Chemotherapy Near the End of Life for Chinese Patients with Solid Malignancies

Abstract

Introduction

There are increasing concerns about the negative impacts of chemotherapy near the end of life (EOL). There is discrepancy among different countries about its use, and little is known about the real‐world situation in China.

Patients and Methods

This retrospective study was conducted at six representative hospitals across China. Adult decedents with a record of advanced solid cancer and palliative chemotherapy were consecutively screened from 2010 through 2014. The prevalence of EOL chemotherapy within the last 1 month of life was set as the primary outcome. The correlations among EOL chemotherapy, clinicopathological features, and overall survival (OS) were investigated.

Results

A total of 3,350 decedents who had had cancer were consecutively included; 2,098 (62.6%) were male and the median age was 56 years (range, 20–88). There were 177 (5.3%), 387 (11.6%), and 837 (25.0%) patients who received EOL chemotherapy within the last 2 weeks, 1 month, and 2 months of life, respectively. We identified inferior OS (median OS, 7.1 vs. 14.2 months; hazard ratio, 1.37; 95% confidence interval [CI], 1.23–1.53; p < .001), more intensive treatments (e.g., admitted to intensive care unit [ICU] in the last month of life, received cardiopulmonary resuscitation and invasive ventilation support), and hospital death (odds ratio, 1.53; 95% CI, 1.14–2.06; p = .005) among patients who received continued chemotherapy within the last month compared with those who did not. However, subgroup analyses indicated that receiving oral agents correlated with fewer ICU admissions and lower rates of in‐hospital death.

Conclusion

This study showed that EOL chemotherapy is commonly used in China. Intravenous chemotherapy at the EOL significantly correlated with poor outcomes and the role of oral anticancer agents warrants further investigation.

Introduction

For patients with incurable solid cancer, palliative chemotherapy is often prescribed with the promise of survival extension and improvement in quality of life [1–3]. Along with the increase in therapeutic possibilities, there has been a recognition of incorporating high‐quality end‐of‐life (EOL) care into routine clinical work. However, the role of EOL chemotherapy is not fully understood [4], nor are the underlying reasons contributing to the overuse of EOL chemotherapy [5]. Meanwhile, it has been reported that EOL treatment for patients with advanced cancer is becoming increasingly aggressive in Western countries [6]. Several studies have explored the association between continuing EOL chemotherapy and unfavorable outcomes, including reduced survival and lower quality of life (QoL) [7, 8]. On the other hand, early incorporation of palliative care can alter chemotherapy use toward the EOL and improve the QoL, mood, and overall survival (OS) [9, 10]. There are increasing calls for establishing a new model to improve the EOL quality of care and clinical outcomes for our patients. To avoid futile EOL chemotherapy, the first steps are to recognize the situation and subsequent outcomes and to identify the potential associated factors.

Although evidence has suggested that EOL care varies geographically, data on the real‐world situation in China are limited [11]. This study, therefore, was designed to describe the prevalence of EOL chemotherapy administered to patients with advanced solid cancers. The potential associations between EOL chemotherapy and EOL care and between EOL chemotherapy and prognosis were also analyzed.

Patients and Methods

Study Population

This was a retrospective study of patients who died of advanced (metastatic or recurrent) solid tumors from January 2010 to December 2014. To be eligible, patients with advanced solid cancer must have had a pathological diagnosis, been at least 18 years old, received chemotherapy with palliative intent, and had documented death due to cancer progression. The trajectory of hematological malignancies is different, in general, from that of patients with solid tumors. In addition, most of the treatment for advanced hematological malignancies is curative. Chemotherapy was more often given to patients who had a more aggressive disease course and considerable chance to achieve complete remission through the curative treatment. Considering these reasons, we set the study exclusion criteria as follows: (1) patients with hematologic malignancy (e.g., leukemia, lymphoma, multiple myeloma); (2) patients who only received local treatment, such as radiotherapy, transcatheter arterial chemo‐embolization, or radiofrequency ablation; (3) patients who received concurrent chemoradiotherapy or any treatment with curable intent; and (4) documented death but not due to cancer progression (e.g., a car accident).

