Although various molecules have been identified as being potential biomarkers of psoriasis, reliable biomarkers for psoriasis severity and therapeutic assessment are still lacking. Betacellulin (BTC) is a membrane protein that belongs to the epidermal growth factor family of peptide ligands, the members of which participate in skin homeostasis. Transcriptome analysis has demonstrated BTC gene downregulation in psoriasis. However, the association of BTC with psoriasis severity and potential BTC use in psoriasis assessment and treatment remain unclear. In this study, BTC gene and protein expression was found to be reduced in skin lesions of psoriasis patients. Furthermore, we observed a significant negative correlation between BTC expression and the PASI and psoriasis-associated genes, including NOS2, IL-23, IL-17A, and IL-17C. Subsequently, we investigated the diagnostic value of BTC expression in plaque psoriasis by using a ROC curve analysis, and the optimal cutoff value for distinguishing patients with plaque psoriasis from healthy individuals was BTC < 5.495, with an area under the ROC curve (AUC) of 0.9413, a sensitivity of 90.39% and a specificity of 90.08%. Moreover, BTC levels in the lesional skin of patients with plaque psoriasis were significantly increased following treatment. Although there was no difference in BTC expression in lesional skin before treatment between responders and nonresponders, BTC expression was significantly correlated with PASI in responders after treatment (but not in nonresponders), suggesting that BTC may not serve as a predictor of treatment response (but rather reflects disease severity) in patients with plaque psoriasis.