Wiskott-Aldrich syndrome (WAS) is a multifaceted monogenic disorder with a broad disease spectrum and variable disease severity and a variety of treatment options including allogeneic hematopoietic stem cell transplantation (HSCT) and gene therapy (GT). No reliable biomarker exists to predict disease course and outcome for individual patients. A total of 577 patients with a WAS variant from 26 countries and a median follow-up of 8.9 years (range, 0.3-71.1), totaling 6118 patient-years, were included in this international retrospective study. Overall survival (OS) of the cohort (censored at HSCT or GT) was 82% (95% confidence interval, 78-87) at age 15 years and 70% (61-80) at 30 years. The type of variant was predictive of outcome: patients with a missense variant in exons 1 or 2 or with the intronic hot spot variant c.559+5G>A (class I variants) had a 15-year OS of 93% (89-98) and a 30-year OS of 91% (86-97), compared with 71% (62-81) and 48% (34-68) in patients with any other variant (class II; P < .0001). The cumulative incidence rates of disease-related complications such as severe bleeding (P = .007), life-threatening infection (P < .0001), and autoimmunity (P = .004) occurred significantly later in patients with a class I variant. The cumulative incidence of malignancy (P = .6) was not different between classes I and II. It confirms the spectrum of disease severity and quantifies the risk for specific disease-related complications. The class of the variant is a biomarker to predict the outcome for patients with WAS.