Highlight
– Between 2015 and 2020 in Ontario, Canada, use of systemic anticancer therapy (SACT) in the last 30 days of life rose, driven largely by increased immunotherapy use.
– Receipt of any SACT in the final 30 days was associated with a more-than-twofold adjusted increase in high health services use (≥2 ED visits, ≥2 hospitalizations, or any ICU admission) and in-hospital death, relative to no SACT.
– All SACT types—chemotherapy, immunotherapy, targeted therapy, and combinations—showed elevated adjusted odds of aggressive, hospital-based care at end of life, underscoring the need to extend guideline recommendations to novel agents and to integrate prognostic conversations and palliative care.
Background
Use of systemic anticancer therapy (SACT) near the end of life (EOL) has long been a marker of potentially aggressive or low-value care. Traditional cytotoxic chemotherapy administered in the last weeks of life is associated with reduced quality of life, limited survival benefit, and greater use of emergency and inpatient services. Over the last decade, the therapeutic landscape shifted with the rapid uptake of immunotherapies and targeted agents that can be less acutely toxic and are often perceived as more tolerable. Whether these novel agents change the balance of benefit and harm when used in the last 30 days of life has been uncertain.
The population-based study by Iqbal and colleagues (J Clin Oncol. 2025) examines recent temporal trends in SACT use in the last 30 days of life and quantifies associations between different SACT classes and measures of high health services use and hospital death among adults with cancer in Ontario. Their findings are timely and relevant to clinicians, health system leaders, and guideline developers wrestling with how to apply EOL care recommendations in an era of rapidly evolving cancer therapeutics.
Study design
This was a population-based, retrospective cohort analysis of adults recorded in the Ontario Cancer Registry who died between March 2015 and March 2021 and had a cancer diagnosis within five years of death. The study included patients with solid tumors and hematologic malignancies who received any systemic anticancer therapy. Receipt of SACT within the last 30 days of life was classified into four mutually exclusive categories: chemotherapy alone, chemotherapy plus immunotherapy, immunotherapy alone, and targeted therapy alone.
Primary outcomes were indicators of high health services use in the last 30 days: multiple (≥2) emergency department (ED) visits, multiple (≥2) hospitalizations, or any intensive care unit (ICU) admission. Hospital death (death occurring in hospital) was a secondary but clinically important outcome. Segmented linear regression characterized monthly trends in SACT use over time; multivariable logistic regression estimated adjusted odds ratios (aORs) comparing each SACT category with no SACT, adjusting for available confounders.
Key findings
Population and overall patterns
The cohort comprised 68,963 decedents with cancer; 18,337 (26.6%) received at least one form of SACT in the last 30 days of life. Across the study period, the proportion of decedents receiving SACT in the final month increased significantly from March 2015 to March 2020 at a rate of 0.072 percentage points per month (P < .001). The upward trend was driven predominantly by immunotherapy alone, which rose by 0.064 percentage points per month (P < .001).
Associations with health services use and hospital death
Compared with patients who did not receive any SACT in the last 30 days of life, those who did had substantially higher adjusted odds of high health services use and of dying in hospital. Key adjusted odds ratios reported were:
- Chemotherapy alone: aOR for high health services use = 2.20; aOR for hospital death = 2.72.
- Chemotherapy plus immunotherapy: aOR for high health services use = 2.36; aOR for hospital death = 3.10.
- Immunotherapy alone: aOR for high health services use = 1.92; aOR for hospital death = 2.27.
- Targeted therapy alone: aOR for high health services use = 1.75; aOR for hospital death = 2.37.
These effect sizes indicate that, regardless of agent class, receipt of SACT in the last 30 days was associated with a roughly twofold (or greater) increase in measures of aggressive, hospital-based care.
Secondary and contextual observations
Although immunotherapy adoption rose markedly over the pre-pandemic period, immunotherapy recipients still had substantially elevated odds of hospital-based care compared with no SACT, albeit somewhat lower than for cytotoxic chemotherapy in some comparisons. Combination chemotherapy and immunotherapy showed the highest odds of hospital death in this analysis. The study timeframe included the early months of the COVID-19 pandemic; segmented regression showed increases up to March 2020, and the pandemic likely introduced access and practice changes afterwards that warrant cautious interpretation of later trends.
Interpretation and clinical implications
1. Novel agents have not eliminated aggressive, hospital-based EOL care. The finding that immunotherapy and targeted agents were associated with increased ED visits, hospitalizations, ICU admissions, and hospital death challenges the assumption that these treatments are intrinsically low-risk in the EOL context. Immune-related adverse events, treatment complications, symptom flares, and rapid clinical deterioration can all precipitate hospital care.
