(1) Background: Little is known about how the COVID-19 pandemic has impacted social support and loneliness over time and how this may predict subsequent mental health problems. This study aims to determine longitudinal trajectories of social support and loneliness in the French general population during the first year of the COVID-19 pandemic and study whether variations in these trajectories are associated with symptoms of depression and anxiety; (2) Methods: Analyses were based on data from 681 French participants in the international COVID-19 Mental Health Study (COMET) study, collected at four periods of time between May 2020 and April 2021. Group-based trajectory modelling (GBTM) was used to determine social support and loneliness trajectories. Associations between the identified trajectories and symptoms of depression and anxiety, measured with the Patient Health Questionnaire (PHQ-9) and Generalized Anxiety Disorder scale (GAD-7), were tested through multivariate linear regression models; (3) Results: Social support trajectories revealed four stable groups: 'poor' (17.0%), 'moderate' (42.4%), 'strong' (35.4%) and 'very strong' (5.1%). Loneliness trajectories also identified four groups: 'low stable' (17.8%), 'low rising' (40.2%), 'moderate stable' (37.6%) and 'high rising' (5.0%). Elevated symptoms of depression were associated with poor social support as well as all identified loneliness trajectories, while high levels of anxiety were associated with moderate stable and high rising loneliness trajectories; (4) Conclusions: High and increasing levels of loneliness are associated with increased symptoms of depression and anxiety during the pandemic. Interventions to address loneliness are essential to prevent common mental health problems during the pandemic and afterwards.