Patients’ demographic and disease characteristics were identified from electronic medical records from six hospitals, including two comprehensive cancer centers—Sun Yat‐sen University Cancer Center and Cancer Hospital of Henan Province—and four general hospitals: The Fifth Affiliated Hospital of Sun Yat‐sen University, The First People’s Hospital of Zhongshan, Peking University Shenzhen Hospital, and The First Affiliated Hospital of Guangzhou University of Chinese Medicine. All of the six hospitals are teaching hospitals associated with a medical college. Additional demographic information was obtained manually from patient charts as required.

Measures and Covariates

Based on previous reports, we focused on chemotherapy use near the EOL, including the proportion of patients receiving chemotherapy within 14 days, 1 month, and 2 months of death. We also designed the study to identify other indicators of aggressive care, including the proportion of emergency room (ER) visits or intensive care unit (ICU) admissions in the last month of life, the proportion of patients who underwent cardiopulmonary resuscitation (CPR) or invasive ventilation support, and the place of death. OS was calculated as the interval (in months) from the time of diagnosis of advanced disease to death. The primary objective was describing the prevalence of EOL chemotherapy for Chinese patients with advanced solid malignancy, with the primary outcome being EOL chemotherapy during the last 1 month of life.

In summary, we collected information on each patient’s sex, age at diagnosis, comorbidities (identified from the International Classification of Diseases, ninth revision, codes for the primary and secondary diagnoses), cancer type, education level, and family history of cancer. Cancer diagnoses were stratified by six types: lung cancer, gastrointestinal tumors, head and neck cancers, gynecological tumors, breast cancer, and a group encompassing other types of cancer, including dermatoma, urologic neoplasms, and thymoma. Antineoplastic agents, which were perceived as “chemotherapy” in this study, included alkylating agents, antimetabolites, anticancer antibiotics, and alkaloids. Hormonal agents, such as aromatase inhibitors and antiandrogens, were also included as anticancer treatments for advanced breast cancer and prostate cancer. Targeted agents, such as small‐molecule tyrosine kinase inhibitors (TKIs) and antibodies, were taken into account during data analyses. No immune checkpoint inhibitors were involved in this cohort. However, immunotherapy for advanced melanoma and renal carcinoma were included according to guidelines and physician’s preference. Because early palliative care has been shown to improve survival and QoL and to reduce the use of intravenous anticancer regimens without impact on oral chemotherapy use in the last days [9, 10], we also analyzed the impact of administration type of final EOL chemotherapy (i.v. or oral) on prognosis.

This study was approved and documented by the institutional review board of each enrolled hospital. Requirements for written informed consent were waived because all personal data were de‐identified before the analyses.

Statistical Analysis

Data were expressed as mean, median, or percentage. Chi‐ square analyses and logistic regression were carried out to explore the factors associated with continuing EOL chemotherapy and to calculate the odd ratios (ORs) and 95% confidence intervals (CIs). Survival analysis was conducted using Kaplan‐Meier method. Cox proportional hazards model was used to calculate hazard ratios (HRs). All analyses were performed using SPSS version 20.0 (IBM Corp., Armonk, NY, http://www.ibm.com/) and Excel for Windows version 13 (Microsoft Corp., Redmond, WA, https://www.microsoft.com). Statistical significance was set at p ≤ .05.

Results

Patients

A total of 3,350 patients from January 2010 to December 2014 were consecutively enrolled. The basic characteristics of enrolled patients are summarized in Table 1. There were 2,098 (62.6%) male and 1,252 (37.4%) female patients and the median age was 56 years (range, 20–88 years). For all patients, the median time between last chemotherapy and death was 23.1 weeks (range, 0.8–222.0 weeks). The mean number of regimens was 2.70 ± 0.06 and of chemotherapy cycles was 7.75 ± 0.14. We observed that up to 85.2% of patients (n = 2,854) received the last chemotherapy intravenously. Among 1,031 patients (30.8%) who progressed on two or more prior lines of chemotherapy, the numbers of patients who had received three or four lines of chemotherapy were 220 (6.6%) and 117 (3.5%), respectively. In addition, patients with lung cancer (n = 921; 27.5%), gastrointestinal cancers (n = 1,294; 37.4%), and head and neck cancers (n = 567; 16.9%) composed the majority.

