IMPORTANCE: Considerable efforts have been dedicated to improving the quality of end-of-life care among patients with advanced cancer in the past decade. Whether the quality has shifted in response to these efforts remains unknown. OBJECTIVE: To examine contemporary patterns of end-of-life care among patients with advanced cancer. DESIGN, SETTING, AND PARTICIPANTS: This retrospective cohort study used a recent linkage of Surveillance, Epidemiology, and End Results and Medicare data to characterize patterns of end-of-life care. The cohort included fee-for-service Medicare decedents aged 66 years or older who were originally diagnosed with distant-stage breast, prostate, pancreatic, or lung cancers and died between 2014 and 2019. Analyses were conducted between June 1, 2023, and July 31, 2024. MAIN OUTCOMES AND MEASURES: Outcomes included monthly use of acute care, systemic therapy, and supportive care (ie, palliative and hospice care and advanced care planning) in the last 6 months of life. Additionally, a claims-based indicator was evaluated of potentially aggressive care in the last 30 days of life, defined as experiencing more than 1 acute care visit, in-hospital mortality, late receipt of systemic therapy, or hospice entry. RESULTS: The study included 33 744 Medicare decedents with advanced cancer (mean [SD] age, 75.7 [6.9] years; 52.1% male). From 6 months before death to month of death, there was an increase in the mean (SE) number of acute care visits (from 14.0 [0.5] to 46.2 [0.5] per 100 person-months), hospice use (from 6.6 [0.4] to 73.5 [0.5] per 100 person-months), palliative care (from 2.6 [0.2] to 26.1 [0.6] per 100 person-months), and advanced care planning (from 1.7 [0.6] to 12.8 [1.1] per 100 person-months). Overall, 45.0% of decedents experienced any indicator of potentially aggressive care. CONCLUSIONS AND RELEVANCE: This study found persistent patterns of potentially aggressive care, but low uptake of supportive care, among Medicare decedents with advanced cancer. A multifaceted approach targeting patient-, physician-, and system-level factors associated with potentially aggressive care is imperative for improving quality of care at the end of life.