Oesophageal cancer survival in Europe: A EUROCARE-4 study

Cancer Epidemiology 36 (2012) 505–512

Contents lists available at SciVerse ScienceDirect

Cancer Epidemiology The International Journal of Cancer Epidemiology, Detection, and Prevention journal homepage: www.cancerepidemiology.net

Oesophageal cancer survival in Europe: A EUROCARE-4 study A.T. Gavin a, S. Francisci b, R. Foschi c, D.W. Donnelly a, V. Lemmens d, H. Brenner e, L.A. Anderson f,* the EUROCARE-4 Working Group1 a

Northern Ireland Cancer Registry, Queen’s University, Belfast, Northern Ireland, United Kingdom Centro Nazionale di Epidemiologia, Istituto Superiore di Sanita, Viale Regina Elena 299, Rome, Italy Evaluative Epidemiology, Istituto Nazionale Tumori, Milan, Italy d Comprehensive Cancer Centre South, Eindhoven, The Netherlands e German Cancer Research Center, Division of Clinical Epidemiology and Aging Research, Heidelberg, Germany f Centre for Public Health, Queen’s University, Belfast, Northern Ireland, United Kingdom b c

A R T I C L E I N F O

A B S T R A C T

Article history: Received 12 December 2011 Received in revised form 18 July 2012 Accepted 22 July 2012 Available online 19 August 2012

Oesophageal cancer survival is poor with variation across Europe. No pan-European studies of survival differences by oesophageal cancer subtype exist. This study investigates rates and trends in oesophageal cancer survival across Europe. Data for primary malignant oesophageal cancer diagnosed in 1995–1999 and followed up to the end of 2003 was obtained from 66 cancer registries in 24 European countries. Relative survival was calculated using the Hakulinen approach. Staging data were available from 19 registries. Survival by region, gender, age, morphology and stage was investigated. Cohort analysis and the period approach were applied to investigate survival trends from 1988 to 2002 for 31 registries in 17 countries. In total 51,499 cases of oesophageal cancer diagnosed 1995–1999 were analysed. Overall, European 1- and 5-year survival rates were 33.4% (95% CI 32.9–33.9%) and 9.8% (95% CI 9.4–10.1%), respectively. Males, older patients and patients with late stage disease had poorer 1- and 5-year relative survival. Patients with squamous cell carcinoma had poorer 1-year relative survival. Regional variation in survival was observed with Central Europe above and Eastern Europe below the European pool. Survival for distant stage disease was similar across Europe while survival rates for localised disease were below the European pool in Eastern and Southern Europe. Improvement in European 1-year relative survival was reported (p = 0.016). Oesophageal cancer survival was poor across Europe. Persistent regional variations in 1-year survival point to a need for a high resolution study of diagnostic and treatment practices of oesophageal cancer. ß 2012 Elsevier Ltd. All rights reserved.

Keywords: Cancer Survival Europe Oesophagus Subtype Stage Trends

1. Introduction In Europe, the annual incidence of oesophageal cancer was 5.39 per 100,000 males and 1.13 per 100,000 females from 2000 to 2004 [1]. Oesophageal cancer is categorised into two main subtypes; adenocarcinoma and squamous cell carcinoma. The incidence of oesophageal adenocarcinoma, thought to be caused by exposure to gastro-oesophageal reflux [2], has increased dramatically over recent decades in developed countries [3]. Many countries, particularly those in Southern and Western Europe, have seen a decline in the incidence of oesophageal squamous cell carcinoma in men [3]; a cancer primarily caused by exposure to alcohol and tobacco smoke [4]. In a recent publication using data from the

* Corresponding author. Tel.: +44 28 90632315; fax: +44 28 90231907. E-mail address: [email protected] (L.A. Anderson). 1 See Appendix A. 1877-7821/$ – see front matter ß 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.canep.2012.07.009

World Health Organisation (WHO) the European Union average mortality rate for males with oesophageal cancer was 5.4 deaths per 100,000; England and Wales, Scotland and Hungary had the highest male mortality rates in 2000–2004 (8.0, 10.9 and 8.4 deaths per 100,000, respectively), while Greece and Bulgaria had the lowest (1.5 and 2.0 deaths per 100,000, respectively) [1]. Oesophageal cancer survival is poor also displaying regional variation across Europe [1,5]. In EUROCARE-2, the 5-year relative survival rate for oesophageal cancer for cases diagnosed 1978– 1989 was 10%; ranging from 3% in Estonia to 11% in Switzerland [5]. Survival is reportedly lower in those with an older age at diagnosis, advanced tumour stage and those not undergoing surgical resection [6]. Body mass, tobacco smoking, educational level [7] and recently alcohol [8] have also been suggested as possible influencing factors on survival. Few studies have investigated survival differences by oesophageal cancer subtype [6,9,10] and none have compared differences across Europe. We compared survival rates and trends in oesophageal cancer survival

A.T. Gavin et al. / Cancer Epidemiology 36 (2012) 505–512

506

across Europe using data from EUROCARE-4 and investigated differences in survival by gender, age, morphology and stage. 2. Methods Details of EUROCARE-4 methodology have been published [11]. Briefly, cancer survival data for patients diagnosed in 1995–1999, with follow-up to the end of 2003, was collected from 66 cancer registries in 23 European countries (Table 1). Additionally patients diagnosed with oesophageal cancer from 1988 to 2002 (from 31 registries in 17 countries) were included in the trend analysis. Cases of oesophageal cancer were identified according to the International Classification of Diseases for Oncology (ICD)-9, ICD10 or ICD-O codes (excluding haematological malignancies) depending on diagnosis period and country of origin [13]. After exclusions for major errors including missing, invalid or inconsistent data (n = 24) the analyses were limited to adults (aged 15–99 years) with primary oesophageal cancer. Those with secondary oesophageal cancers and those identified by death certificate only or by autopsy were excluded (n = 1286). Consistent with previous EUROCARE-4 reports due to small numbers in some registries countries were grouped into five regions for the majority of our analyses: Central Europe (Austria, Belgium, France, Germany, the Netherlands and Switzerland), Eastern Europe (Czech Republic, Slovakia and Poland), Northern Europe (Denmark, Finland, Iceland, Norway and Sweden), Southern Europe (Italy, Malta, Portugal, Slovenia and Spain) and United Kingdom & Ireland (England, Wales, Scotland, Ireland and Northern Ireland) [11]. A pooled estimate for the overall European survival was created using data from all 66 cancer registries. Patients were categorised into the following age categories: 15–44 years, 45–54 years, 55–64 years, 65–74 years and 75–99 years. Histological subtypes included squamous cell carcinoma (ICD-0-3: M8050–8078, 8083–8084), adenocarcinoma (ICD-0-3: M8140–8141, 8143–8145, 8190–8231, 8260–8263, 8310, 8401, 8480–8490, 8550–8551, 8570–8574 and 8576), and other which

included unknown morphology. Nineteen registries collected sufficient (over 45% of cases) information on cancer stage. Stage at diagnosis was classified as local (confined to primary site), regional (spread to regional lymph nodes), distant (metastatic) and no stage. Survival was investigated by gender, age, morphology and stage. If there were insufficient numbers of cases (<10) in any of the stratified categories the country was excluded from the analysis. 2.1. Trend analysis Data on 64,755 oesophageal cancer cases diagnosed from 1988 to 2002 with active follow-up were used to investigate trends in survival. Cases <15 years and those diagnosed by death certificate or autopsy only were excluded. Thirty-one registries, from 17 countries, with survival data from 1988 were included in the analysis. 2.2. Statistical analysis Relative survival was calculated using the Hakulinen approach [14] for generating expected survival, with age-standardisation using the International Cancer Survival Standard [15]. Countries with small numbers, i.e. Malta and Iceland were not reported individually but included in the pooled analyses. Cohort analysis was applied for all survival analysis except for persons diagnosed in 2000–2002, where 5-year follow-up was incomplete, for which the period approach was utilised [16]. Relative excess risks of death compared to Europe were derived by taking the ratio of the natural logarithm of the relative survival rate for each region to that of the European pool of 66 registries. Survival trend significance was evaluated as the pvalue (two-sided, <0.05) of the slope coefficient deriving from a linear regression on the relative survival trend in Europe (the pool of 31 European registries). The test for trend was limited to the period 1988–1999 as 2000–2002 was not directly comparable. The regression was applied separately at 1-year and 5-years survival,

