A Systematic Review of the Burden of Pancreatic Cancer in Europe: Real-World Impact on Survival, Quality of Life and Costs
- 1.6k Downloads
The purpose of this study was to assess the overall burden of pancreatic cancer in Europe, with a focus on survival time in a real-world setting, and the overall healthy life lost to the disease.
Real-world data were retrieved from peer-reviewed, observational studies identified by an electronic search. We performed two de novo analyses: a proportional shortfall analysis to quantify the proportion of healthy life lost to pancreatic cancer and an estimation of the aggregate life-years lost annually in Europe.
Ninety-one studies were included. The median, age-standardised incidence of pancreatic cancer per 100,000 was 7.6 in men and 4.9 in women. Overall median survival from diagnosis was 4.6 months; median survival was 2.8–5.7 months in patients with metastatic disease. The proportional shortfall analysis showed that pancreatic cancer results in a 98 % loss of healthy life, with a life expectancy at diagnosis of 4.6 months compared to 15.1 years for an age-matched healthy population. Annually, 610,000–915,000 quality-adjusted life-years (QALYs) are lost to pancreatic cancer in Europe. Patients had significantly lower scores on validated health-related quality of life instruments versus population norms.
To the best of our knowledge, this is the first study to systematically review real-world overall survival and patient outcomes of pancreatic cancer patients in Europe outside the context of clinical trials. Our findings confirm the poor prognosis and short survival reported by national studies. Pancreatic cancer is a substantial burden in Europe, with nearly a million aggregate life-years lost annually and almost complete loss of healthy life in affected individuals.
KeywordsPancreatic cancer Disease burden Epidemiology Survival Quality of life Proportional shortfall
Pancreatic cancer is among the most deadly malignancies; despite being responsible for only 3 % of all new cancer diagnoses, it was the fourth most common cause of cancer death in the United States in 2013  and is expected to be the second most common cause by 2030 . The reported incidence is highest in developed countries, which likely reflect more accurate diagnosis, with Europe bearing a significant part of the burden . In 2008, Europe carried one quarter of the global burden, despite comprising only one ninth of the world population . Accurate data on cancer incidence and mortality in Europe are therefore crucial in assessing the burden of disease, the effectiveness of control programmes and for budgeting and planning at national and regional levels.
Current treatment options for pancreatic cancer are limited, with surgical resection presently the only potentially curative treatment option [5, 6]. However, owing to the frequently advanced stage at diagnosis, 80–90 % of patients have unresectable tumours , and long-term survival after surgical resection is poor . New treatments have recently been approved and are expected: chemotherapy has demonstrated a favourable impact in overall survival when prescribed after surgery with curative intent ; gemcitabine and erlotinib are approved in the USA and Europe for use in metastatic disease; and nab-paclitaxel has recently been approved in Europe and the USA to treat metastatic disease. In order to improve long-term outcomes for pancreatic cancer patients, it is imperative that patients have access to the range of new treatment options. Key stakeholders in this process are healthcare payers. In order to support the introduction of these treatments, real-world survival times provide a meaningful context in which to evaluate improvements in patient outcomes. Survival times are an important way of capturing the unmet medical needs of a patient as they allow the calculation of shortfall. They also allow us to capture the unmet medical needs at a societal level through the number of aggregate life-years lost to the disease .
To our knowledge, no published studies have systematically reviewed the real-world survival time in Europe outside context of clinical trials. The objective of the systematic review was therefore to perform a systematic, macroscopic analysis of the burden of pancreatic cancer in Europe on individuals and health systems, to evaluate the direct and indirect costs of the disease, and to assess the extent of the unmet needs.
The research comprised two phases: a systematic review of published data and related de novo analyses.
