Posttraumatic growth at end of life, a positive psychological consequence of dealing with a life crisis, may be catalyzed by the trauma of facing mortality. Studies on cancer patients' posttraumatic growth have overwhelmingly examined early-stage survivors. Of the few studies on advanced cancer patients, none anchored posttraumatic growth with the patient's death. Therefore, the purposes of this study were to describe longitudinal changes in and to identify determinants of posttraumatic growth over the dying process. A convenience sample of 313 cancer patients was recruited and followed until death. Posttraumatic growth was measured by the Posttraumatic Growth Inventory. Determinants were evaluated from sociodemographics, cancer characteristics, social context variables, and coping strategies using a multiple linear regression model. The dying process may be too threatening and stressful for cancer patients to experience profound posttraumatic growth as manifested by low Posttraumatic Growth Inventory scores (26.13 ± 21.59 to 40.33 ± 27.71 on a 0-105 scale) without significant changes as death approached. However, terminally ill cancer patients who were not defeated by the initial threat of death experienced posttraumatic growth. Patients were at risk for low-level posttraumatic growth if they were male, were non-middle aged, had low educational attainment, had a non-metastatic disease, recently recognized terminally ill status, had severe symptom distress, had high functional dependence and weak social support, and were unaware or had low acceptance of their prognosis. Interventions are needed to appropriately relieve symptom distress, facilitate functional independence, enhance social support, and improve prognostic awareness and acceptance to inspire positive changes for cancer patients at end of life.