Management and survival analysis of elderly patients with a cancer in the digestive system who refused to receive anticancer treatments

The cancer types of the 57 cases included esophageal squamous cell carcinoma, gastric adenocarcinoma, duodenal adenocarcinoma, periampullary carcinoma, pancreatic ductal adenocarcinoma, primary hepatocellular carcinoma, intrahepatic/perihilar/distal cholangiocarcinoma, colon cancer, and rectal cancer. Of these 57 cases, 49 were at an advanced stage (12 at stage II, 12 at stage III, and 25 at stage IV), while the remaining eight (including five pancreatic ductal adenocarcinomas, two hepatocellular carcinomas, and one duodenal adenocarcinoma) cases were at an early stage. The median overall survival time was 11 months, but advanced-stage patients survived for a shorter period with their median overall survival being 7.28 months. The survival time of the early-stage subgroup was longer than that of the advanced-stage subgroup, as the median overall survival of five patients with an early-stage pancreatic cancer was 27 months, while the remaining nine patients with an advanced-stage pancreatic cancer survived for only 5 months. Within the advanced-stage subgroup, duodenal adenocarcinoma, primary hepatocellular carcinoma, and colon cancer patients tended to have a relatively longer survival, while pancreatic cancer and intrahepatic cholangiocarcinoma patients survived for the shortest time. Because we only had one case each of esophageal squamous cell carcinoma, periampullary carcinoma, and rectal cancer, these cases were excluded from the comparisons. Among the advanced-stage patients, two cases of pancreatic cancer and one case of hilar bile duct carcinoma showed the shortest survival time (2 months). The longest survival time was 36 months in a patient with an advanced cancer in the distal common bile duct. In the early-stage group, a patient with a hepatocellular carcinoma (stage Ia) survived for 61 months, which was the longest in our cohort, followed by a case of an early duodenal adenocarcinoma who lived for 53 months. The longest survival time of early pancreatic cancer in our cohort was 35 months.

During the course of their diseases, almost all patients experienced a decline in function of one or more major organs, such as heart, lung, liver, kidney, hematopoietic system, and gastrointestinal tract as well as total body homeostasis. As shown in Fig. 1, imbalance of homeostasis, including acid-and-base imbalance and electrolytic disturbance, was the most common morbidity and occurred in 56 patients. The only exception was the patient with esophageal carcinoma, who died suddenly of massive hemorrhage from the tumor. It was our anticipation that this patient might have shown imbalanced homeostasis as well, had he survived longer. The percentage of the patients with impairments of the lung, liver, or kidneys was 89.47, 77.19, or 66.67%, respectively, suggesting that these organs might be more easily affected. Common clinical symptoms included incontrollable lung infection and ensuing respiratory failure, as well as renal or hepatic injury and dysfunction caused by tumor invasion, infection, or the drugs administered. All patients in our cohort manifested impairment or dysfunction of two or more organ systems, with the average number of malfunctioning organs per patient being 3.68.

All patients received the best available supportive care and palliative care to alleviate the pain and improve QOL over their terminal stage of life, with the comprehensive therapies employed summarized in Table 1. Antibiotics were administered to all patients, as they all experienced infection in the lung, bile duct, and/or urinary tract. Twenty-one (36.8%) patients experienced cancer-caused pain and were given analgesics, and 77.2% of the patients were administered liver protective drugs due to impaired hepatic function. Bile duct stenting through endoscopic retrograde cholangiopancreatography (ERCP), or percutaneous transhepatic cholangial drainage (PTCD), was performed on 15 pancreatic, bile duct, or duodenal cancer patients with obstructive jaundice. The median overall survival from the jaundice-relieving operation to death was 160 days. Tracheal intubation was done in 27 patients, and the median overall survival from this procedure to death was a mere 12 days. Enteral and/or parenteral nutrition was applied to all 57 patients, of whom 77.2% had a nasogastric tube or duodenal catheterization while 96.5% received central venous catheterization or had a peripherally inserted central catheter (PICC). Plasma albumin, pre-albumin, and hemoglobin were determined 3 to 5 days before death, and their levels were 27.64 ± 3.99 g/L, 13.08 ± 3.72 mg/dL, and 83.7 ± 23.1 g/L, respectively.

Although impairment of two or more organs occurred in all patients enrolled, only 54.4% of the patients eventually died of multiple-organ functional failure (MOSF). After active supportive and palliative cares, some patients died of single organ dysfunction, including lung failure after refractory lung infection, bile duct infection, liver failure, heart failure, and tumor bleeding causing hemorrhagic shock with ensuing heart failure (Fig. 2). Functional failure of an organ is determined, as elsewhere, based on not only the patient’s clinical symptoms but also data from laboratories and relevant instruments. Direct cause of the death is defined as the primary one, such as lung infection or tumor hemorrhage, which triggers severe functional impairment of the affected organ and probably also other organ(s), eventually leading to heart failure and ensuing death of the patient.

