- Email: [email protected]
Lung cancer survival in Germany: A population-based analysis of 132,612 lung cancer patients
G Model LUNG-4959; No. of Pages 6
ARTICLE IN PRESS Lung Cancer xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Lung Cancer journal homepage: www.elsevier.com/locate/lungcan
Lung cancer survival in Germany: A population-based analysis of 132,612 lung cancer patients Andrea Eberle a,∗ , Lina Jansen b , Felipe Castro b , Agne Krilaviciute b , Sabine Luttmann a , Katharina Emrich c , Bernd Holleczek d , Alice Nennecke e , Alexander Katalinic f , Hermann Brenner b,g , the GEKID Survival Working Group a
Bremen Cancer Registry, Leibniz-Institute for Prevention Research and Epidemiology—BIPS, Bremen, Germany Division of Clinical Epidemiology and Aging Research, German Cancer Research Center (DKFZ), Heidelberg, Germany c Cancer Registry of Rhineland-Palatinate, Institute for Medical Biostatistics, Epidemiology and Informatics, University Medical Center, Johannes Gutenberg University Mainz, Mainz, Germany d Saarland Cancer Registry, Saarbrücken, Germany e Hamburg Cancer Registry, Authority for Health and Consumer Protection, Hamburg, Germany f Cancer Registry of Schleswig-Holstein, Institute of Cancer Epidemiology, University of Lübeck, Lübeck, Germany g German Cancer Consortium (DKTK), Heidelberg, Germany b
a r t i c l e
i n f o
Article history: Received 18 August 2014 Received in revised form 18 May 2015 Accepted 4 October 2015 Keywords: Lung cancer survival Population-based study Cancer registry Germany Period analysis
a b s t r a c t Objectives: Lung cancer is the most common cancer-related death worldwide. In Germany it accounts for 25% of cancer deaths in men, and 14% in women. The aim of this study is to provide an overview of 5-year relative survival by sex, age, histology, and tumour stage in Germany representing a population of 26.7 million people. Materials and methods: The study is based on a pooled German dataset including data from 12 populationbased cancer registries covering around one third of the German population. A total of 132,612 patients diagnosed with lung cancer from 2002 to 2010 were included in the analysis. Survival estimates for the time period 2007–2010 were calculated using period analysis. Differences in survival between sexes were tested for statistical significance by model-based period analysis (poisson regression model). The relative excess risk (RER) of death (women vs. men) was extracted from the model with the p value for the difference in RER. Results: The overall age adjusted 5-year relative survival was 15.5% (standard error (SE) 0.2) for men and 20.3% (SE 0.3) in women. Survival differed markedly according to age (men: <60 years 18.5% vs. 80+ years 8.4% and women 23.7% vs. 10.6%, respectively), histology (largest difference between histological groups: men 25.7 and women 44.4% points) and stage (men: UICC Ia 62.9%, vs. UICC IV 4.6% and women 75.2% vs. 7.0%, respectively). Our study showed survival advantages for women compared to men, most notably in younger aged patients (RER 0.83, p < 0.0001), patients with adenocarcinoma (RER 0.80, p < 0.0001), and patients with lower stage cancer (RER 0.62, p < 0.0001). Conclusions: This study presents up-to-date survival estimates for lung cancer in Germany. Compared to other European countries survival was relatively high. Women showed higher survival than men independent of age, histology and stage. The reasons for the survival differences require further clarification. © 2015 Elsevier Ireland Ltd. All rights reserved.
1. Introduction
∗ Corresponding author at: Leibniz-Institute for Prevention Research and Epidemiology—BIPS, Department Biometry and Data Management, Unit Statistical Modelling of Secondary and Registry Data, Achterstrasse 30, D-28359 Bremen, Germany. Fax: +49 421 218 56941. E-mail address: [email protected] (A. Eberle).
In Germany as well as in other European industrialised countries, lung cancer is among the three leading forms of cancer both in terms of incidence and mortality. Since the end of the 1990s age-standardised incidence and mortality rates are showing an opposing trend between men and women: Incidence and mortality have decreased steadily for men by around 20% but have increased
http://dx.doi.org/10.1016/j.lungcan.2015.10.007 0169-5002/© 2015 Elsevier Ireland Ltd. All rights reserved.