Table 1 Characteristics by country: patients diagnosed 1995–1999. Country

Number of oesophageal cancer cases

% death certificate only cases

Number of oesophageal cancer cases included in survival analysis

Gender % males

Age mean

Ratio squamous cell carcinoma: adenocarcinoma

% with staging information available

% microscopically verified

% lost to follow-up

Austria Belgium Czech Republic Denmark England Finland France Germany Iceland Ireland Italy Malta Netherlands Northern Ireland Norway Poland Portugal Scotland Slovakia Slovenia Spain Switzerland Sweden Wales

1426 1067 132 1560 24,406 913 1786 348 65 1461 3234 56 1752 727 705 632 389 3587 1287 399 1767 460 1521 1819

18.1% 0.0% 3.9% 2.6% 0.0% 2.5% 0.0% 4.5% 0.0% 2.5% 2.5% 0.0% 0.0% 1.5% 0.6% 4.5% 0.0% 1.5% 21.0% 4.2% 3.0% 2.0% 0.0% 15.8%

1207 1067 127 1520 24,406 891 1786 333 65 1426 3153 56 1751 716 701 605 389 3535 1052 383 1716 450 1521 1571

82% 78% 87% 72% 61% 59% 85% 79% 75% 61% 77% 71% 68% 61% 73% 77% 84% 59% 93% 81% 84% 76% 72% 58%

64 66 60 68 71 69 65 63 73 70 68 69 67 70 69 65 64 70 59 64 64 66 70 71

2.4:1 1.3:1 3.1:1 –b 0.7:1 2.5:1 5.3:1 3.6:1 1.7:1 1.2:1 3.7:1 1.0:1 0.8:1 0.6:1 1.6:1 4.9:1 4.6:1 0.9:1 8.4:1 7.3:1 4.3:1 1.8:1 1.8:1 0.8:1

76% 0% 50% 0% 0% 75% 0% 66% 0% 61% 0% 0% 44% 0% 100% 74% 49% 0% 76% 86% 37% 64% 0% 0%

78% 91% 85% 92% 98% 93% 89% 94% 100% 87% 86% 77% 98% 87% 96% 73% 93% 91% 76% 86% 91% 94% 99% 45%

0.0% 0.0% 0.0% 0.1% 0.0% 0.0% 2.0% 0.0% 0.0% 0.0% 0.3% 0.0% 0.1% 0.0% 0.0% 0.3% 0.0% 0.0% 0.0% 0.3% 0.2% 0.9% 0.1% 0.0%

Europea

51,499

2.0%

50,427

67%

70

1.2:1

14%

92%

0.1%

a b

European pooled estimate. Morphology data not provided to EUROCARE-4.

Table 2 1- and 5-year age-standardised relative survival and relative excess risks by sex and countryc compared to the European pool: patients diagnosed 1995–1999. Country

Male

Female

1-Year

5-Year

1-Year

5-Year

Relative excess risk (95% CI)

Relative survival (95% CI)

Relative excess risk (95% CI)

Relative survival (95% CI)

Relative excess risk (95% CI)

Relative survival (95% CI)

Relative excess risk (95% CI)

Austria Belgium Czech Republicb Denmark Finland France Germany Ireland Italy Netherlands Norway Poland Portugal Slovakia Slovenia Spain Sweden Switzerland England Northern Irelandb Scotland Wales

36.0% 43.0% 25.7% 27.4% 31.2% 43.2% 41.3% 30.9% 37.6% 35.6% 29.1% 24.5% 35.3% 23.2% 19.8% 35.6% 35.2% 42.1% 31.9% 36.7% 33.1% 33.5%

39.5%) 46.7%) 34.7%) 30.5%) 35.9%) 46.0%) 49.4%) 34.3%) 39.7%) 38.4%) 33.7%) 29.4%) 41.7%) 27.4%) 25.0%) 38.8%) 38.4%) 47.9%) 32.8%) 41.4%) 35.3%) 36.9%)

0.92 0.76 1.23 1.17 1.05 0.76 0.80 1.06 0.88 0.93 1.12 1.27 0.94 1.32 1.47 0.93 0.94 0.78 1.03 0.92 1.00 0.99

10.2% 17.2% 3.4% 5.4% 8.3% 11.5% 17.1% 11.0% 10.0% 11.6% 8.3% 5.5% 13.3% 6.8% 3.9% 9.8% 11.8% 11.2% 8.1% 10.4% 9.9% 13.0%

0.96 0.74 1.42 1.22 1.04 0.91 0.74 0.93 0.96 0.87 1.04 1.22 0.85 1.13 1.36 0.97 0.90 0.92 0.99 0.95 0.97 0.86

30.1% 45.0% 15.3% 28.8% 43.6% 42.0% 34.8% 37.0% 42.1% 40.3% 33.7% 27.0% 30.5% 31.2% 22.5% 37.5% 38.8% 52.0% 34.4% – 35.1% 34.5%

(32.0%, 38.2%) (30.2%, 39.6%)

1.16 0.77 1.81 1.20 0.80 0.84 1.02 0.96 0.83 0.88 1.05 1.26 1.14 1.12 1.44 0.94 0.91 0.63 1.03 – 1.01 1.03

12.6% 20.9% – 4.3% 19.2% 15.4% 23.1% 16.8% 16.2% 11.0% 7.0% 9.4% 13.7% 8.4% 11.6% 10.5% 18.2% 17.5% 10.1% – 11.7% 15.5%

(9.6%, 14.2%) (11.7%, 19.8%)

0.97 0.73 – 1.48 0.77 0.87 0.68 0.83 0.85 1.06 1.24 1.02 0.93 1.16 1.01 1.05 0.80 0.81 1.01 – 1.00 0.87

Europea

35.4% (34.6%, 36.2%)

1.00

37.7% (36.2%, 39.4%)

1.00

13.4% (12.2%, 14.6%)

1.00

a b c

(32.9%, (39.7%, (19.5%, (24.8%, (27.3%, (40.6%, (34.8%, (27.9%, (35.6%, (33.2%, (25.3%, (20.6%, (30.1%, (19.8%, (15.9%, (32.8%, (32.4%, (37.1%, (31.1%, (32.0%, (31.0%, (30.5%,