There were 51 (89.5%) cases of lung infection, 44 (77.2%) cases of liver functional injury, and 38 (66.7%) cases of kidney impairment in our cohort. There were 17.6% (nine cases), 15.9% (nine cases), and 18.4% (seven cases) of the patients whose lung infection, liver injury, and kidney impairment, respectively, were well-controlled (Fig. 3). None of our patients died of acid/base imbalance or electrolyte disturbance.

Aging is typically associated with a higher risk of comorbidity. Elders often manifest a decline in physical and cognitive functions, and their social supports are often reduced. It is more common, compared with younger patients, that tumors in elders are already at an advanced stage or even already have distant metastases at the time of diagnosis, albeit tumors in elders usually grow and progress more slowly than in younger patients. Because of these factors, elderly cancer patients usually show a poorer prognosis and a shorter survival time, compared with younger adult patients. It has been reported that elderly cancer patients are less likely to be treated with surgery, chemotherapy, or radiation [8]. A recent population-based study from Italy revealed that only 58 of 1183 (5.8%) elderly patients at ages of ≥ 80 years received chemotherapy, in contrast to an average of 34.3% of their junior counterparts [9]. Due to the toxicity of chemotherapy or because of other concerns such as infection or other comorbidities, a higher percentage of elderly patients are hospitalized than their young counterparts. Chemotherapy recipients have a substantially higher hospitalization rate for infection or fever, hematologic complications, dehydration, and pulmonary embolism (PE) or deep vein thrombosis (DVT), compared with those who have not received chemotherapy [10]. Modern chemotherapy and targeted therapy have improved the overall outcomes of patients for all ages. However, the results observed in real clinical practice are often different from those reported in clinical trials, especially in elderly patients, according to our own experiences. Albeit most cancer patients are at a senior age, there are few specific treatment-based guidelines for elderly cancer patients. In our opinion, this may be due to the limited number of such patients recruited in clinical trials. Of course, there are some data to support that the general health situation of some elderly patients is good enough to tolerate modified therapies. In fact, individualized treatment for elderly cancer patients requires concerns other than their age. When treating elderly cancer patients, oncologists are advised to make a comprehensive assessment, using such tools as a geriatric assessment or predictive chemotherapy toxicity tools, as the basis for making an optimal therapy regimen. The International Society for Geriatric Oncology and the NCCN guidelines both recommend performing a geriatric assessment in all elderly cancer patients [11, 12]. Factors such as functional status of major organs, social support, patient’s preference, presence of comorbidities, and life expectancy should be taken into consideration when formulating an optimal treatment regimen. Therefore, for elderly cancer patients, it is important to weigh the risk of dying from cancer against the risk of dying from a possible comorbidity or from a treatment-caused complication.

A large percentage of patients with an advanced cancer receive a long course of aggressive treatments, including chemotherapy and/or radiotherapy, until the moribund period of the patient’s life, despite the fact that this may actually reflect a poor quality of care. A survey was recently conducted on the family members of elderly lung or colorectal cancer patients who eventually died. The results show that an earlier hospice enrollment, avoidance of ICU admission within 30 days of death, and death at a non-hospital location are associated with a perception of a better end-of-life care [13]. In 2012, the American Society of Clinical Oncology (ASCO) published a provisional clinical opinion (PCO) advising its members that “… combined standard oncology care and palliative care should be considered early in the course of illness for any patient with metastatic cancer and/or a high symptom burden” [14]. This recommendation is based on several randomized clinical trials of palliative care interventions during conventional anticancer treatments of patients with a metastatic cancer. To date, it has become possible to drive some malignancies into a manageable, chronic, situation via current treatments and managements, making it possible for some incurable patients to live with the cancer in relative peace and comfort. And, perhaps most importantly, many patients and families have “dying with dignity” as their main goal. Realizing this, fulfilling this goal should also be important to the oncologist. The patients in our cohort were provided with the best available supportive and palliative care regimens instead of purely medical anticancer therapies, and thus, their “from early-to-terminal-stage” course of cancer progression is relatively closer to the natural one compared with the one shown in those patients receiving route anticancer treatments. The median overall survival of our group is similar to that of routine anticancer therapy groups reported in the literature [15‐20]. The newest SEER data (from 1988 to 2012) indicates that the rates of 1-year survival in liver/intrahepatic bile duct cancer and pancreatic cancer patients over 75 years of age are 25.6 and 15.8%, respectively [1]. In our cohort, four out of five patients with an advanced liver cancer survived 12 months or longer, while two patients with an advanced pancreatic cancer survived 12 months after diagnosis. Therefore, our data, although just from a relatively small number of patients, imply that active supportive and palliative therapies alone can provide a relatively good QOL and survival times comparable to a traditional, aggressively treated, group of senior patients with a lethal and advanced malignancy.

