Genetic context unmasks olfactory loss as a marker for dyskinesia risk via levodopa exposure trajectories in Parkinson's disease.

BACKGROUND: Olfactory impairment is common in Parkinson's disease (PD), but its prognostic value for levodopa-induced dyskinesia (LID) remains uncertain. We examined whether genetic background modifies the olfaction-LID association and how levodopa exposure influences this relationship. METHODS: We analyzed longitudinal PPMI data from 1254 participants with early PD (921 sporadic; 333 genetic carriers, including GBA, LRRK2, and other variants). Baseline olfaction was classified as normosmia, hyposmia, or anosmia using the University of Pennsylvania Smell Identification Test. The outcome was time from motor symptom onset to first LID. Cox models tested olfaction-by-genetics interactions with sequential adjustment for average, point, and time-varying levodopa equivalent daily dose (LEDD), adjusting for age at onset, sex, baseline MDS-UPDRS Part III, and Hoehn and Yahr stage. RESULTS: Olfactory category was not associated with LID across the whole cohort (P = 0.607). In genetic PD, hyposmia (HR 1.88, 95% CI 1.16-3.06; P = 0.011) and anosmia (HR 2.19, 95% CI 1.32-3.64; P = 0.003) were associated with higher LID hazard, whereas no significant associations were observed in sporadic PD (P ≥ 0.099). Genetic-by-olfaction interactions were significant without LEDD adjustment (hyposmia: HR 2.71, P = 0.002; anosmia: HR 2.61, P = 0.004). After accounting for changes in LEDD over time, these interactions were attenuated but remained approximately two-fold (hyposmia: HR 2.00, P = 0.049; anosmia: HR 2.04, P = 0.054). Time-varying LEDD was strongly associated with LID. CONCLUSIONS: Olfactory impairment was associated with LID primarily in genetic PD, and this pattern persisted after accounting for longitudinal levodopa exposure. These findings support a context-dependent rather than universal prognostic role of olfaction in PD.