(0.84, (0.69, (0.97, (1.07, (0.93, (0.70, (0.64, (0.97, (0.83, (0.86, (0.98, (1.11, (0.79, (1.17, (1.26, (0.85, (0.86, (0.67, (1.00, (0.80, (0.94, (0.90,

1.01) 0.84) 1.49) 1.27) 1.18) 0.82) 0.96) 1.16) 0.94) 1.00) 1.25) 1.43) 1.09) 1.47) 1.67) 1.01) 1.02) 0.90) 1.06) 1.03) 1.06) 1.08)

(8.0%, 12.8%) (14.6%, 20.0%) (1.4%, 7.1%) (4.0%, 7.0%) (5.7%, 11.5%) (9.8%, 13.5%) (11.2%, 23.9%) (8.6%, 13.7%) (8.7%, 11.5%) (9.6%, 13.7%) (5.8%, 11.4%) (3.7%, 7.8%) (8.8%, 18.7%) (4.0%, 10.6%) (2.3%, 6.2%) (7.7%, 12.1%) (9.6%, 14.3%) (7.8%, 15.3%) (7.6%, 8.6%) (7.5%, 13.9%) (8.4%, 11.5%) (10.6%, 15.7%)

10.2% (9.7%, 10.9%)

1.00

(0.85, (0.67, (1.07, (1.11, (0.89, (0.84, (0.58, (0.83, (0.90, (0.83, (0.90, (1.06, (0.69, (0.92, (1.15, (0.88, (0.81, (0.78, (1.02, (0.82, (0.90, (0.77,

1.06) 0.80) 1.76) 1.34) 1.19) 0.97) 0.90) 1.02) 1.02) 0.98) 1.19) 1.37) 1.00) 1.33) 1.57) 1.07) 0.98) 1.06) 1.09) 1.08) 1.04) 0.94)

(24.3%, 37.8%) (38.8%, 52.5%) (6.9%, 47.0%) (24.4%, 34.3%) (37.8%, 50.5%) (36.5%, 48.7%) (25.6%, 48.8%) (32.7%, 42.1%) (38.2%, 46.6%) (36.6%, 44.2%) (26.7%, 43.3%) (20.1%, 37.3%) (19.3%, 42.4%) (22.2%, 45.4%) (15.4%, 34.7%) (30.4%, 47.0%) (33.6%, 44.9%) (43.8%, 62.3%) (33.2%, 35.7%)

(0.94, (0.62, (0.90, (1.03, (0.66, (0.70, (0.71, (0.83, (0.74, (0.78, (0.81, (0.96, (0.77, (0.78, (1.05, (0.73, (0.77, (0.47, (0.98,

1.37) 0.92) 2.71) 1.37) 0.94) 0.98) 1.33) 1.08) 0.93) 0.97) 1.28) 1.56) 1.52) 1.47) 1.83) 1.16) 1.06) 0.85) 1.07)

(0.92, 1.10) (0.89, 1.16)

(8.4%, 17.7%) (15.6%, 26.8%) (2.5%, 6.7%) (14.1%, 25.0%) (10.9%, 20.6%) (13.1%, 34.9%) (13.1%, 20.9%) (12.9%, 19.8%) (8.5%, 13.8%) (3.5%, 12.1%) (4.7%, 16.1%) (6.2%, 24.1%) (3.2%, 17.1%) (5.3%, 20.6%) (5.8%, 16.8%) (13.7%, 23.3%) (10.8%, 25.5%) (9.3%, 11.0%)

(0.79, 1.14) (0.60, 0.86) (1.24, (0.64, (0.73, (0.46, (0.72, (0.75, (0.92, (0.95, (0.82, (0.62, (0.77, (0.69, (0.80, (0.67, (0.62, (1.02,

1.71) 0.90) 1.03) 0.91) 0.95) 0.95) 1.15) 1.54) 1.40) 1.24) 1.54) 1.32) 1.30) 0.92) 1.30) 1.12)

(0.91, 1.10) (0.75, 1.00)

A.T. Gavin et al. / Cancer Epidemiology 36 (2012) 505–512

Relative survival (95% CI)

European average was obtained after directly weighting the region-specific survival estimates with the mean population of each region over 1995–1999. There were insufficient numbers (<10) of females in Northern Ireland. Malta and Iceland were excluded due to the small number of cases but were included in the European average.

507

5.0%)

8.2%)

5.7%)

12.0%)

9.0%)

30.5%)

25.6%)

23.2%)

22.0%)

7.2% (5.6%, 6.7% (3.2%, 4.3% (3.1%, 6.3% (4.7%, 4.5% (4.1%, 31.6%)

29.2% (26.8%, 17.7% (12.9%, 23.3% (21.2%, 27.9% (25.2%, 21.3% (20.5%, 33.4%)

35.2%)

28.8%)

38.5%)

14.2% (12.6%, 15.9%) 7.1% (4.7%, 10.0%) 10.0% (8.4%, 11.8%) 10.7% (9.0%, 12.4%) 9.4% (8.7%, 10.0%) 42.9%)

40.8% (38.7%, 24.7% (20.8%, 32.8% (30.4%, 36.1% (33.6%, 32.4% (31.5%, 44.1%)

38.4%)

29.7%)

39.3%)

14.8% (13.2%, 16.6%) 5.2% (3.4%, 7.5%) 11.8% (9.8%, 14.0%) 11.8% (10.0%, 13.7%) 11.9% (11.0%, 12.8%) 48.2%)

45.9% (43.6%, 25.7% (21.9%, 35.3% (32.3%, 41.5% (38.8%, 38.0% (36.7%, 44.2%)

50.8%)

31.3%)

43.8%)

16.0% (13.8%, 18.3%) 6.4% (4.3%, 9.1%) 10.0% (7.6%, 12.9%) 13.0% (10.5%, 15.7%) 13.4% (12.1%, 14.8%) 52.8%)

49.9% (46.8%, 27.0% (23.0%, 39.7 (35.5%, 47.1% (43.3%, 42.3% (40.3%, 50.8%)

51.9%)

37.6%)

50.9%)

19.3% (14.9%, 24.1%) 8.8% (4.7%, 14.4%) 17.7% (10.9%, 26.0%) 16.5% (11.2%, 22.7%) 14.1% (11.4%, 16.9%) 49.8%)

UK & Ireland

Southern

1-Year relative survival (95% CI) 5-Year relative survival (95% CI) 1-Year relative survival (95% CI) 1-Year relative survival (95% CI)

44.2% (38.4%, 29.9% (22.5%, 42.4% (32.4%, 43.3% (35.6%, 47.1% (43.1%, Northern

European 1-year relative survival rates, based on the pool of 19 cancer registries, were 54.8%, 32.8% and 18.2% for patients with

Eastern

3.4. Survival by stage

Age class 45–54 years

European 1-year relative survival rates were significantly higher in patients with oesophageal adenocarcinoma compared to survival in patients with oesophageal squamous cell carcinoma, Table 4. However, this was not as evident for 5-year relative survival. One-year and 5-year relative survival was highest in Central Europe and lowest in Eastern Europe for all subtypes investigated, Table 4. Five-year relative survival rates were higher in Central Europe for oesophageal adenocarcinomas compared to oesophageal squamous cell carcinomas.

