Cohort study of the overall survival of patients with pancreatic cancer in a hospital of specialties of Quito-Ecuador in the period 2007–2017

Introduction

Pancreatic neoplasms are associated with one of the higher mortalities around the world [1, 2]. It is the 14th most prevalent neoplasm and the seventh cause of death for neoplasm worldwide [3]. According to GLOBOCAN 2018, the age-standardized incidence rate (ASR) of pancreatic cancer is 5.5 per 100,000 men, and 4.0 per 100,000 women. ASR mortality is 5.1 per 100,000 men and 3.8 per 100,000 women [4]. In Ecuador, there were 373 deaths due to pancreatic neoplasms in 2017, according to the register of Instituto Nacional de Estadísticas y Censos (INEC) [5].

The risk factors for developing pancreatic neoplasms include genetics, tobacco and alcohol consumption, obesity, and diabetes [6]. The time it takes to develop a pancreatic neoplasm is considered to start between 10 and 17 years before the final stages of the disease. Early identification is associated with a better prognosis [7, 8].

The symptoms depend on the localization of the tumor and the stage of the disease. The most common symptoms include abdominal pain, jaundice, nausea, vomit, and steatorrhea [8]. The mean medial survival rate at 5 years is 9% in all stages of the disease according to the American Society of Cancer between 2008 and 2014 [9].

Today, there is no recommendation for routine screenings of pancreatic neoplasms because of its lower incidence in the general population [3]. The stage of the disease at diagnosis is the most important factor associated with survival rates in patients [1]. Less than 20% of the patients can receive surgical treatment considering the diagnosis is usually made in the late stages of the disease [3, 9, 10]. Other types of treatments include chemotherapy, radiotherapy, and others [11].

Methods

Results

For the period of time of the study, 357 clinical records of patients with pancreatic cancer were found. The mean age in pancreatic cancer patients was 62.7 years old (SD 14.3), with the predominance of females (55.74%). The most frequent personal history was smoking and alcohol consumption (12 and 51.8%, respectively), and 60% of patients had a history of diabetes. The main sign of admission was abdominal pain (50.98%). Sixty-eight percent% of patients had a positive antigen Ca 19-9. Adenocarcinoma was the most frequent type of cancer (56.86%), the most frequent location was the head of the pancreas (59.10%). More than two-thirds (69.19%) of the patients were diagnosed in the late stages of the disease, we account for stages III and IV that represented 15.13 and 54.06%, respectively, as of late stages for pancreatic neoplasm. Of these 38.3% (n=137) patients received surgical treatment, in this group 27.17% had surgery alone and 11.2% had surgery plus another treatment (Table 1).

Kaplan–Meier survival curves were created and then followed by a logrank test to compare survival functions. These tests showed a statistically significant difference in survival between groups of treatment (p<0.001). Survival time from the moment of diagnosis was highest among the group that received surgical treatment at 1, 2, 5 years, and at the end of follow-up (Table 2). The lowest survival proportion for this group was of 0.51 at 3 years of follow up, compared to zero when receiving no treatment or palliative care, 0.16 when receiving chemotherapy, and 0.29 when receiving combined treatment.

The overall survival rate based on treatment modality is shown in Figure 1. The median survival time from the moment of diagnosis for no treatment was 2 months, for chemotherapy 5 months, for surgical treatment 17 months, for combined therapy 7 months, and for palliative care 2 months.

Figure 1:

Kaplan–Meier estimates by type of cancer.

The overall survival rates based on the type of cancer are shown in Figure 2. There is no statistically significant difference in survival rates if the patient was diagnosed with exocrine cancer compared to endocrine cancer (p=0.605).

Figure 2:

Kaplan–Meier estimates by type of treatment.

The stage at which the cancer was detected clearly marks a difference in survival rate. The median survival time was 14 months for stage II, 5 months for stage III, and 2 months for stage IV.

Figure 3:

Kaplan–Meier estimates by stage of diagnosis.

An unadjusted Cox Regression model that analyzed the relationship between survival time and age, gender, stage at diagnosis, type of cancer, and type of treatment was done to determine the variables that had a p value of 0.25 or under. A saturated model which included stage of diagnosis and type of treatment was performed with variables that met the inclusion criteria. We also included the type of cancer due to its clinical relevance. By backward variable reduction, a final model which included stage of diagnosis and type of treatment was chosen. The researchers furtherly performed a sensitivity analysis by eliminating all patients diagnosed at stage 0, which significantly changed our results, leading us to exclude stage 0 patients. It was found that when adjusting for stage of cancer, there is a 0.49 Hazard Ratio when patients are treated with surgery compared to no treatment (p=0.013) and of 0.34 when patients are treated with both surgery and chemotherapy (p<0.001) as described in Table 3.

Discussion

New diagnosis techniques and treatments have been discovered for pancreatic neoplasms in the last decade. Despite all these advantages, the survival rates have not improved [3, 12].

In our study, 69.9% of the patients were diagnosed in the late stages of the disease. This could be compared with McGuigan’s review of patients, who noted that this was a result of the unspecific symptoms and the rapid invasion of blood vessels [3].

The mean survival rate of patients at 5 years was 2–9% in a study performed in Shanghai [12], comparable to a 14% survival rate in our study [13].

The study results show a survival time of 14 months from the moment of diagnosis for patients at stage III compared with only two months in patients at stage IV [14], this data could be compared to the results of Cancer Research UK in which the mean survival time for patients at stage III was 6–11 months and 2–6 months for patients at stage IV [13]. Other studies have found similar results [15–17].

Surgery was associated with the best survival time in patients with pancreatic neoplasms. The mean survival time for patients who underwent surgery was 17 months compared with 5 months for those patients who received chemotherapy. Similar results have been described by Nakeeb [18], Kardosh [16], Nakagawa [19], Brooks [20], and Cancer Research UK [13].

A limitation of the study was the exclusion of some patients because their clinical records were incomplete. There were fewer registered patients with pancreatic neoplasms in the years previous to 2010 and some of the patients did not continue with follow-up treatment due to a lack of subsequent appointments at the hospital.

Conclusion

The statistically significant difference of survival time from the moment of diagnosis between types of treatment is the most relevant finding in this study. When compared to all other types of treatments, surgery yields the best results. An early diagnosis increases the possibility of receiving surgery and having a better survival time in patients.