Systematic Review Methods
Study inclusion and exclusion criteria
Study types eligible for inclusion
Study types that were excluded
• Observational/uncontrolled cohort studies (prospective/retrospective longitudinal studies, cross-sectional studies)
• Studies with both observational and interventional phases
• Database studies, registries
• Studies with no available abstract
• Interventional trials (parallel and crossover design, double-blind, single-blind, open label) with no observational phases
• Phase I, II, III or IV studies
• Case reports or case series
• Case–control studies
• In vitro studies/genetic studies/molecular epidemiology studies
• Imaging/diagnostic studies; studies assessing techniques
• Classification schemes (e.g. studies assessing staging schemes)
• Studies focusing on pathophysiology or prevention
• Systematic reviews
• Editorials, letters, comments, non-systematic review articles
• Policy/prioritisation papers/research recommendations
Two independent reviewers performed the review and full-text extraction; any differences were resolved by a third reviewer. No quantitative syntheses, e.g. meta-analysis, were performed. Overall medians of median values reported in individual studies were calculated as summary measures.
De Novo Analysis: Proportional Shortfall
Quality-adjusted life-year (QALY) expectation or ‘health life’ in the absence of pancreatic cancer was calculated by multiplying the mean life expectancy of patients aged 71 in EU 28 countries (15.1 years) , by the mean health-related quality of life (HRQoL) of this age group (0.78 = weighted health state index for the UK population aged 65–74) . Pancreatic cancer-related QALYs were calculated by multiplying the median patient survival in months, derived from the current systematic review, by the EQ-5D utility obtained from a German study of 45 pancreatic cancer patients (mean EQ-5D utility 0.65; mean age 64 years) .
De Novo Analysis: Aggregate Life-Years Lost
Aggregate life-years lost to pancreatic cancer was calculated from the difference in survival between pancreatic cancer patients and the general population multiplied by the annual number of incident pancreatic cancer cases across Europe obtained from three different sources: GLOBOCAN incident cases derived from crude incidence estimates (79,331 cases) ; crude incidence rates from the current systematic review multiplied by the EU 28 population of 507.2 million  (62,203 cases); and age-adjusted (European standard) rates derived from the current systematic review multiplied by the EU 28 population (52,872 cases). QALYs lost per person (calculated as per the PS analysis) were multiplied by the incident cases of pancreatic cancer in Europe to derive the aggregate QALYs lost. Analyses were performed for all patients with pancreatic cancer and metastatic patients (based on an estimated 62.5 % patients having Stage IV disease as a proportion of all patients in Stages I to IV ).
Search and Selection of Studies
Incidence of Pancreatic Cancer
Median annual incidence rates of pancreatic cancer in Europe by sex: age-standardised and crude rates
Annual incidence rate per 100,000: median (range)
Age-standardised: World (33 studies)
Age-standardised: Europe (4 studies)
Crude: (5 studies)
Survival Rates at 1 and 5 Years
Combined survival data for both sexes were available from 32 studies or sub-studies reported in 17 publications [45, 46, 47, 48, 49, 44, 34, 50, 37, 38, 33, 51, 52, 53, 54, 29, 43] (Figure S1 in the online resource). Combined 1-year survival rates ranged from 10 to 23 %. The largest study in a population of 30,025 Dutch patients  reported a 1-year survival rate of 16 %. The overall median 1-year survival across studies was 15 %.
Combined 5-year survival rates ranged from 0.5 % in Sweden  to 9 % in Slovenian  and German populations . The two largest studies reported 5-year survival rates of 3 and 5 % [49, 48]. Overall median 5-year survival across studies was 3 %.
Survival Rates at 1 and 5 Years by Sex
Survival at 1 and 5 years from diagnosis
Survival (%): median (range)
15 (9 [Denmark]–29 [Malta])
14 (9 [Malta]–22 [Sweden and Finland])
4 (1.3 [Slovenia]–7.5 [Czech Rep.)
4 (0 [Switzerland]–7 [Estonia])
Survival by Disease Stage
The effect of disease stage (irrespective of intervention) on median survival from diagnosis was reported by three studies [41, 43, 39]. As expected, survival rates decreased with advancing stage. In a Dutch population between 1985 and 2001, survival rates at 1 year were 40, 30, 35, 25 and 5 % for stages I, II, III, IVa and IVb, respectively . In a Finnish population between 1947 and 1980, 1-year survival rates were 53, 19, 17 and 12 % for stages I, II, III and IV, respectively . In the aforementioned Dutch population, survival at 5 years was 15, 1, 5 and 1 % among patients with stages I, II, III and IV disease, respectively .