Age class 15–44 years

3.3. Survival rates by morphology

European region

European 1- and 5-year relative survival rates decreased with increasing age. There was a stepwise reduction in 1-year survival rates with increasing age, but the effect was less pronounced with 5-year survival. One-year relative survival rates were highest in Central Europe in those aged 45–99 years and lowest in Eastern Europe for all age groups, Table 3. Five-year survival was highest in Central Europe across all age groups and lowest in Eastern Europe except in those aged 75–99 years where Northern Europe had the lowest survival rates, Table 3.

Table 3 1- and 5-year relative survival by age and region: patients diagnosed 1995–1999.

3.2. Survival rates by age

Central

Age class 55–64 years

Overall, females had lower 1-year (RER female vs male 0.94, 95% CI 0.92–0.97) and 5-year (RER female vs male 0.89, 95% CI 0.86– 0.91) relative excess risks of death than males, Table 2. One-year relative survival was highest in France, Belgium and Switzerland in males and Switzerland, Belgium and Finland in Females. One-year relative survival was above the European pool for both males and females in Belgium, Italy and Switzerland, Table 2. One-year relative survival was below the European pool for both males and females in Denmark, Slovakia and Slovenia, Table 2. Five-year relative survival rates were highest in Belgium, Germany and Portugal in males and Germany, Belgium and Finland in females, Table 2. Five-year relative survival rates were lowest in the Czech Republic in males and Denmark in females. In Denmark the 5-year relative survival rates were below the European pool for both males and females in Denmark, Table 2.

1-Year relative survival (95% CI)

3.1. Survival rates by gender

5-Year relative survival (95% CI)

Age class 65–74 years

In total 51,499 cases of oesophageal cancer, diagnosed from 1995 until end 1999, were included. Slovakia had the highest death certificate only rates (21%) and the highest percentage males (93%), Table 1. Several countries, including England, Iceland, France, Sweden, Belgium, Malta, the Netherlands and Portugal, had no death certificate only oesophageal cancer cases, Table 1. Denmark had no information on histological type. England, Netherlands, Scotland, Northern Ireland and Wales had a lower proportion of patients with squamous cell carcinomas compared to patients with adenocarcinoma, Table 1. The overall European pooled relative 1-year and 5-year survival rates for oesophageal cancer were 33.4% (95% CI: 32.9–33.9%) and 9.8% (95% CI: 9.4–10.1%), respectively.

1-Year relative survival (95% CI)

Age class 75–99 years

3. Results

5-Year relative survival (95% CI)

respectively. The SEER-Stat software [17] was used for all analysis. For further details see the detailed description of the EUROCARE-4 database and statistical methods [12].

5-Year relative survival (95% CI)

A.T. Gavin et al. / Cancer Epidemiology 36 (2012) 505–512

5-Year relative survival (95% CI)

508

A.T. Gavin et al. / Cancer Epidemiology 36 (2012) 505–512 Table 4 1- and 5-year relative survival by morphology and region: patients diagnosed 1995–1999.

European Region

1-Year relative survival (95% CI)

5-Year relative survival (95% CI)

Local

Central Eastern Northern Southern UK & Ireland Europea

64.8% 42.4% 55.1% 52.7% 63.7% 54.8%

(58.9%, (35.5%, (49.6%, (47.1%, (54.9%, (52.2%,

70.1%) 49.1%) 60.2%) 57.9%) 71.3%) 57.3%)

36.9% 13.6% 25.1% 16.8% 37.5% 24.5%

Regional

(14.3%, 17.8%) (2.2%, 8.0%) (10.2%, 15.2%) (11.1%, 16.5%) (9.0%, 10.1%) (10.1%, 11.1%)

Central Eastern Northern Southern UK & Ireland Europea

42.1% 24.4% 44.5% 33.1% 30.6% 32.8%

(39.0%, (20.7%, (37.7%, (28.1%, (28.6%, (31.5%,

45.1%) 28.4%) 51.0%) 38.2%) 32.6%) 34.1%)

9.6% 7.1% 8.8% 8.1% 8.5% 8.4%

(7.7%, (4.5%, (5.3%, (5.2%, (7.2%, (7.5%,

11.8%) 10.6%) 13.4%) 11.8%) 9.9%) 9.3%)

Distant

(7.0%, (5.9%, (4.4%, (5.3%, (5.8%, (6.4%,

Central Eastern Northern Southern UK & Ireland Europea

20.1% 14.4% 15.8% 13.0% 20.5% 18.2%

(17.2%, 23.2%) (11.0%, 18.3%) (13.1%, 18.8%) (9.1%, 17.4%) (17.5%, 23.6%) (16.8%, 19.7%)

2.6% 2.1% 2.9% 3.5% 4.9% 3.8%

(1.6%, (1.2%, (1.7%, (1.0%, (3.3%, (3.1%,

4.0%) 3.5%) 4.5%) 8.8%) 6.9%) 4.8%)

Unstaged

Central Eastern Northern Southern UK & Ireland Europea

34.9% 20.1% 34.7% 25.6% 36.2% 28.9%

(30.6%, (16.4%, (28.5%, (20.7%, (30.9%, (26.9%,

European region

1-Year relative survival (95% CI)

5-Year relative survival (95% CI)

Squamous cell carcinoma

Central

39.8% (38.2%, 41.4%)

11.9% (10.7%, 13.1%)

Eastern Northern Southern UK & Ireland Europea

23.9% 34.0% 38.4% 32.8% 33.9%

(20.9%, (31.7%, (36.6%, (31.8%, (33.3%,

27.0%) 36.3%) 40.1%) 33.8%) 34.6%)

5.9% 11.2% 10.0% 10.4% 10.1%

(4.0%, (9.6%, (8.9%, (9.6%, (9.7%,

Central Eastern Northern Southern UK & Ireland Europea

43.8% 30.4% 39.5% 41.1% 36.8% 37.9%

(41.7%, (23.7%, (36.2%, (37.6%, (36.0%, (37.1%,

45.9%) 37.3%) 42.8%) 44.6%) 37.7%) 38.6%)

16.0% 4.4% 12.6% 13.6% 9.5% 10.6%

Central Eastern Northern Southern UK & Ireland Europea

28.7% 18.7% 26.2% 26.8% 20.0% 23.0%

(25.4%, (14.7%, (24.1%, (23.5%, (18.7%, (22.0%,

32.1%) 23.1%) 28.5%) 30.3%) 21.4%) 24.0%)

9.2% 9.2% 5.6% 7.1% 6.6% 7.0%

Other

a

Table 5 1- and 5-year relative survival by stage and region: patients diagnosed 1995–1999. Stage

Subtype

Adenocarcinoma

509

8.2%) 13.0%) 11.3%) 11.1%) 10.6%)

11.8%) 13.4%) 6.9%) 9.2%) 7.6%) 7.7%)

European pooled estimate.

local, regional and distant stage disease, respectively. Eastern Europe had the lowest 1-year relative survival rates for local, regional and distant stage cancers and for those with no staging information provided to EUROCARE-4, Table 5. European 5-year relative survival rates were 24.5%, 8.4% and 3.8% for patients with local, regional and distant stage disease, respectively. Five-year relative survival was highest in the UK for local, distant and unstaged cancers and lowest in Eastern Europe for all stages, Table 5. Patients with distant stage disease had a higher 1-year and 5year excess risk of death than those with local stage disease (RER 3.25, 95% CI 3.06–3.44 and RER 2.76, 95% CI 2.57–2.96, respectively). 3.5. Trends Overall, European 1-year relative survival rates, based on the pool of 31 registries, improved for patients diagnosed between 1988–1990 and 1997–1999 (p = 0.016), Table 6. One-year relative survival increased from 27.8% in 1988–1990 to an estimated 37.9% in 2000–2002, Table 6. Five-year relative survival increased from

39.3%) 24.1%) 40.9%) 30.6%) 41.4%) 31.0%)

13.1% 4.9% 10.0% 11.0% 18.2% 10.2%

(30.7%, 43.1%) (8.8%, 19.5%) (20.1%, 30.5%) (12.6%, 21.5%) (28.6%, 46.4%) (22.1%, 26.9%)

(9.9%, 16.7%) (2.7%, 8.1%) (5.2%, 16.5%) (7.4%, 15.4%) (13.9%, 23.1%) (8.8%, 11.8%)

‘Unstaged’ refers to those patients with stage blank, unknown or invalid stage information’. a European pooled estimate.

