Background
Pancreatic cancer (PC) generally has a very poor prognosis, and is the eighth leading cause of cancer deaths in men and the ninth in women worldwide [
1]. In addition, the number of patients with PC is increasing globally [
2]. Although surgical resection is the only potentially curative treatment, it is only possible in approximately 10–20% of newly diagnosed patients [
1].
Gemcitabine (GEM) or S-1 based chemotherapy has been the primary treatment for patients with unresectable PC in Japan, although recently new regimens have emerged that have given improved outcomes. For example, a combination chemotherapy regimen consisting of oxaliplatin, irinotecan, fluorouracil, and leucovorin (FOLFIRINOX) provided a significant survival advantage compared to GEM [
3]. Albumin-bound paclitaxel combined with GEM also improved overall survival in patients with metastatic PC [
4]. We previously conducted a retrospective chart review between 2001 and 2010 for 1248 consecutive patients with PC in 10 centers [
5]. This revealed that long-term outcomes in PC had improved, suggesting that chemotherapy should be offered to patients previously only considered for best supportive care (BSC).
Many elderly patients with malignancies often choose BSC even though they meet the eligibility criteria for chemotherapy. Under-treatment because of age, regardless of any comorbidity, has been previously reported in patients with malignancies such as lung, colorectal, and breast carcinomas, and lymphoma [
6,
7]. The reasons for this bias may include patient preferences, the tendency of physicians to treat patients according to their age, and a lack of evidence-based guidelines for treating elderly patients [
6]. Nevertheless, approximately 60% of all new cancer cases and 70% of all cancer-related deaths occur in the elderly (i.e. those aged ≥ 65 years) [
8].
The age-specific incidence of PC is also increasing amongst patients aged ≥70 years in Japan, whereas the incidence amongst younger patients remains unchanged [
9]. Despite an ageing population, there are relatively few reports on the efficacy and safety of GEM-based chemotherapy for elderly PC patients compared to younger patients [
10‐
13], and those that have been published only included a small number of patients. Here, we report the results of a study conducted by the Ehime Pancreato-Cholangiology (EPOCH) Study Group that used a large patient cohort to examine the efficacy of chemotherapy and the issues faced by elderly patients with unresectable PC.
Methods
Study design
This study was a retrospective chart review that included patients who had been diagnosed with PC at Ehime University Hospital or one of its 9 affiliated community hospitals (EPOCH Study Group). The study protocol conformed to the ethical guidelines of the 1975 Declaration of Helsinki, and it was approved by the relevant institutional ethics committee at the Ehime University Graduate School of Medicine (approval number: 1204066). Data were stored in a secure database and patients were numerically coded to anonymize data.
Patients
We collected the data of consecutive PC patients at 10 centers between January 2001 and December 2010, as previously described [
5]. Briefly, PC was diagnosed on the basis of abdominal imaging reported by board certified radiologists, together with histologic findings reported by board certified pathologists if a pathological specimen was retrieved [
5]. We included patients who were diagnosed with pancreatic ductal adenocarcinoma, and excluded patients who underwent surgical resection, were diagnosed with a pancreatic neoplasm other than PC, or for whom the relevant data was missing. The final clinical stage was determined on the basis of imaging results. Treatments were divided into 2 categories: chemotherapy (with or without radiation), and BSC. For the purposes of this study, the patients were divided into an elderly group (aged ≥65 years) and a younger group (aged <65 years), as defined by the World Health Organization [
14] and the Ministry of Health, Labour and Welfare [
15].
Data collection
Data collected included demographics (age and sex), date of diagnosis, tumor location, clinical stage (JPS TNM classification [
16]), treatment, cause of death, and outcome. Some differences in the TNM classification between the JPS and UICC staging systems with respect to tumor factors (T) and lymph node metastases (N) need to be noted. Briefly, T4 in UICC indicates tumor invasion limited to the trunk of the celiac artery or superior mesenteric artery, whereas invasion to any major vessels, neural plexus, or adjacent organs are included in the JPS classification. Positive lymph node metastasis in UICC is designated N1, while in JPS, various grades are used depending on the distance from the main tumor (N1, N2, or N3).
Analyses of the reason for choosing BSC
We identified the reason for selecting BSC from medical charts to evaluate whether the patients choosing BSC received an optimized treatment. Eligibility criteria for chemotherapy included performance status 0–2, serum aspartate aminotransferase ≤ upper limit of normal (ULN) × 2.5, serum alanine aminotransferase ≤ ULN × 2.5, serum total bilirubin ≤ ULN × 2, serum creatinine ≤ ULN, blood urea nitrogen ≤ ULN, white blood cell count ≥ 4000 /μL, hemoglobin ≥ 9.5 g/dL, platelet count ≥ 10 × 104 /μL, and no complications or dementia that would affect the decision to administer chemotherapy. We defined a non-optimized BSC group as containing patients who met the eligibility criteria for chemotherapy but who chose BSC because (a) they were not informed that they had the disease, (b) the family members preferred this option, or (c) the physician selected this treatment because of the patient’s age.
Outcome measures
The primary outcome measure was the proportion of patients choosing BSC. The secondary outcome measures included the proportion of patients treated with chemotherapy, the total median survival time (MST), the MST of patients treated with chemotherapy or BSC, and the proportion of patients who choose BSC inappropriately.