Prevalence of EOL Chemotherapy and Potential Associated Factors

The proportions of patients who received chemotherapy during the last 2 weeks, 1 month, and 2 months of life were 5.3% (n = 177), 11.6% (n = 387), and 25.0% (n = 837), respectively. More than one third of patients received chemotherapy during the last 3 months (n = 1,202; 35.9%). However, this result varied significantly among different cancer types. As shown in Figure 1, patients with lung cancer experienced the most aggressive chemotherapy near the EOL: 72 patients with lung cancer (7.8%) received chemotherapy within 2 weeks of death, and 156 cases (16.9%) received anticancer agents within 1 month of death. Although male patients tended to receive more aggressive chemotherapy, this was not confirmed by multivariate analyses (p = .435). Comorbidity status or family history were not associated with EOL chemotherapy (p = .942 and p = .246, respectively). Younger patients and those who had higher levels of education or who had received more than two lines of chemotherapy had increased probability of continuing chemotherapy within the month before death (Table 1). A remarkable difference in EOL chemotherapy use was observed between comprehensive cancer centers and general hospitals (10.8% vs. 14.1%; p = .014).

Association Between Aggressive Palliative Chemotherapy and Poor Prognosis, and Aggressive Care Near Death

Another aim of this study was to identify the potential impact of continuing EOL chemotherapy on prognosis. The median OS of the whole population was 13.7 months (95% CI, 13.1–14.0 months). We found that aggressive EOL chemotherapy was associated with inferior OS. Supplemental online Figure 1 shows the survival curve of two groups. For those continuing chemotherapy within 2 weeks or 1 month of EOL, median OS was 6.8 months (95% CI, 5.1–8.5 months) and 7.1 months (95% CI, 6.4–7.8 months), respectively. Conversely, those who ceased chemotherapy at the corresponding time had significantly longer survival times (median OS, 13.9 and 14.2 months, respectively). The HRs for the risk for death were 1.48 (95% CI, 1.27–1.72; p < .001) and 1.37 (95% CI, 1.23–1.53; p < .001), respectively.

To investigate the consistency of aggressive EOL chemotherapy as a prognostic factor, we performed subgroup analyses based on patients’ characteristics. As summarized in supplemental online Figure 2,continuing chemotherapy within the last month or 2 weeks of EOL was consistently associated with inferior survival in most subgroups. Interestingly, we found an exception in the breast cancer group: the risk for death was reduced 40% or 54% for patients who continued EOL chemotherapy within 1 month or 2 weeks, respectively.

In terms of aggressive care indicators, including ICU admission, ER visit during the last month of life, and intensive support (i.e., CPR or invasive ventilation) near death, the prevalences were all significantly higher in those continuing chemotherapy within the last month of life (Table 2). Patients who continued chemotherapy within the last month of life had significantly higher risk for ICU admission (OR, 10.7; 95% CI, 6.5–17.6; p < .001) and were more likely to have gone to the emergency room (OR, 2.64; 95% CI, 2.10–3.31; p < .001). In terms of place of death, 44.2% of patients who received EOL chemotherapy within the last month of life died in the hospital, whereas only 16.6% of those who ceased chemotherapy died in the hospital (OR, 1.53; 95% CI, 1.14–2.06; p = .005).

Table 3 summarizes the differences in aggressive care indicators between intravenous and oral anticancer agents. We found that patients with breast cancer received less i.v. chemotherapy at the EOL stage (OR, 0.27; 95% CI, 0.10–0.79; p = .017), whereas patients continuing oral agents had lower risk for hospital death (OR, 0.38; 95% CI, 0.18–0.79; p = .009) and fewer ICU admissions (OR, 0.30; 95% CI, 0.13–0.73; p = .008). Continuing i.v. chemotherapy during the last month or 2 weeks of life was a universally negative prognostic factor (Fig. 2).

Discussion

The prescription of chemotherapy for patients with advanced cancer has increased with the development of novel and high‐ efficacy anticancer agents, along with increasing concerns about its negative effects on survival and QoL at the EOL stage. It has been shown that early involvement of palliative care will improve the QoL and even prolong the overall survival for patients with advanced non‐small cell lung cancer (NSCLC) [9, 10]. To better understand the current situation of EOL chemotherapy for the population of Chinese patients with cancer and to explore appropriate cutoff for improving EOL care, we conducted this investigation with consecutive data from 3,350 Chinese patients who had died of cancer. The descriptive analyses indicated that 5.3%, 11.6%, and 25.0% of patients were prescribed chemotherapy within the last 2 weeks, 1 month, and 2 months, respectively (Fig. 1). Those continuing chemotherapy at the EOL experienced more extensive aggressive management and impaired overall survival.