7.9% in 1988–90 to an estimated 11.2% in 2000–2002, Table 6, but the trend for patients diagnosed between 1988–1990 and 1997– 1999 was not significant (p = 0.065). 4. Discussion Our analysis of EUROCARE-4 data showed that European 1-year and Central European 5-year relative survival rates were better for patients with oesophageal adenocarcinomas compared to those with oesophageal squamous cell carcinomas. One- and 5-year relative survival rates were also higher in females, younger patients and those with local stage disease. There was evidence of regional variations in survival among oesophageal cancer patients throughout Europe. Overall, 1- and 5-year relative survival rates for oesophageal cancer were highest in Central Europe. Eastern Europe had significantly lower 1- and 5-year survival rates

Table 6 1- and 5-year age-standardised relative survival by period of diagnosis and European region. Interval and European region

Relative survival (95% CI) 1991–1993

1988–1990 1-Year Central Eastern Northern Southern UK & Ireland Europea

34.5% 17.2% 29.7% 28.6% 27.2% 27.8%

5-Year Central Eastern Northern Southern UK & Ireland Europea

8.9% 6.2% 8.1% 6.1% 8.8% 7.9%

a b

European pooled estimate. Estimated using period analysis.

(31.9%, (14.3%, (27.1%, (25.8%, (25.9%, (26.9%,

(7.3%, (4.0%, (6.4%, (4.6%, (7.9%, (7.4%,

37.1%) 20.4%) 32.2%) 31.5%) 28.5%) 28.7%)

10.7%) 9.1%) 10.1%) 7.9%) 9.7%) 8.5%)

33.2% 17.4% 31.4% 31.7% 28.8% 28.7%

8.9% 5.3% 8.2% 7.7% 8.4% 7.9%

1994–1996

(30.7%, (14.7%, (28.9%, (29.1%, (27.5%, (27.9%,

(7.3%, (3.3%, (6.7%, (6.1%, (7.5%, (7.3%,

35.6%) 20.3%) 33.9%) 34.5%) 30.0%) 29.6%)

10.5%) 8.1%) 9.9%) 9.6%) 9.3%) 8.4%)

2000–2002b

1997–1999

37.2% 21.4% 34.4% 33.9% 30.4% 31.4%

(34.8%, (18.6%, (32.0%, (31.0%, (29.2%, (30.5%,

39.6%) 24.5%) 36.8%) 36.8%) 31.6%) 32.2%)

38.2% 25.9% 34.6% 37.4% 33.3% 33.7%

(35.9%, (22.7%, (32.3%, (34.4%, (32.2%, (32.9%,

40.5%) 29.2%) 36.9%) 40.4%) 34.5%) 34.5%)

43.9% 22.7% 38.3% 37.9% 38.0% 37.9%

(42.1%, (19.9%, (36.0%, (35.6%, (37.3%, (37.3%,

45.7%) 25.7%) 40.7%) 40.2%) 38.8%) 38.5%)

12.3% 6.3% 10.8% 11.6% 9.0% 9.6%

(10.6%, 14.1%) (4.2%, 8.9%) (9.1%, 12.6%) (9.5%, 13.9%) (8.2%, 9.8%) (9.0%, 10.2%)

12.0% 7.5% 11.3% 9.4% 10.3% 10.3%

(10.3%, 13.7%) (5.2%, 10.2%) (9.6%, 13.1%) (7.5%, 11.6%) (9.5%, 11.1%) (9.7%, 10.9%)

15.1% 7.3% 11.2% 10.2% 11.2% 11.2%

(13.8%, 16.5%) (5.1%, 9.8%) (9.6%, 12.9%) (8.9%, 11.6%) (10.6%, 11.8%) (10.8%, 11.7%)

510

A.T. Gavin et al. / Cancer Epidemiology 36 (2012) 505–512

compared to the European pool irrespective of subtype. However, only a small proportion of Eastern European countries were represented in the current analysis and our findings may therefore be unrepresentative of Eastern Europe as a whole. Changes in environmental and lifestyle exposures, diagnostic methods, classification, disease management and cancer registration practices may account for some of the regional variation in survival, however, the RER of death in European regions did not alter compared to the European average over time. Despite trend analyses showing improvements in European 1year relative survival rates over time, overall 1- and 5-year relative survival rates in 1995–1999 in EUROCARE-4 (33.4% and 9.8%, respectively) were similar to those reported in the 1978–1989, EUROCARE-2, study (33% and 10%, respectively) [5]. The discrepancy between these findings may be attributed to the differing countries represented in EUROCARE-2 (17 countries) and EUROCARE-4 (24 countries) and differing analytical methods utilised. One-year relative survival has shown a more marked improvement compared to 5-year relative survival. Improvements in 1-year survival may be attributed to earlier detection, the ongoing centralisation of oesophageal cancer surgery in high-volume centres leading to decreases in hospital mortality rates, the introduction of chemo and/or radiotherapy for advanced oesophageal cancers [10,18–20] and/or other interventions such as endoscopic palliative treatments [21]. Increased proportions of patients undergoing curative resection improvements, more adequate patient selection for resection, earlier detection, staging and pre-/peri-operative chemoradiotherapy, staging and patient selection may contribute to improved longer-term survival [22]. As highlighted for other cancer sites [23] the pooled European 5-year relative survival rate for oesophageal cancer (9.8%) is lower than that reported by the Surveillance Epidemiology and End Results (SEER) for the USA (14.0%; 1998–2001) [24]. Survival rates were lower in the UK and Ireland, Northern, Southern and Eastern Europe compared to the USA. However, 5-year relative survival rates in Central Europe were estimated at 15.1% in 2000–2002. Microscopic verification of cases in EUROCARE 4 (approx. 90%) [12] was less than in SEER (97%) [24] but is unlikely to explain the survival advantage in the USA. Surveillance for oesophageal adenocarcinoma, population awareness, clinical referral patterns, access to interventions and treatments and differing health care systems may explain some of the regional variation in survival within Europe and between Europe and the USA. The survival rates reported in SEER [24] and Central Europe point to achievable targets and health gain opportunities. Women had better 5-year relative survival than men as reported for multiple cancer sites in EUROCARE-4 [25]. Overall 1- and 5-year survival rates for oesophageal cancer have also been reported to be higher in women compared to men in England and Wales [26] and elsewhere [27,28]. Differences in histological subtypes, risk factor exposure, e.g. tobacco use, alcohol use, comorbidity [29] and/or stage at presentation could all contribute to the observed survival advantage in females. Unsurprisingly, increasing age was associated with a marked decrease in overall 1- and 5-year survival, despite taking account of background mortality. Surgical resection during the time period investigated was the only treatment to cure oesophageal cancer. Compared to younger patients, elderly patients have higher operative, post-operative and in-hospital mortality [30]. Curative resection may therefore not be offered to elderly patients particularly if they have co-morbid conditions. Smoking is strongly associated with risk of squamous cell carcinoma [4], and to a lesser extent oesophageal adenocarcinoma [31] and is associated with several co-morbidities such as chest and circulatory disease which will influence treatment options and affect recovery. There may be differences due to different levels of alcohol and tobacco use in the