2. Increased use of SACT in the final month of life raises questions about appropriateness and prognostication. Some SACT administered in the last 30 days may be intended for symptom control or within clinical trials, but the observational design cannot distinguish intent. Nevertheless, the strong association with aggressive healthcare use suggests that a substantial proportion of these treatments may reflect continuing disease-directed care with limited patient-centered benefit.
3. Guidelines and quality metrics must evolve to incorporate novel therapies. Most quality statements that discourage chemotherapy in the last 30 days predate the immunotherapy and targeted therapy era or do not comment specifically on them. The results argue for explicit guideline language and quality measures that address all classes of SACT when assessing aggressive EOL care.
4. Integration of palliative care and improved communication are essential. The data support early and proactive goals-of-care conversations, routine prognostic awareness checks, and timely referral to palliative care, particularly when considering initiation or continuation of SACT in patients with limited life expectancy.
Expert commentary and limitations
Strengths of the study include its large, population-based sample and the contemporary timeframe capturing the rapid diffusion of immunotherapies. The use of objective, health administrative outcomes (ED visits, hospitalizations, ICU admissions, location of death) lends practical relevance for health system planners and payers.
Key limitations are those inherent to retrospective administrative data. Important clinical variables that influence treatment decisions and EOL trajectories—performance status, symptom burden, patient preferences, goals-of-care documentation, and the treating clinician’s intent—were not available. Residual confounding is therefore possible: patients selected to receive SACT near death may differ in unmeasured ways from those who did not. Causality cannot be inferred; SACT may be a marker of aggressive illness rather than the direct cause of increased hospital use. The study was conducted in Ontario, Canada; health system factors, access patterns, and cultural norms around EOL care may limit generalizability to other jurisdictions.
Nevertheless, the consistency of associations across SACT types and the magnitude of effect sizes underscore a robust signal that merits action in clinical practice and policy.
Recommendations for clinicians and policy makers
- Discuss prognosis and likely benefits and harms explicitly before initiating or continuing any SACT in patients with limited life expectancy; do not assume that newer agents obviate the need for such conversations.
- Incorporate palliative care earlier and more systematically for patients with advanced cancer who are candidates for SACT—this has been shown to improve symptom burden and may reduce aggressive EOL care.
- Update local and national guidelines and quality metrics to address immunotherapy and targeted agents, clarifying thresholds for considering treatment cessation when life expectancy is short.
- Develop decision aids and stopping-rule frameworks for clinicians and patients that account for the heterogeneous toxicity profiles and onset of benefit of modern SACT.
- At a system level, monitor SACT use in the last 30 days of life as a quality indicator, stratified by drug class and adjusted for patient case-mix.
Research gaps and future directions
Prospective studies and trials are needed to determine which patients, if any, derive a symptomatic or survival benefit from late-line SACT and to quantify harms more precisely. Studies should integrate patient-reported outcomes, performance status, and clinician intent. Implementation research can test whether structured prognostic communication and earlier palliative care reduce the use of SACT in the last month of life and ameliorate hospital-based care.
Conclusion
In this large, population-based study, use of systemic anticancer therapy in the last 30 days of life increased between 2015 and early 2020, largely driven by immunotherapy. Receipt of any SACT in the final month was associated with substantially higher odds of multiple ED visits, multiple hospitalizations, ICU admission, and death in hospital. These findings call for an update of EOL care guidelines to explicitly include novel anticancer agents and for strengthened practice strategies to align treatment decisions with patient goals, prognosis, and palliative care principles.
Funding and clinicaltrials.gov
The analysis summarized here is a population-based, observational study; it is not a registered clinical trial. For details on funding and disclosures, see the original publication: Iqbal J et al., J Clin Oncol. 2025;43(30):3279-3291 (doi: 10.1200/JCO-24-02816).
References
Please refer to the primary report for full author disclosures and detailed methods: Iqbal J, Moineddin R, Quinn KL, Booth CM, Earle CC, Lheureux S, Grant R, Lau J, Le LW, Tanuseputro P, Downar J, Rodin G, Seow H, Tsai J, Fowler RA, Hannon B, Krzyzanowska MK, Zimmermann C. Novel Systemic Anticancer Treatments and Health Services Use at the End of Life Among Adults With Cancer. J Clin Oncol. 2025 Oct 20;43(30):3279-3291. doi: 10.1200/JCO-24-02816. Epub 2025 Jun 4. PMID: 40466035; PMCID: PMC12527759.