This study aimed to identify latent classes of suicidal ideation trajectories among university students and to examine the predictors of class membership, thereby providing longitudinal evidence regarding suicide risk in this population. Given the increasing concerns about suicide among young adults, this research provides critical insights into subgroup-specific risk factors. Data were collected from 300 undergraduate students enrolled in 4-year universities across the nation at three time points: T1 (January 2025), T2 (April 2025), and T3 (June 2025). Suicidal ideation was measured using the Suicidal Ideation Attributes Scale (SIDAS; 11-point Likert scale), and the sum score (range: 0-50) of its five sub-dimensions (frequency, uncontrollability, closeness to attempt, distress, interference with daily activities) was utilized. Psychological distress was assessed with the Brief Symptom Inventory-18 (BSI-18; 5-point Likert scale), focusing on four sub-dimensions : depression, somatization, anxiety, and panic. Analyses were conducted using Growth Mixture Modeling (GMM), which classified participants into three groups: C1 (I=29.30***, S=-0.90*), C2 (I=2.76***, S=-0.64***), and C3 (I=17.17***, S=-0.39). Accordingly, C1 was designated as the "High-decreasing group," C2 as the "Low-increasing group," and C3 as the "Moderately stable group." Regarding predictors, when the "High-decreasing group" was used as the reference group, only somatization (OR: 0.202, 95% CI: 0.067-0.609) significantly predicted membership in the "Low-increasing group." This finding suggests that somatic symptoms are a salient factor associated with elevated risk of suicidal ideation. Furthermore, when the "Low-increasing group" was employed as the reference group, depression (OR: 2.663, 95% CI: 1.418-5.001) significantly predicted membership in the "Moderately stable group." This indicates that depressive symptoms may play a pivotal role in maintaining a moderately elevated level of suicidal ideation. Collectively, these findings suggest that the predictors of suicide risk differ across subgroups and underscore the importance of interventions tailored to the specific characteristics of each group.