prevalence of co-morbidity across regions and time for which we were not able to adjust for in the current analyses. A populationbased study in the Netherlands indeed showed co-morbidity to differ between patients with oesophageal adenocarcinoma and squamous cell carcinoma [29]. Few studies have compared survival of oesophageal cancer patients by morphology. We found higher European 1-year relative and Central European 5-year relative survival rates for oesophageal adenocarcinoma. Siewert et al. [32] reported that adenocarcinoma patients have a survival advantage over patients with squamous cell carcinoma following resection and that patients with oesophageal squamous cell carcinoma have higher rates of micrometastases. Resection rates have also been reported to be higher in patients with oesophageal adenocarcinoma compared to squamous cell carcinoma [7]. Most studies reporting overall survival by morphology did not present direct survival comparisons but showed similar survival rates between subgroups [33]. Surveillance for Barrett’s oesophagus, a pre-malignant condition leading to oesophageal adenocarcinoma [34], has increased in many localities [35]. Survival differences by morphology may be attributed to earlier stage at presentation in those with oesophageal adenocarcinoma, which has been shown to be the main predictor of long-term survival [28]. However, Alcedo et al. [36] suggest that surveillance may have little impact on oesophageal cancer survival due to the low progression rate of Barrett’s oesophagus. Other factors predictive of survival in oesophageal cancer are body mass, smoking and educational level [7] which may vary by subtype. In EUROCARE-4 only 19 registries had sufficient information on cancer stage at diagnosis. We classified stage into the broad categories of local, regional and distant stage to limit some of the regional variations in coding. In all regions survival from late stage disease was poor and showed little variation. Survival variations in persons with early disease point to opportunities for improvements in treatment and survival. Eastern Europe had 1-year relative survival rates below the European average for local, regional and unstaged cancers. This may be attributed to a staging shift, differences in staging practices or investigations undertaken between countries. Five-year relative survival rates were below the European average in Eastern and Southern Europe for those with local stage disease pointing to a need for a high resolution study of diagnostic and treatment practices of oesophageal cancer. A detailed description of the quality of EUROCARE-4 data has been published elsewhere [12]. Information on oesophageal cancer registration and survival was obtained from over 66 registries across 24 countries covering approximately 30% of the European population making this one of the largest studies to investigate oesophageal cancer survival to date. Several limitations of the data should be noted. Firstly, difficulty with case ascertainment in some regions has resulted in a large number of cases identified by death certificate only during the time period of this study (e.g. 10.2% in Austria, 12.5% in Wales and 14.1% in Thames, UK) [12]. However, death certificate only rates were low overall in EUROCARE 4 at 2.7% and even lower for oesophageal cancer at 2.0%. These cases were excluded from the current analyses potentially leading to an overestimation of the survival rates [37]. Secondly, there is no reliable information as to the anatomical sub-site of the oesophageal cancer. Surgical approach and response to treatment may differ by anatomical sub-site and may therefore carry a prognostic value. Thirdly, population coverage of cancer registration is not complete for all countries; only 14 of the 24 included in this analysis have 100% coverage [12]. Between country comparisons therefore need to be interpreted with caution as some registries (e.g. Germany, Czech Republic, Poland, Spain, Switzerland, France and Italy) may not be representative of the population. Fourthly, the staging variable,

A.T. Gavin et al. / Cancer Epidemiology 36 (2012) 505–512

defined as the ‘‘extent of disease’’, was not validated by the EUROCARE 4 team hence some misclassification of staging may be present. Fifthly, racial variation in oesophageal cancer survival has been reported, with blacks having poorer overall survival than whites in the USA [38]. We were unable to adjust for race as it was not available in the EUROCARE dataset but it is likely that the majority of patients in the EUROCARE dataset were of white European descent. Due to regional differences in the staging of oesophageal cancers bias may have resulted in the survival analyses by region and stage. For stage, we used a cut-off of 45% which is close to the International Cancer Benchmarking standards of 50%. However, it is likely that the reported survival rates by stage are not reflective of all patients with oesophageal cancer and a more complete breakdown by stage in future analysis is required. Finally, the historic nature of this data does not take into consideration recent changes in the diagnosis, staging and treatment of oesophageal cancer. In conclusion, we found continued regional variations in oesophageal cancer survival in Europe. Regional differences in relation to stage of disease presentation, point to opportunities for improvements in the diagnosis of oesophageal cancer patients in some regions. There is a need for a high resolution study of diagnostic and treatment practices of oesophageal cancer to identify the exact determinates by which this may be achieved. Conflict of interest statement None declared. Acknowledgements The EUROCARE-4 project was supported by the Compagnia di San Paolo di Torino, Italy. The Northern Ireland Cancer Registry is funded by the Public Health Agency for Northern Ireland. Appendix A Members of the EUROCARE-4 Working Group include: Austria: W Oberaigner (Tyrol Cancer Registry); M Hackl (Austrian National Cancer Registry); Belgium, E Van Eycken; K Henau (Belgian Cancer Registry), Czech Republic: J Holub, L Jurickova (West Bohemia Cancer Registry); Denmark: HH Storm; G Engholm (Danish Cancer Society, Dept. Cancer Prevention & Documentation); Finland: T Hakulinen (Finnish Cancer Registry); France: A Belot (FRANCIM); G He´delin, M Velten (Bas-Rhin Cancer Registry); I Tron, E Le Gall (Bretagne Childhood Cancer Registry); G Launoy (Calvados Digestive Cancer Registry); AV Guizard (Calvados General Cancer Registry); J Faivre, AM Bouvier (Coˆte d’Or Digestive Cancer Registry); PM Carli, M Maynadie´ (Coˆte d’Or Haematological Malignancies Registry, EA 4184); A Danzon (Doubs Cancer Registry); A Buemi (Haut-Rhin Cancer Registry); B Tretarre (He´rault Cancer Registry); B Lacour, E Desandes (Lorraine Childhood Cancer Registry); M Colonna (Ise`re Cancer Registry), F Molinie´ (Loire Atlantique Breast and Colon Cancer Registry); S Bara (Manche Cancer Registry); C Schvartz (Marne Thyroid Cancer Registry); O Ganry (Somme Cancer Registry); P Grosclaude (Tarn Cancer Registry); Germany: H Brenner (German Cancer Research Center, Heidelberg); P Kaatsch (German Childhood Cancer Registry); H Ziegler (Saarland Cancer Registry); Iceland: L Tryggvadottir (Icelandic Cancer Registry); Ireland: H Comber (National Cancer Registry of Ireland); Italy: F Berrino (Project Leader), C Allemani, P Baili, I Casella, C Margutti, L Ciccolallo, G Gatta, A Micheli, P Minicozzi, M Sant, S Sowe, C Tereanu, G Zigon (Fondazione IRCCS; ‘‘Istituto Nazionale dei Tumori’’); G Tagliabue, P Contiero (Cancer