In general, the prevalence of EOL chemotherapy was in line with recent published data. The current relative situation in China was not striking. We found that 387 patients (11.6%) continued chemotherapy within the last month of life, in accordance with the results from a single‐center investigation in Switzerland [12]. However, a study from Portugal showed that the prevalence of EOL chemotherapy within the last month near death was as high as 37% [13]. Several reasons contribute to this discrepancy. Resource availability, social culture environment, and patient characteristics vary among countries and hospitals [14]. It has been recognized that in well‐resourced countries, patients have a high rate of exposure to futile treatments and medicalized death [15]. China is castigated for inefficient use of scarce resources and for far‐from‐satisfactory performance of standardized palliative care for patients with cancer patient near EOL [16]. Therefore, no evidence yet links the relatively lower prevalence of EOL chemotherapy with the quality of palliative care in China.

In reality, evidence from Western countries has revealed that patients fail to get or accept the truth of palliative intent and believe their disease was supposed to be curative [17]. Patients are still willing to accept cytotoxic agents even if the potential benefit of chemotherapy is pessimistically presented by their physicians [5, 18]. However, the factors influencing patients’ decision on anticancer regimens are complex and unidentified in China. Moreover, Chinese medicine is worthy of attention because it is often resorted to by patients with cancer through a self‐help process intimately associated with deep cultural grounding [19]. Based on the experience of professional oncologists, approximately 88% of patients with terminal cancer ever received traditional Chinese medicine. In addition, compared with those from cancer‐specialty hospitals, we found a higher prevalence of EOL chemotherapy and better prognosis in patients from general hospitals. According to data in supplemental online Table 1, patients’ characteristics may account for this result. On the other hand, adjuvant therapy along with palliative chemotherapy may vary among different hospitals. However, further interpretation of the role of traditional Chinese medicine at the EOL was not available because of the lack of detailed records.

Hypothetically, the use of anticancer agents near the EOL is not a guarantee of benefit [20]. Our findings refuted the positive effect of EOL chemotherapy on survival prolongation and demonstrated that overuse of chemotherapy significantly correlated with more‐invasive rescue management near death in or out of the hospital. In accordance with what we found, a prospective study has revealed that patients receiving palliative chemotherapy were more likely to die in an intensive care unit rather than at home, compared with those not receiving this treatment [21]. Through comparison, we found the median age of our cohort was relatively younger than what others have reported [4, 12, 13]. In our study, patients younger than 56 years underwent more aggressive EOL chemotherapy but presented significantly better survival. Potential reasons were analyzed to interpret this contradiction. Among these were that a higher proportion of younger patients were well educated, and they were more inclined to receive Western medicine rather than traditional Chinese medicine, which may explain why the prevalence of EOL chemotherapy was higher in this group. Patients in the younger group had less comorbidity and thereby could receive more lines of cytotoxic agents.

It was not possible to compare many major prognostic features among these patients. Therefore, the result of factors associated with reduced survival with chemotherapy at the EOL should be considered preliminary. Given the potential heterogeneity, we conducted subgroup analyses to validate the consistency of aggressive EOL chemotherapy as a negative prognostic factor. Our results indicated that continuing chemotherapy within the last days of life was associated with poor survival, irrespective of the number of prior treatments and many other clinicopathological factors. However, patients with more aggressive disease may be treated more intensively under certain circumstances. Overuse of chemotherapy was associated with poor prognosis, but the causation needs to be further demonstrated.

This study also demonstrated several factors associated with the probability of continuing chemotherapy, such as cancer type, age, stage at diagnosis, education level, and type of hospital. Whether specific clinicopathological factors correlate with aggressive chemotherapy near the EOL remains controversial [7, 22–26]. In our study, the prevalence of EOL chemotherapy was prominent in patients with lung cancer, which was diametrically opposite the prevalence in patients with breast cancer. Continuing EOL chemotherapy was consistently associated with poor survival in most subgroups, except breast cancer. In response to concerns about survival improvement with the availability of more active agents in the course of treatment [27], further analyses were performed to explore potential reasons. In our study, patients who received two or more lines of chemotherapy had better OS (supplemental online Table 2). Moreover, patients with breast cancer received more lines of prior therapy but less i.v. chemotherapy within the last days of life, compared with those with lung cancer (supplemental online Table 3).