Median Survival by Intervention
Five studies reported survival from diagnosis by intervention; four were in Scandinavian populations [44, 37, 38, 33] and one in an Italian population . Patients who underwent resection or radical surgery consistently showed the longest median survival, ranging from 11 to 25.7 months. In patients who underwent chemotherapy, radiotherapy, palliative surgery or exploratory laparotomy, median survival ranged from 2 to 8.1 months. Patients who received no surgery or best supportive care had a median survival of 1.1 month (Figure S2 in the online resource).
The shorter survival among patients treated with chemotherapy, radiotherapy, palliative surgery or exploratory laparotomy reflects the more advanced stage of disease in these patients (54 % of patients treated with chemotherapy/radiotherapy and 52 % with best supportive care had stage IV disease, versus 3 % of patients receiving resective/radical surgery).
Annual mortality rate per 100,000 in Europe as whole and in European regions
Age-standardised, world (124 studies/sub-studies)
Annual mortality rate per 100,000: median (range)
Age-standardised mortality rates based on the ESP in women and men ranged from 7.4 to 9.7 per 100,000, respectively, in England/Wales to 11.8 and 15.3, respectively, in the Netherlands [68, 30, 31]. Crude mortality rates per 100,000 ranged from 11.2  to 16.7  in women and from 12.5  to 16.5 [63, 64] in men.
Data were not evenly distributed over time, with the majority coming from the period 2000 to 2006. Women had lower mortality rates than men at all time periods where data were collected.
Three studies evaluated depression and anxiety in patients with advanced disease [70, 40, 12]. In a Swedish population, 42 % of patients reported moderate or severe anxiety and depression. In a German study, the number of patients experiencing anxiety/depression was around 10-fold higher than population norms .
Three studies reported direct costs [39, 73, 74] and two indirect costs [73, 74] associated with pancreatic cancer. In Sweden, the mean monthly total cost of care from 2005 to 2007 was just over €6500 per month , corresponding to €16,066 over the residual lifetime. In Germany, total costs were considerably higher at €31,375 over the residual lifetime or €48,900 per year . Data on the relative contribution of indirect versus direct costs to the overall cost were not consistent between countries. In Sweden, indirect costs accounted for a greater proportion of overall costs than direct costs (€83,109 versus 16,066, respectively) , whereas the opposite was reported in Germany (€28,164 versus 3210) .
Five studies reported direct cost data [75, 39, 73, 34, 74]. Hospitalisation accounted for the major component of direct costs per residual lifetime (€7981–16,264), followed by interventions (radiology, surgery and chemotherapy; €1575–9761) and chemotherapy alone (€1423–3569).
Two studies reported indirect costs [73, 74]. In Germany, 24 % of diagnosed patients were actively employed, resulting in a mean productivity loss of €2972 . In Sweden in 2009, indirect costs were slightly higher for male versus female patients aged ≤64 years, with mean short-term productivity loss of €87,205 for men and €49,895 for women; mean productivity loss per patient due to mortality was €238,843 in men and €220,543 in women .
De Novo Analysis of Proportional Shortfall
Proportional shortfall due to pancreatic cancer
Pancreatic cancer patients
Life expectancy at age 71
0.38 years (4.6 months)
Utility (ages 65–74)
Proportional shortfall (C)/(A)
De Novo Analysis of Aggregate Life-Years Lost
Aggregate life-years and QALYs lost to pancreatic cancer across Europe (EU 28)
Aggregate life-years lost
Aggregate QALYs lost
All pancreatic cancer
Systematic review (crude)
Systematic review (ESP)
Metastatic pancreatic cancer
Systematic review (crude)
Systematic review (ESP)
The clinical studies identified represented all regions of Europe, although the North of Europe, in particular Scandinavian countries, represented almost half studies. Survival times in pancreatic cancer are very short; this is partly explained by the lack of effective treatments and the difficulties in diagnosing the disease early. Data from large cohort studies and national/regional registries showed 1- and 5-year survival rates of about 15 and 4 %, respectively. Our estimated median survival from diagnosis of 4.6 months is in line with a median of 3 months reported in a US study of 32,452 adults with distant metastatic pancreatic cancer  and 10 weeks in a recent Dutch study of 3099 patients with metastatic disease  and reflects the high number of patients with metastatic disease at diagnosis. The longest median survival from diagnosis was reported in patients who underwent resection or radical surgery. It is likely, however, that these data are skewed by disease stage, as patients who underwent resection are more likely to have had earlier stage disease.