Please cite this article in press as: A. Eberle, et al., Lung cancer survival in Germany: A population-based analysis of 132,612 lung cancer patients, Lung Cancer (2015), http://dx.doi.org/10.1016/j.lungcan.2015.10.007
G Model LUNG-4959; No. of Pages 6
ARTICLE IN PRESS A. Eberle et al. / Lung Cancer xxx (2015) xxx–xxx
2
in women by approximately 30% [1,2]. This development is caused by the changing behaviour with respect to smoking. Nonetheless, currently there are still two-times more men than women diagnosed with lung cancer in Germany [3]. The prognosis for lung tumours is unfavourable and improvements in survival in recent decades have been minimal. Between the mid-1970s and the beginning of the millennium, 5-year relative survival from lung cancer in the USA increased only modestly from 13% to 16% [4]. For Europe, the EUROCARE-5 study (European Registry Based Study on Survival and Care of Cancer Patients) estimated a mean age-standardised 5-year lung cancer survival of 13% for patients diagnosed in 2000–2007. Relative survival varied nearly twofold between European countries, with the lowest survival being observed in Bulgaria (6.2%), UK with Wales (8.6%) and Scotland (8.7%), and the highest survival in Germany (15.6%) and Austria (16.7%) [5]. Differences in health care systems as well as in the distribution of diagnostic and treatment practices in the whole population are possible explanations for geographical differences in survival [6]. Further explanations are differences in the used data base especially regarding the quality of cancer registration which includes completeness of data and ascertainment of vital status [7]. In this study, using population-based cancer registry data, we analysed lung cancer survival in Germany with respect to sex, age, stage at diagnosis and histology. The aim of this investigation is the identification of factors which affect the relatively high survival rates for lung cancer in Germany. 2. Materials and methods 2.1. Study population The study is based on a pooled national dataset including data from cancer registries in 12 out of 16 federal states in Germany, covering a population of 26.7 million people. Only regions with sufficient data quality were included in the analysis data set. Data quality was assessed by proportion of death certificate only notified and autopsy only (DCO) cases over all malignant cancers. Cancer registries were included if the overall proportion of DCO cases was below 13% in 2002–2010, or decreased steadily over time and was overall below 14%. A cut-off at 13% was chosen, as the same cut-off was used in the European Cancer Registry based study on survival and care of cancer patients (EUROCARE-5 study) [5]. If these criteria were not fulfilled for the whole region covered by a cancer registry, only those administrative districts that fulfilled these criteria were included. The analysis included patients with primary invasive lung cancer (ICD-10: C33-34) diagnosed in 2002–2010 who were at least 15 years old at the time of diagnosis. Patients were followed up with respect to vital status until December 31, 2010. Cases notified by DCO were excluded. Topography and histology of tumours were coded according to ICD-O-3 [8]. The histology codes were grouped into the following six categories: adenocarcinoma, squamous cell carcinoma, large cell carcinoma, small cell carcinoma, other specified tumours and other unspecified tumours [9]. Tumour stages were coded in accordance with the tumour–node–metastasis (TNM) classification of malignant tumour [10]. For the analyses, patients were categorized by sex, age at diagnosis (15–59, 60–69, 70–79 and 80+ years), tumour histology and stage at diagnosis. 2.2. Statistical methods Five-year relative survival estimates for the time period 2007–2010 were calculated using period analysis, providing more
up-to-date survival estimates than traditional cohort based analysis. The period analysis included only survival experience during the period from 2007 to 2010 [11]. Relative survival was derived as a ratio of the observed survival of the included patients divided by the expected survival of the underlying population. Expected survival was computed according to the Ederer II method [12] using national life tables stratified by age, sex, and calendar year obtained from the German Federal Statistical Office. We calculated 5-year relative survival stratified by sex, age group, stage and histological type. Age-standardised survival was estimated using weights defined by the International Cancer Survival Standard [13]. Model-based period analysis was used to investigate differences in survival between men and women and to perform statistical significance testing [14]. Poisson regression models were fit to the number of excess deaths, using year of follow-up and sex as explanatory variables (of categorical type) and included the logarithm of the person-years at risk as offset. The relative excess risk (RER) of death for women compared to men was extracted from the model together with the corresponding p value for the difference in RER between the sexes. Analyses were repeated with additional adjustment for age, histology group, and stage at diagnosis. All analyses were performed with the SAS software (version 9.2), using special macros for period analysis [15] and their adaptation for model-based analysis [14]. Statistical significance was tested two-sided using a significance level of 0.05.
