Previous genome-wide association studies investigating the relationship between the HLA-DRB1 and the risk of Parkinson's disease (PD) have shown limited racial diversity and have not explored clinical heterogeneity extensively. The study consisted of three parts: a case-control study, a cross-sectional study, and a longitudinal cohort study. The case-control study included 477 PD patients and 477 healthy controls to explore the relationship between rs660895 and PD susceptibility. The cross-sectional study utilized baseline data from 429 PD patients to examine the correlation between rs660895 and PD features. The longitudinal study included 388 PD patients who completed a 3-year follow-up to investigate the effects of rs660895 on PD progression. In the case-control study, HLA-DRB1 rs660895-G allele was associated with a decreased risk of PD in allele model (adjusted OR=0.72, p = 0.003) and dominant model (AG + GG vs. AA: adjusted OR = 0.67, p = 0.003). In the cross-sectional analysis, there was no association between rs660895 and the onset age, motor phenotype, or initial motor symptoms. In the longitudinal analysis, PD patients with the G allele exhibited a slower progression of motor symptoms (MDS-UPDRS-III total score: β = -5.42, p < 0.001, interaction ptime × genotype < 0.001) and non-motor symptoms (NMSS score: β = -4.78, p = 0.030, interaction ptime × genotype < 0.001). Our findings support HLA-DRB1 rs660895-G allele is a protective genetic factor for PD risk in Chinese population. Furthermore, we also provide new evidence for the protective effect of rs660895-G allele in PD progression.