We found that age-standardisation methodology had a notable influence on the incidence rate, with median rates ranging from approximately 8–16 per 100,000 annually in men, and 5–15 per 100,000 in women, depending on the method used. Incidence rate standardised to the WSP gave the lowest combined incidence (6.29/100,000), which was similar to those reported by GLOBOCAN (7/100,000) , which includes estimates from all EU countries. The combined crude incidence rate estimated from five studies was higher at 12.27/100,000 annually. Crude incidence rates give a picture of the actual rate in a population. However, incidence (and other rates) of cancer are strongly age-dependent; therefore, comparisons of crude rates between populations may be misleading if the age composition of the populations differ. Although age-standardisation facilitates comparisons, standardised rates can be deceiving if the age structure of the reference population is different to that of the real population. Owing to the substantially lower rates observed with age-adjustment to the WSP, we used only crude estimates or figures adjusted to ESP in the de novo analyses.
Mortality rates were higher in men than in women in all the studies identified; given the similar survival rates between the sexes, this is most likely driven by the higher incidence in men. The majority of studies reported mortality rates as age-standardised based on the WSP; however, similar to the incidence data, this resulted in substantially lower rates than ESP age-adjusted and crude estimates, making it questionable whether this standardisation method is appropriate for the European population. Mortality rates across European regions and individual countries were comparable to the Europe-wide figures. However, one study that used data from the WHO database reported appreciably higher mortality rates in eastern European accession countries compared with the European Union .
Expectedly, HRQoL scores were significantly lower and levels of depression and anxiety higher in patients with pancreatic cancer compared with population norms, indicating that there is scope to improve supportive care of patients until screening or early diagnosis is improved, or treatments that provide clear survival advantages become available.
A limitation of the current study is the lack of quantitative analysis, such as meta-analysis, of the epidemiological data. Nonetheless, we were able to quantify the aggregate number of life-years and QALYs lost due to pancreatic cancer in Europe each year. Various methods of incorporating equity concerns into standard QALY-based economic evaluations have been implemented, such as considering different incremental cost-effectiveness ratio (ICER) thresholds . The PS approach that we have used includes an equity-weighting component that demonstrates the nearly complete loss of healthy life associated with pancreatic cancer. Recent developments such as value-based pricing in the UK and policies in other countries reflect an increasing willingness of decision-makers to account for such weightings in QALY-related assessments [81, 82]. The estimates of aggregate life-years and QALYs lost to pancreatic cancer demonstrate the magnitude of the disease’s impact in Europe and raise the question of how health systems are managing this burden.
We found few studies that assessed the cost of pancreatic cancer; in those identified, methodologies varied between studies. Hospitalisation accounted for the major component of direct costs, followed by interventions and chemotherapy, with some variation between countries. This was consistent with data from a large US study showing that hospitalisations and cancer-directed procedures accounted for the largest fraction of health care costs .
The findings of this systematic review show that the real-world median survival time of patients with pancreatic cancer in Europe was less than 5 months and less than 10 % of patients survived beyond 5 years. Survival at 1 year was highest in patients with early stage disease who undergo resection, highlighting the need for improved early diagnosis. Our results confirm those of individual pan-European registry studies and US studies that report similar incidence and mortality rates for pancreatic cancer, reflecting the poor prognosis. Our analyses demonstrate the considerable burden of pancreatic cancer in Europe, with almost a million aggregate life-years lost and an almost complete loss of healthy life in affected individuals. Quantifying the number of life-years and QALYs lost provides a method to compare the relative burden of different cancers in the context of value-based pricing and to evaluate treatment options in view of the projected increase in disease burden  over the next 15 years.
This work was funded by Celgene Corporation.