511

Registry Unit – Varese Cancer Registry, Fondazione IRCCS, Istituto Nazionale dei Tumori); F Bellu` (Registro Tumori Adige/Tumor register Su¨dtirol); A Giacomin (Biella Cancer Registry); S Ferretti (Ferrara Cancer Registry); D Serraino L Dal Maso, M De Dottori, A De Paoli, L Zanier (Friuli Venezia Giulia Cancer Registry, Udine); M Vercelli, MA Orengo, C Casella, A. Quaglia (Liguria Cancer Registry, IRCCS AOU S. Martino-IST/Univ. Genova); S Vitarelli (Macerata Province Cancer Registry, Childhood Cancer Registry of Marche); M Federico, I Rashid, C Cirilli (Modena Cancer Registry); M Fusco (Napoli Cancer Registry); A Traina (Palermo Breast Cancer Registry); V De Lisi, F. Bozzani (Parma Cancer Registry); C Magnani, G Pastore (Piedmont Childhood Cancer Registry), R Tumino, MG La Rosa, E Spata, A Sigona (Cancer Registry Azienda Ospedaliera ‘‘Civile M.P. Arezzo’’ Ragusa, Italy); L Mangone (Reggio Emilia Cancer Registry); F Falcini, F Foca, S Giorgetti (Romagna Cancer Registry – I.R.S.T.); G Senatore, A Iannelli (Salerno Cancer Registry); M Budroni (Sassari Cancer Registry); R Zanetti, S Patriarca, S Rosso (Torino Cancer Registry); S Piffer (Trento Cancer Registry); E Paci, E Crocetti (Tuscan Cancer Registry); F La Rosa, F Stracci, T Cassetti (Umbria Cancer Registry); P Zambon, S Guzzinati (Veneto Cancer Registry, Istituto Oncologico Veneto – IRCCS, Padova); M Caldora, R Capocaccia, E Carrani, R De Angelis, S Francisci, E Grande, R Inghelmann, H Lenz, L Martina, P Roazzi, M Santaquilani, A Simonetti, A. Tavilla, A Verdecchia (Centro Nazionale di Epidemiologia, Istituto Superiore di Sanita`, Rome); Malta: M Dalmas, K England, R Micallef (Malta National Cancer Registry); Norway: F Langmark, F Bray, TB Johannesen (Cancer Registry of Norway); Poland: J Rachtan (Cracow Cancer Registry), S Go´z´dz´, U Siudowska, R Me˛z˙yk (Holycross Cancer Centre); M Bielska-Lasota (Independent Unit of Oncological Education, M. Sklodowska-Curie Cancer Centre, Warsaw); M Zwierko (Warsaw Cancer Registry); Portugal: A Miranda (Southern Portugal Cancer Registry); Slovakia: CS Diba, I Plesko (National Cancer Registry of Slovakia); Slovenia: M Primic-Zˇakelj (Cancer Registry of Slovenia); Spain: A Mateos (Albacete Cancer Registry); I Izarzugaza (Basque Country Cancer Registry); A Torrella-Ramos, Oscar Zurriaga (Comunitat Valenciana Childhood Cancer Registry/Castellon Cancer Registry); R Marcos-Gragera, ML Vilardell, A Izquierdo (Girona Cancer Registry); C Martinez-Garcia, MJ Sa´nchez (Granada Cancer Registry); C Navarro, MD Chirlaque (Murcia Cancer Registry and CIBER Epidemiologı´a y Salud Pu´blica (CIBERESP)); R Peris-Bonet (Registro Nacional de Tumores Infantiles (RNTI-SEHOP), Universitat de Vale`ncia and CIBER Epidemiologı´a y Salud Pu´blica (CIBERESP)); E Ardanaz, C Moreno (Navarra Cancer Registry and CIBERESP); J Galceran (Tarragona Cancer Registry); Sweden: A˚ Klint, M Talba¨ck, S Khan (Cancer Registry of Sweden); Switzerland: G Jundt (Basel Cancer Registry); M Usel (Geneva Cancer Registry); H Frick (Grisons Cancer Registry); SM Ess (St. Gall Cancer Registry); A Bordoni (Ticino Cancer Registry); I Konzelmann (Valais Cancer Registry); S Dehler (Zurich Cancer Registry); JM Lutz, P. Pury (Co-ordinating Centre); The Netherlands: S Siesling, O Visser, R Otter (Comprehensive Cancer Centre the Netherlands); JWW Coebergh, ML Janssen-Heijnen, Louis van der Heijden (Eindhoven Cancer Registry); UK – England: DC Greenberg (Eastern Cancer Registration and Information Centre); MP Coleman, Laura Woods (London School of Hygiene and Tropical Medicine); T Moran (North West Cancer Intelligence Service); D Forman (Northern and Yorkshire Cancer Registry and Information Service); N Cooper (Office for National Statistics); M Roche (Oxford Cancer Intelligence Unit), J Verne (South West Cancer Intelligence Services); H Møller (Thames Cancer Registry); D Meechan, J Poole (Trent Cancer Registry); G Lawrence (West Midlands Cancer Intelligence Unit); UK – England/Wales: C Stiller (Childhood Cancer