In our opinion, the word “manage” may be more proper than “treat” to describe how we should approach therapy in our daily oncological practice. Nutritional support, maintenance of internal homeostasis, management of various complications (pain, infection, jaundice), protection of organ functions, and even psychological intervention are fundamental elements of a comprehensive and systematic implementation for cancer patients. All patients in our cohort had supplemental enteral and/or parenteral nutrition support, resulting in a relatively high level of serum pre-albumin. Proper levels of albumin and hemoglobin are important for the maintenance of whole-body physiological function and are significant factors in the patients’ survival [21]. Up to two-thirds of all elderly patients develop pain as a result of the cancer or as a consequence of its treatment [22], but in this study, only 36.8% of the patients accepted pain-relieving drugs when apparently needed. This phenomenon may be partly because some tumors may not cause as much pain as we think, or merely because some seniors are less likely to complain of pain [23]. Obstructive jaundice in patients with pancreatic cancer, cholangiocarcinoma, or hepatocarcinoma may predict an unfavorable survival, and drainage of jaundice will help to improve liver function [24, 25]. The obstructive jaundice patients in our cohort obviously benefited from positive jaundice-reducing procedures such as PTCD or bile duct stenting, since the median overall survival is significantly prolonged after these treatments. Less than half of our patients accepted tracheal intubation, with the median overall survival after the mechanical ventilation to death being merely 12 days, suggesting that such invasive manipulation was not beneficial to the patients and thus, in our opinion, should not often be a primary choice.

When asked “how does a cancer kill the patient?”, most oncologists can only give such examples as “liver cancer patients may die of tumor hemorrhage” and “some lung cancer patients may die of infection”, but few, if any, oncologists can give good statistics about how each direct cause, such as infection, heart failure, or renal failure, may account for the percentage of deaths for any given cancer type. This is in part because different patients with the same type of cancer may die from different causes. Textbooks of medicine generally describe that patients with end-stage cancer die of MOSF, which is true in a broad sense, since the body cannot survive when one or more important organs have lost function. Patients who have or have not received anticancer treatments may die from different reasons, because the treatments themselves likely alter, and usually damage, the functions of major organs, including the immune system. Our report is one of the few, if not the only, studies of this kind to provide a percentage of common causes of death for elderly patients with a cancer in the digestive system who have not received any direct anticancer treatments but who have received the best supportive and palliative cares. Oncology peers can make their own evaluations on the value of such cares on the patients’ survival time with our data as a reference.

Organ failure could be regarded as the direct cause of the patient’s death. The following are the situations often encountered during our clinical practice: severe infection inducing septic shock, lung infection leading to respiratory failure, bile duct obstruction causing liver failure, hypercoagulation prompting a myocardial infarction, tumor rupture causing massive hemorrhage and ensuing hemorrhagic shock, complex hematologic complications, kidney failure due to various reasons. Many patients could have an even more complex situation, because they have more than two organs involved and eventually develop MOSF. However, only 54.39% of the patients in our cohort died of MOSF, with the rest dying from single organ dysfunction or from a single complication such as massive hemorrhage. In our cohort, 15.9% of the patients with liver impairment were well controlled, as were 18.4% of the patients with kidney impairment.

It is worth noting that cancer patients at terminal stages are usually bedridden, which easily causes infection in the lung or the urinary system, as shown in our cohort in which nearly 89.5% of the patients manifested lung infection with functional impairment. Uncontrolled infection will certainly accelerate the patient’s death, and therefore antibiotics, in most cases routed via intravenous infusion, become inevitable for most patients. Fortunately, about 17.6% of the patients with lung infection were finally well controlled. Therefore, foreseeing possible dysfunction of an organ and actively preventing its occurrence are fundamental in the management of elders with an advanced cancer. These management goals can improve patients’ QOL and prolong their survival time. The complexity of available treatments poses a challenge to oncologists in discussing the choice of cancer treatment with their patients, since chemotherapy, radiotherapy, or even targeted therapy are not the only important factors that influence the patients’ survival and QOL. Routine chemotherapy and radiotherapy are likely to be associated with toxicity and are thus associated with a significantly increased risk of organ impairment [10]. Some of these weaknesses could be avoided by prophylactic interventions. In our humble opinion, (1) foreseeing and (2) diminishing a possible organ dysfunction should be two key elements of cancer management.