Statistical analyses
All results are expressed as the mean ± standard deviation. Data were analyzed using the chi-square test, Student t-test, or Mann-Whitney U-test, as appropriate. Outcomes were analyzed using the Kaplan-Meier method. Differences in survival were analyzed using the log-rank test. Two-tailed significance was defined in all analyses as a P-value < 0.05. All statistical analyses were performed using the JMP statistical software package (version 11; SAS Institute Japan, Tokyo, Japan).
Discussion
In our study, GEM-based chemotherapy appeared to be effective for both elderly and younger PC patients. However, we found that a significantly greater proportion of elderly patients chose BSC, which led to a poorer prognosis. In addition, the preferences of elderly patients who were treated with BSC despite being eligible for chemotherapy were not taken into account, mainly because their family had asked that they not be informed of their diagnosis.
In elderly patients, the use of chemotherapy is often limited due to the increased risk of adverse events related to comorbidities, poor performance status, or poor cognitive function [
17]. However, some elderly patients are eligible for chemotherapy, and each case needs to be considered on its own merit. It has been reported that elderly cancer patients are still under-represented in new cancer treatment clinical trials [
18‐
21], and only a few studies have provided information on the tolerability and benefits of chemotherapy in the elderly [
22‐
24]. Consequently, the safety and efficacy of new chemotherapy regimens for PC such as FOLFIRINOX or combined albumin-bound paclitaxel and GEM remains unclear in elderly patients [
3,
4]. On the other hand, the safety and efficacy of conventional chemotherapy for PC, mainly GEM-based regimens, has been proven in a number of studies with small patient cohorts, as described above [
10‐
13]. In fact, in the large cohort study we report here, GEM-based chemotherapy was equally effective in younger and elderly patients. This is consistent with the results of previous studies [
10‐
13], and suggests that GEM-based chemotherapy may continue to play a more important role than novel regimens in elderly patients until the safety and efficacy of the latter are proven.
Our data also showed that quite a large number of elderly patients chose BSC even though they met the eligibility criteria for chemotherapy, indicating that simply providing the best treatment option may improve the prognosis of these patients. A previous study showed that many patients receiving chemotherapy for incurable cancers might not understand that chemotherapy is unlikely to be curative [
25]. On the other hand, there may be a number of reasons why elderly patients tend to choose BSC over chemotherapy, which are likely to include organ dysfunction, drug toxicity, and a refusal to give informed consent. Yamagishi et al. suggested that these include patient disappointment that chemotherapy does not cure unresectable cancer, the greater susceptibility of elderly patients to severe drug toxicity that is most likely due to the age-related deterioration of organ function and/or drug clearance, and other disorders that may contraindicate chemotherapy [
10]. A particular point of concern is the high prevalence of patients who were not informed about their medical condition based on a family member’s wishes and who then chose BSC (41.1%). Although the informed consent rate at a cancer institute was as high as 75% in 1997, data from the Ministry of Health, Labor and Welfare in Japan shows that the informed consent rate among cancer patients in general practice has been gradually increasing but is still not acceptably high: 18% in 1992, 20% in 1994, 50% in 1998, 66% in 2006, and 74% in 2012. These studies included all cancer patients, but when the analysis was limited to patients who were predicted to survive less than 6 months, the informed consent rate was only 46% in 2005.
In order to better understand and improve inaccurate beliefs about chemotherapy, sufficient recognition of its limitations in elderly patients is needed among medical staff and the patients’ family. Care should be taken to avoid misleading the patient due to insufficient knowledge and a lack of evidence-based informed consent, which in turn would allow the patients’ preferences to be better reflected. Although the cut-off age of 65 years might not match the current clinical view of what “elderly” means, our findings held true even when the age cut-off between elderly and non-elderly patients was increased to 70, 75, or even 80 years.
Our study has several limitations. First, not all patients had histologically proven adenocarcinoma because fine needle aspiration was not widely used in general practice, especially in the first half of the study period and this retrospective study was conducted mainly among community hospitals where fine needle aspiration is not yet available. Nevertheless, given that the vast majority (87%) of Japanese PC patients who were assessed histologically were diagnosed with adenocarcinoma [
26], this may have had only a small impact on the treatment strategy and prognosis in the current study population. Second, the reason for choosing BSC was not identified in 54 elderly patients (17.1%) and this may overestimate the reason why they chose BSC. However, the high proportion (78 of 133, 58.6%) of cases in which BSC may have been selected inappropriately by elderly patients is the more important issue. Third, our study mainly included patients treated with GEM-based chemotherapy, rather than novel regimens, because these new treatments had not been approved within this study period in Japan. Further clinical trials are required to address adaptations of these novel regimens for elderly PC patients. We also need to assess drug toxicity, which was not possible in this study. Finally, some patients who are eligible for chemotherapy have an unfavorable prognosis and do not benefit from it. Indeed, in our study, 27 elderly patients (7.8%) who received chemotherapy survived no longer than 90 days. Furthermore, eligibility criteria for chemotherapy other than age, performance status, liver function, and renal function may exist. It is therefore important to identify predictive factors for whether chemotherapy will benefit elderly patients.
Acknowledgements
Not applicable.