Objective Glomerular basement membrane (GBM) thickening is a typical and essential histopathological characteristic for the diagnosis of primary membranous nephropathy (PMN). The present study aimed to explore the relationship between GBM thickness and treatment response in PMN patients. Methods A total of 128 patients with nephrotic syndrome concurrent with PMN were studied. The highest GBM thickness was measured from at least five glomerular capillary loops using an electron microscope, and the mean value was obtained. Patients were categorized into three groups according to the tertiles of GBM thickness as follows: Group 1 (GBM thickness ≤ 1100 nm, n = 48), Group 2 (1100 nm < GBM thickness ≤ 1300 nm, n = 40), Group 3 (GBM thickness >1300 nm, n = 40). Clinicopathological features and treatment response were compared among the three groups. The associations of GBM thickness with complete remission (CR) were assessed by Cox proportional hazard analyses and a cubic spline curve. Results During a median follow-up period of 25.80 months, 69 (53.9%) patients achieved CR. Kaplan–Meier analysis showed that the non-CR probability was significantly higher in the highest tertile of GBM thickness (p˂0.001). Univariate Cox proportional hazard analysis indicated that GBM thickness was associated with CR (HR per SD 0.617, 95% CI [0.471–0.809], p˂0.001). After adjusting for age, duration of PMN, estimated glomerular filtration rate (eGFR), urinary protein excretion, grade of C3 deposition, and titer of serum anti-phospholipase A2 receptor (PLA2R) antibody, GBM thickness remained an independent predictor of CR (HR per SD 0.580, 95% CI [0.436–0.772], p˂0.001). Further multivariable-adjusted restricted cubic spline analysis confirmed a significant reverse linear association between GBM thickness and CR (p for nonlinear = 0.1261). Conclusions GBM thickness is an independent risk factor of CR. PMN patients with an increased level of GBM thickening at diagnosis have a lower probability of achieving CR.