Rheumatoid Arthritis (RA) is a chronic, inflammatory, autoimmune disease affecting diarthrodial joints and extra-articular tissues; in the absence of an effective treatment, it is characterized by persistent symmetrical and erosive synovitis which leads to structural joint damage and lifelong disability. Several autoantibodies have been associated with RA such as rheumatoid factor (RF) and anti-citrullinated protein antibodies (ACPA). B cells have been shown to play a crucial role in the pathogenesis of RA by producing autoantibodies and promoting synovial inflammation through antigen presentation, T cells activation and cytokines production [1]. Although biologic agents have notably improved disease outcome and patients' quality of life, currently around 30-40% of subjects do not respond to treatment and the mechanisms leading to resistance are still not known [2]. For this reason, new prognostic biomarkers and predictors of response are needed. We and others have postulated that the development of biomarkers for patients' stratification prior therapeutic intervention may be possible through a better understanding of the different histopathological patterns present both in early and established individual RA patient and the related underlying cellular and molecular mechanisms. To date, Tumor Necrosis Factor (TNF)-α has been shown to be one of the master elements of inflammation in RA; however, even though therapies aimed at blocking this key cytokine have emerged as a major tool in the treatment of RA, a large proportion of patients (approximately 30-40%) do not achieve a meaningful clinical response assessed by either the American College of Rheumatology (ACR) or the European League Against Rheumatism (EULAR) criteria. The same limitation can be applied to the use of rituximab, a chimeric monoclonal antibody directed against CD20, which is uniquely expressed by all B-lymphocytes during the maturation process from late stage pro-B cells to memory cells. The clinical efficacy of rituximab has been proved by several clinical studies [3] but it is highly variable. Currently, NICE guidelines recommend rituximab in patients with inadequate response to a first-line biologic therapy, including at least one anti-TNFα agent independently of their pre-treatment chance to respond. In all cases, whether considering biologics used for several years in RA patients (anti-TNFα or rituximab) or relatively newer biologics in clinical use (i.e. tocilizumab, an anti-interleukin (IL) -6 receptor blocking monoclonal antibody or abatacept, a CTLA4 inhibitor fusion protein designed to target the T cell co-stimulatory signal mediated through the CD28-CD80/86 pathway), no validated biomarkers predictive of clinical response currently exist. Consequently, to date a "trial-and-error" approach is used in the prescription of biologics in RA, which has the obvious disadvantage of potentially exposing patients to drugs that they may not respond, with potential unnecessary side-effects, delaying use of an effective treatment and causing a significant economic burden to society. Therefore, identifying pre-treatment predictors of response with a customized stratification approach would be of invaluable importance in RA, also in consideration of the large number of biologics in development targeting novel pathways currently being tested in clinical trials. In this manuscript, we review existing data and provide future perspectives with regard to the role of synovial histopathology as a potential prognostic biomarker for patient stratification in RA, in particular regarding the use of specific biologic therapies.