (NSCLC). However, patients received the TKIs showed diverse responses, indicating the functional restriction of genetic test to guide the TKI prescription. Several studies reported that the expression levels of mutations in DNA, RNA and proteins levels are not always quantitatively correlated. Thus, we aimed to develop a mass spectrometry (MS)-based proteogenomics strategy for multiplexed screening of somatic EGFR mutations in protein level. The proteogenomics strategy integrated bioinformatics analysis of mutant EGFR protein sequences, affinity purification of EGFR protein complex, parallel enzymatic gel-assisted digestions, LC-DDA-MS/MS and LC-PRM-MS analyses, customized database searching using multiple engines, and construction of reference intervals using synthesized isotopic standard peptides, for unambiguous identification and absolute quantification of wild-type and mutant EGFR proteins. The strategy offered unambiguous identification of peptides covering 34 mutated and 33 wild-type sites in EGFR as well as EGFR-interacting proteins. The quantitation results revealed concomitant and heterogeneous expressions of mutated and wild-type EGFR proteins in a series of NSCLC cells and xenograft tumors that harbored different EGFR genotypes, suggesting diverse and heterogenous expressions of the same EGFR mutations in individual tumors. Our developed proteogenomics strategy can robustly determine the multiple somatic mutations in EGFR at protein level which may provide a better prediction on adaption to TKI treatments and improve the understanding towards the molecular impact of EGFR mutations during cancer progression.