Background Allergic rhinitis (AR) is a highly prevalent chronic inflammatory disease of the nasal mucosa and a growing public health, medical and economic problem worldwide. Currently, allergen-specific immunotherapy was referred to as etiological treatment, which could restore immune homeostasis but limited by specific allergens. The development of potential therapeutic targets and drugs for allergic rhinitis is urgently needed. Methods A proteome-wide Mendelian randomization (MR) study was used to identify potential therapeutic targets and drugs for AR. The statistics were obtained from three European AR consortiums and then validated the results by the FinnGen and UK Biobank cohorts. The genetic tools for 1699 plasma proteins were from genome-wide studies and validated by deCODE proteins. To ensure the robustness of the results and the reliability of the conclusions, we performed sensitivity analyses, including bidirectional Mendelian randomization, heterogeneity testing, pleiotropy testing, Steiger filtering, phenotype scanning and colocalization analyses. Results MR analysis revealed that expression of C1GALT1C1, CLEC5A and IL1R1 were associated with an increased risk of AR, while the expression of B2M, VCAM1 and IGF1R were associated with a decreased risk of AR. Then the small molecule ochratoxin-a, ZG-10, pidorubicin and the JAK3 inhibitor VI were identified to be potential drugs for AR treatment through joint analysis by connectivity map (cMAP) and selected plasma proteins. Conclusion The identified proteins are causally related to the risk of AR and are promising potential drug targets for AR. In addition, the identified small molecules show potential as effective drugs and warrant further clinical investigation.