Conflict of Interest
AP and KD are employed by Celgene and own shares in the company. KAA, AH, CP and IP are employees of Dolon Ltd., which received payment from Celgene for services undertaken in support of this work. AC and JV have acted as a paid consultant to Celgene. AF has acted as a paid consultant to Celgene and also Merck Serono, Amgen, Bayer, Sanofi, Roche and Lilly. MD has acted as a paid consultant to Celgene, Novartis, Sanofi, Merck and Roche.
- 2.Rahib L, Smith BD, Aizenberg R, Rosenzweig AB, Fleshman JM, Matrisian LM. Projecting cancer incidence and deaths to 2030: the unexpected burden of thyroid, liver, and pancreas cancers in the United States. Cancer Res. 2014;74(11):2913–21. doi: 10.1158/0008-5472.can-14-0155.PubMedCrossRefGoogle Scholar
- 7.Neoptolemos JP, Stocken DD, Friess H, Bassi C, Dunn JA, Hickey H, Beger H, Fernandez-Cruz L, Dervenis C, Lacaine F, Falconi M, Pederzoli P, Pap A, Spooner D, Kerr DJ, Buchler MW, European Study Group for Pancreatic C. A randomized trial of chemoradiotherapy and chemotherapy after resection of pancreatic cancer. N Engl J Med. 2004;350(12):1200–10. doi: 10.1056/NEJMoa032295.CrossRefGoogle Scholar
- 8.Lichtenberg FR. Pharmaceutical innovation, mortality reduction, and economic growth. In: Murphy KM, Topel RH, editors. Measuring the gains from medical research: an economic approach. Chicago: University of Chicago Press; 2010. p. 74–84.Google Scholar
- 10.Eurostat Table: Life expectancy by age and sex (2013). http://epp.eurostat.ec.europa.eu/portal/page/portal/statistics/search_database. Accessed 26 March 2014.
- 11.Kind P, Hardman G, Macran S (1999) UK Population Norms for EQ-5D. University of York Centre for Health Economics - Discussion paper. http://www.york.ac.uk/media/che/documents/papers/discussionpapers/CHE Discussion Paper 172.pdf. Accessed 26 March 2014.
- 13.GLOBOCAN 2008 v2.0, Cancer Incidence and Mortality Worldwide: IARC CancerBase No. 10. http://globocan.iarc.fr. Accessed 06 December 2013.
- 14.Eurostat Table: Population on 1 January by age and sex. http://epp.eurostat.ec.europa.eu/portal/page/portal/statistics/search_database. Accessed 26 March 2014.
- 50.Desauw C, El Hajbi F, Ligier K, Duhamel A, Richard F, Rose C. Pathological proof and survival for patients with billiary-tract or pancreatic tumor. Vienna: ESMO annual meeting. European Society for Medical Oncology (ESMO); 2012.Google Scholar
- 76.GLOBOCAN 2012 v1.0, Cancer Incidence and Mortality Worldwide: IARC CancerBase No. 11. http://globocan.iarc.f. Acessed 06 December 2013.
- 77.Worni M, Guller U, White RR, Castleberry AW, Pietrobon R, Cerny T, et al. Modest Improvement in Overall Survival for Patients With Metastatic Pancreatic Cancer: a Trend Analysis Using the Surveillance, Epidemiology, and End Results Registry From 1988 to 2008. Pancreas. 2013. doi: 10.1097/MPA.0b013e318291fbc5.PubMedGoogle Scholar
- 78.Bernards N, Haj Mohammad N, Creemers GJ, de Hingh IH, van Laarhoven HW, Lemmens VE. Ten weeks to live: a population-based study on treatment and survival of patients with metastatic pancreatic cancer in the south of the Netherlands. Acta Oncol. 2015;54(3):403–10. doi: 10.3109/0284186x.2014.953257.PubMedCrossRefGoogle Scholar
- 81.NICE ‘central’ to value-based pricing of medicines (2013). http://www.nice.org.uk/news/article/nice-central-to-valuebased-pricing-of-medicines. Accessed 26 March 2015.
- 82.Analysing the Potential Impact of UK “Value-Based Pricing” (2013). http://www.ohe.org/case-studies/analysing-potential-impact-uk-value-based-pricing. Accessed 26 March 2015.
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.