Abstract: The present disclosure provides methods for determining whether a patient diagnosed with lung cancer will benefit from or is predicted to be responsive to treatment with a therapeutic agent that targets HER2. These methods are based on detecting elevated levels of HER2 dimerization in a biological sample obtained from a lung cancer patient. Kits for use in practicing the methods are also provided.

Abstract: Improved methods for treating lung cancer are provided. Tumor samples from patients are analyzed (i) by DNA sequencing to detect the presence of HER2 mutations and (ii) by mass spectrometric proteomic analysis to determine whether HER2 protein is expressed in the tumor cells. Patients respond to therapy with trastuzumab emtansine (T-DM1) or an equivalent antibody-drug conjugate when unique HER2 protein fragments are detected in the patient's tumor cells that harbor HER2 mutations. Conversely, patients do not respond to T-DM1 therapy when the tumor cells contain HER2 mutations but the unique protein fragments are not detected. Detection of HER3 in the tumor cells is a positive predictor of response to treatment.