In the last few decades, recognition of oncogenic driver mutations and derived targeted agents has changed the paradigm of cancer treatment, especially the first‐line options for advanced NSCLC [28–30]. Patients are recommended to receive targeted therapy if a specific oncogenic driver gene is identified. Therefore, traditional therapeutic agents, most of which are given intravenously, are postponed after first‐line treatment fails. However, the overall survival was similar among patients with advanced NSCLC who first received TKIs or chemotherapy; consequently, there is no preferred sequence for second‐line oral targeted agents or second‐line chemotherapy [31, 32]. We found that patients with breast cancer received more oral agents, such as lapatinib and capecitabine, but those in other cancer groups receive more i.v. chemotherapy at the EOL. Continuing EOL chemotherapy for patients with breast cancer was initially a positive prognostic factor (supplemental online Fig. 2). However, subgroup analyses from patients receiving i.v. chemotherapy indicated that chemotherapy overuse at the EOL was associated with poor survival (Fig. 2). These results suggest that the balance between anticancer efficacy and toxic effects may account for the aforementioned prognostic difference. Oral agents tend to have lower toxicity and include most of the targeted therapies and hormonal therapies. The difference between intravenous chemotherapy and oral agents is thereby considerable.

Based on this information, we propose that the toxicity or adverse effects of intravenous chemotherapy overwhelm its benefit during the whole course of treatment, especially at the EOL stage. With the growing availability of oral therapeutic agents, it is imperative to understand their role at the EOL stage. Moreover, a better organized sequence of available anticancer drugs may contribute to improvement of the quality of EOL care for patients with cancer. Prospective and well‐designed studies about sequencing available anticancer drugs for patients with cancer will provide more valuable evidence.

One of the limitations of this work derives from it being a retrospective study, which may impose selection biases. We imposed restrictions on consecutive data of cancer decedents and performed our investigation at six representative hospitals across China, including cancer‐specialty hospitals and general hospitals. However, one should be cautious about applying our findings to community hospitals. Another limitation comes from unavailable data on performance status and treatment response for detailed analyses. As a result of inadequate data of treatment expense, we were unable to present a cost‐effectiveness analyses; however, to the best of our knowledge, the present study is the first and largest multicenter study demonstrating the real‐world situation of EOL chemotherapy for Chinese patients with solid cancers. We met the primary objective through analyses of a large sample of patients and detailed illustration of subsequent prognosis. We propose that the role of oral therapeutic agents at the EOL stage and in optimal whole‐process management with available drugs needs further research. In summary, we described the current status of EOL chemotherapy in China and highlighted the importance of avoiding futile treatment and improving palliative care for patients with terminal cancer.

Conclusion

In this study, we researched the current prevalence of EOL chemotherapy use in Chinese patients with solid cancers. Intravenous chemotherapy at the EOL was significantly associated with poor outcomes, and we found that the role of oral anticancer agents warrants further investigation. Efforts are also warranted to improve the quality of EOL care in China.

Author Contributions

Conception and Design: Yan Huang, Hong‐Yun Zhao, Li Zhang

Provision of study material or patients: Jin Sheng, Xiao‐Bo He, Gui‐Nan Lin, Xuan Wu, Ning Li, Jing Zhang, Lin‐Zhu Zhai, Yuan‐Yuan Zhao

Collection and/or assembly of data: Jin Sheng, Gui‐Nan Lin, Xuan Wu, Ning Li, Jing Zhang, Yuan‐Yuan Zhao, Lin‐Zhu Zhai

Data analysis and interpretation: Jin Sheng, Wen‐Feng Fang, Yun‐Peng Yang, Ning‐Ning Zhou

Manuscript writing: Jin Sheng, Ya‐Xiong Zhang

Final approval of manuscript: Jin Sheng, Ya‐Xiong Zhang, Xiaobo He, Wen‐Feng Fang, Yun‐Peng Yang, Gui‐Nan Lin, Xuan Wu, Ning Li, Jing Zhang, Lin‐Zhu Zhai, Yuan‐Yuan Zhao, Yan Huang, Ning‐Ning Zhou, Hong‐Yun Zhao, Li Zhang

Disclosure of Potential Conflicts of Interest

The authors indicated no financial relationships.

References

Author notes