Background Adolescent depression prevention research has focused on mean intervention outcomes, but has not considered heterogeneity in symptom course. Here, we empirically identify subgroups with distinct trajectories of depressive symptom change among adolescents enrolled in two indicated depression prevention trials and examine how cognitive‐behavioral (CB) interventions and baseline predictors relate to trajectory membership. Methods Six hundred thirty‐one participants were assigned to one of three conditions: CB group intervention, CB bibliotherapy, and brochure control. We used group‐based trajectory modeling to identify trajectories of depressive symptoms from pretest to 2‐year follow‐up. We examined associations between class membership and conditions using chi‐square tests and baseline predictors using multinomial regressions. Results We identified four trajectories in the full sample. Qualitatively similar trajectories were found in each condition separately. Two trajectories of positive symptom course (low‐declining, high‐declining) had declining symptoms and were distinguished by baseline symptom severity. Two trajectories of negative course (high‐persistent, resurging), respectively, showed no decline in symptoms or decline followed by symptom reappearance. Participants in the brochure control condition were significantly more likely to populate the high‐persistent trajectory relative to either CB condition and were significantly less likely to populate the low‐declining trajectory relative to CB group. Several baseline factors predicted trajectory classes, but gender was the most informative prognostic factor, with males having increased odds of membership in a high‐persistent trajectory relative to other trajectories. Conclusions Findings suggest that CB preventive interventions do not alter the nature of trajectories, but reduce the risk that adolescents follow a trajectory of chronically elevated symptoms.