512

A.T. Gavin et al. / Cancer Epidemiology 36 (2012) 505–512

Research Group); UK – Northern Ireland: A Gavin (Northern Ireland Cancer Registry); UK – Scotland: RJ Black, DH Brewster (Scottish Cancer Registry); UK – Wales: JA Steward (Welsh Cancer Intelligence and Surveillance Unit). References [1] Bosetti C, Levi F, Ferlay J, Garavello W, Lucchini F, Bertuccio P, et al. Trends in oesophageal cancer incidence and mortality in Europe. Int J Cancer 2008;122(5):1118–29. [2] Vial M, Grande L, Pera M. Epidemiology of adenocarcinoma of the esophagus, gastric cardia, and upper gastric third. Recent Results Cancer Res 2010;182:1–17. [3] Botterweck AA, Schouten LJ, Volovics A, Dorant E, van Den Brandt PA. Trends in incidence of adenocarcinoma of the oesophagus and gastric cardia in ten European countries. Int J Epidemiol 2000;29:645–54. [4] Steevens J, Schouten LJ, Goldbohm RA, van den Brandt PA. Alcohol consumption, cigarette smoking and risk of subtypes of oesophageal and gastric cancer: a prospective cohort study. Gut 2010;59(1):39–48. [5] Faivre J, Forman D, Este`ve J, Gatta G. Survival of patients with oesophageal and gastric cancers in Europe. Eur J Cancer 1998;14:2167–75. [6] Crane LMA, Schaapveld M, Visser O, Louwman MW, Plukker JT, van Dam GM. Oesophageal cancer in The Netherlands: increasing incidence and mortality but improving survival. Eur J Cancer 2007;43(9):1445–51. [7] Sundelo¨f M, Lagergren J, Ye W. Patient demographics and lifestyle factors influencing long-term survival of oesophageal cancer and gastric cardia cancer in a nationwide study in Sweden. Eur J Cancer 2008;44(11):1566–71. [8] Thrift AP, Nagle CM, Fahey PP, Russell A, Smithers BM, Watson DI, et al. The influence of prediagnostic demographic and lifestyle factors on esophageal squamous cell carcinoma survival. Int J Cancer 2012;131(5):E759–68. [9] Sundelo¨f M, Ye W, Dickman PW, Lagergren J. Improved survival in both histologic types of oesophageal cancer in Sweden. Int J Cancer 2002;99 (5):751–4. [10] van de Poll-Franse LV, Lemmens VE, Roukema JA, Coebergh JW, Nieuwenhuijaen GA. Impact of concentration of oesophageal and gastric cardia cancer surgery on long-term population-based survival. Br J Surg 2011;98(7):956–63. http://dx.doi.org/10.1002/bjs.7493. [11] Berrino F, Verdecchia A, Lutz JM, Lombardo C, Micheli A, Capocaccia R, EUROCARE Working Group. Comparative cancer survival information in Europe. Eur J Cancer 2009;45:901–8. [12] DeAngelis R, Francisci S, Baili P, Marchesi F, Roazzi P, Belot A, et al. The EUROCARE-4 database on cancer survival in Europe: data standardisation, quality control and methods of statistical analysis. Eur J Cancer 2009;45: 909–30. [13] Fritz A, Percy C, Jack A, Shanmugarathnam K, Sobin L, Parkin DM, et al. International classification of diseases for oncology, 3rd ed., Geneva: WHO, 1990. [14] Hakulinen T. Cancer survival corrected for heterogeneity in patient withdrawal. Biometrics 1982;38(4):933–42. [15] Corazziari I, Quinn M, Capocaccia R. Standard cancer patient population for estimating age standardising survival ratios. Eur J Cancer 2004;40:2307–16. [16] Brenner H, Gefeller O. Deriving more up-to-date estimates of long-term patient survival. J Clin Epidemiol 1997;50:211–6. [17] Surveillance research program, National Cancer Institute SEER*Stat software. Version 6.4.4; 2008, http://www.seer.cancer.gov/seerstat. [18] Rao S, Cunningham D. Survival from cancer of the oesophagus in England and Wales up to 2001 – clinical commentary. Br J Cancer 2008;99(Suppl. 1):S14–5. [19] Mitry E, Rachet B, Quinn MJ, Cooper N, Coleman MP. Survival from cancer of the oesophagus in England and Wales up to 2001 – survival analysis. Br J Cancer 2008;99(Suppl. 1):S11–3. [20] Wouters MW, Wijnhoven BP, Karim-Kos HE, Blaauwgeers HG, Stassen LP, Steup WH, et al. High-volume versus low-volume for esophageal resections for

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28]

[29]

[30]

[31]

[32]

[33] [34]

[35]

[36]

[37]

[38]

cancer: the essential role of case-mix adjustments based on clinical data. Ann Surg Oncol 2008;15(1):80–7. Thompson AM, Rapson T, Gilbert FJ, Park KGM. Endoscopic palliative treatment for esophageal and gastric cancer: techniques, complications and survival in a population-based cohort of 948 patients. Surg Endosc 2004; 18:1257–62. Hofstetter W, Swisher SG, Correa AM, Hess K, Punam Jr JB, Ajani JA, et al. Treatment outcomes of resected esophageal cancer. Ann Surg 2002; 236(3):376–84 [discussion 384–5]. Verdecchia A, Francisci S, Brenner H, Gatta G, Micheli A, Mangone L, et al. Recent cancer survival in Europe: a 2000-02 period analysis of EUROCARE-4 data. Lancet Oncol 2007;8(9):784–96. Ries LAG, Young JL, Keel GE, Eisner MP, Lin YD, Horner M-J (editors). SEER Survival Monograph: cancer survival among adults: U.S. SEER Program, 1988– 2001, Patient and tumor characteristics. Chapter 3: cancers of the esophagus, stomach & small intestine. Bethesda, MD: National Cancer Institute, SEER Program, NIH Pub. No. 07-6215; 2007. Micheli A, Ciampichini R, Overaigner W, Ciccolallo L, de Vries E, Izarzugaza I, et al. The advantage of women in cancer survival: an analysis of EUROCARE-4 data. Eur J Cancer 2009;45(6):1017–27. Newnham A, Quinn MJ, Babb P, Kang JY, Majeeds A. Trends in oesophageal and gastric cancer incidence, mortality and survival in England and Wales 1971– 1998/1999. Aliment Pharmacol Ther 2003;17:655–64. Rouvelas I, Zeng W, Lindblad M, Viklund P, Ye W, Lagergren J. Survival after surgery for oesophageal cancer: a population-based study. Lancet Oncol 2005;6(11):864–70. Trivers KF, De Roos AJ, Gammon MD, Vaughan TL, Risch HA, Olshan AF, et al. Demographic and lifestyle predictors of survival in patients with oesophageal or gastric cancers. Clin Gastroenterol Hepatol 2005;3:225–30. Koppert LB, Janssen-Heijnen ML, Louwman MW, Lemmens VE, Wijhoven BP, Tilanus HW, et al. Comparison of comorbidity prevalence in oesophageal and gastric carcinoma patients: a population-based study. Eur J Gastroenterol Hepatol 2004;16(7):681–8. Cijs TM, Verhoef C, Steyerberg EW, Koppert LB, Tran TC, Wijhoven BP, et al. Outcome of esophagectomy for cancer in elderly patients. Ann Thorac Surg 2010;90(3):900–7. Cook MB, Kamangar F, Whiteman DC, Freedman ND, Gammon MD, Bernstein L, et al. Cigarette smoking and adenocarcinomas of the esophagus and esophagogastric junction: a pooled analysis from the International BEACON consortium. J Natl Cancer Inst 2010;102(17):1344–53. Siewert JR, Stein HJ, Feith M, Bruecher BL, Bartels H, Fink U. Histologic tumor type is an independent prognostic parameter in esophageal cancer: lessons from more than 1,000 consecutive resections at a single center in the Western world. Ann Surg 2001;234(September (3)):360–7 [discussion 368–9]. Oliver SE, Robertson CS, Logan RFA. Oesophageal cancer: a population-based study of survival after treatment. Br J Surg 1992;79:1321–5. Yousef F, Cardwell C, Cantwell MM, Galway K, Johnston BT, Murray L. The incidence of esophageal cancer and high-grade dysplasia in Barrett’s esophagus: a systematic review and meta-analysis. Am J Epidemiol 2008;168(3): 237–49. van Soest EM, Dieleman JP, Siersema PD, Sturkenboom MC, Kuipers EJ. Increasing incidence of Barrett’s oesophagus in the general population. Gut 2005;54:1062–6. ˜ a F, Ortego J, Sainz R, et al. Trends in Alcedo J, Ferrandez A, Arenas J, Sopen Barrett’s esophagus diagnosis in Southern Europe: implications for surveillance. Dis Esophagus 2009;22(3):239–48. Robinson D, Shankila R, Hakulinen T, Moller H. Interpreting international comparisons of cancer survival: the effects of incomplete registration and the presence of death certificate only cases on survival estimation. Eur J Cancer 2007;43:903–13. Greenstein AJ, Litle VR, Swanson SJ, Divino CM, Packer S, McGinn TG, et al. Racial disparities in esophageal cancer treatment and outcomes. Ann Surg Oncol 2008;15(3):881–8.