FairTTE is the first comprehensive framework for analyzing fairness in time-to-event prediction using medical imaging, revealing widespread bias across modalities and the limitations of existing fairness methods, especially under distribution shifts.