Biomarkers that show high sensitivity and specificity are needed for the early diagnosis and prognosis of cancer. An immune response to cancer is elicited in humans, as demonstrated, in part, by the identification of autoantibodies against a number of tumor-associated antigen (TAAs) in sera from patients with different types of cancer. Identification of TAAs and their cognate autoantibodies is a promising strategy for the discovery of relevant biomarkers. During the past few years, three proteomic approaches, including serological identification of antigens by recombinant expression cloning (SEREX), serological proteome analysis (SERPA) and, more recently, protein microarrays, have been the dominant strategies used to identify TAAs and their cognate autoantibodies. In this review, we aim to describe the advantages, drawbacks and recent improvements of these approaches for the study of humoral responses. Finally, we discuss the definition of autoantibody signatures to improve sensitivity for the development of clinically relevant tests.