524 - Prognostic impact of end-of-life chemotherapy in the last weeks for patients with advanced cancer

Background

Appropriately timed cessation of chemotherapy is important for patients with advanced cancer at the end-of-life stage. There are frequent discussions about over-treatment of end-of-life chemotherapy currently, but the definition of end-of-life chemotherapy remains unknown.

Methods

We analyzed patients’ data between August 2011 and August 2016. The primary endpoints were prognostic factors (age, sex, primary site, clinical stage, comorbidity, Eastern Cooperative Oncology Group (ECOG) performance status (PS), and Glasgow Prognostic Score (GPS)) following the last administration of chemotherapy within 14 days of death. Patients with either C-reactive protein levels >1.0 mg/dL or albumin levels <3.5 g/dL were classified as GPS1; patients with both were classified as GPS2, and those with neither of the two were classified as GPS0. The secondary endpoints were prognostic factors following the last administration of chemotherapy within 30 days of death. The associations between end-of-life chemotherapy and the frequency of end-of-life symptoms (delirium, cancer pain, dyspnea, nausea and vomiting, and fatigue) and treatment (hydration, continuous sedation, and opioids) were evaluated.

Results

We obtained 300 patients’ data including complete information about the last administration of chemotherapy. The number of patients within 14 and 30 days of death from the last administration of chemotherapy were 16 (5.3%) and 50 (16.7%), respectively. In multivariate analysis for end-of-life chemotherapy, ECOG-PS (odds ratio 0.26, p = 0.046) and GPS2 (odds ratio 0.19, p = 0.043) were significant prognostic factors within 14 days of death, while ECOG-PS (odds ratio 0.34, p = 0.046), GPS1 (odds ratio 0.28, p = 0.010), and GPS2 (odds ratio 0.22, p < 0.001) were significant prognostic factors within 30 days of death. The median survival time from the last administration of chemotherapy of patients with both GPS2 and EGOG-PS (≥2) was 41 days, and that of patients with both GPS0 and EGOG-PS (0-1) was 162.5 days. Prevalence rates at the end-of-life stage for nausea and vomiting (25.0%) within 14 days of death were significantly higher than those (7.4%) over 14 days of death. The mean amounts of hydration (0.50 L/day) at the end-of-life stage within 14 days of death were significantly higher than those (0.20 L/day) over 14 days of death.

Conclusions

GPS and ECOG-PS were significant prognostic factors for over-treatment of end-of-life chemotherapy. Information about these factors can aid clinical decision-making in individual patient risk stratification, especially in palliative care settings, and in the further development of prospective cohort studies about end-of-life chemotherapy.

Clinical trial identification

Legal entity responsible for the study

Mitsubishi Kyoto Hospital.

Funding

Has not received any funding.

Editorial Acknowledgement

The past report was that overly aggressive care is evident when measures such as chemotherapy given within last 14 days of life, new chemotherapy initiated within last 30 days of life. But it’s still defined uncertain. Therefore the aim of this study is to analysis relationship between prognostic factor and last chemotherapy for advanced cancer patients, to describe the end of life chemotherapy was the hypothesis patients underwent last chemotherapy was dead within last 14 days of life.

Disclosure

All authors have declared no conflicts of interest.