3. Results After exclusion of DCO cases (13.5%) a total of 132,612 lung cancer cases were included in the analysis (Table 1). The percentage of DCO cases varied between registries, ranging from 8.0% in Saarland to 16.4% in Lower Saxony. The proportion of cancers with microscopic confirmation was 93.3%. Nearly three-quarters of lung cancer tumours (72%) were diagnosed in men. The basic characteristics of the patients included in the analyses according to sex, histological group, age, period of diagnosis, and stage are presented in Table 2. Among men, the most common histological category was squamous cell carcinoma (32.0%), with a substantially higher proportion in men than in women (17.1%). Adenocarcinoma was the second most common histological group (27.3%). Among women, the most common histological groups are adenocarcinoma (37.6%), other unspecified tumours (17.5%), and squamous cell carcinoma (17.1%). Large cell carcinoma was the least frequent histological group in both sexes (3.1%). In both sexes, lung cancer patients in the oldest age group (80+ years) had the highest proportion of unspecified tumours (men 32.8%, women 39.2%) whereas patients in the youngest age group (15–69 years) had the highest proportion of adenocarcinoma (men 31.7%, women 43.7%). In both sexes, the number of lung cancer patients rose steadily from 39,207 cases in 2002–2004 to 48,326 cases in 2008–2010. Staging of cancer at the time of diagnosis is the most important predictor of survival, and treatment options are based on stage. In our study, information on stage of tumours at diagnosis was often missing (44%). Because lung cancer is asymptomatic in early stages, only 14% of cases with stage information were classified in stage I. There was no difference in the proportion of tumours with earlier stages (Ia,b) between men and women. In both sexes, small cell carcinoma patients had the highest proportion of stage IV tumours (men: 72.4%; women: 69.1% of those with stage information). The overall age adjusted 5-year relative survival for the time period 2007–2010 was 16.9%, with substantial variation between men (15.5%) and women (20.3%; Table 3). Survival varied by age, with increasing age being associated with reduced relative survival.
Please cite this article in press as: A. Eberle, et al., Lung cancer survival in Germany: A population-based analysis of 132,612 lung cancer patients, Lung Cancer (2015), http://dx.doi.org/10.1016/j.lungcan.2015.10.007
G Model
ARTICLE IN PRESS
LUNG-4959; No. of Pages 6
A. Eberle et al. / Lung Cancer xxx (2015) xxx–xxx
3
Table 1 Description of participating cancer registries and of the dataset used in period survival analysis of lung cancer patients diagnosed between 2002 and 2010 in Germany. Registry
Underlying population 2010 (million)
Brandenburg Bremen Hamburg Lower Saxonya Mecklenburg-Vorpommern North Rhine-Westphaliab Rhineland-Palatinatec Saarland Saxony Saxony-Anhaltd Schleswig-Holsteine Thüringen
a b c d e
2.50 0.66 1.79 7.91 1.64 2.59 0.51 1.02 4.15 0.32 1.38 2.24 26.7
Total
Cases
Exclusion based on death certificate or autopsy only notification (%)
Cases in analysis
Number of deaths (%)
Proportion male %
Median age at diagnosis
Microscopically confirmed cases (%)
14,815 5075 11,339 39,895 10,245 16,489 3081 7632 21,971 2169 8741 11,919
14.0 8.8 14.3 16.4 10.6 12.4 15.0 8.0 12.5 11.0 16.2 12.4
12,741 4630 9721 33,348 9157 14,450 2618 7025 19,224 1930 7323 10,445
9710 (76.2%) 3600 (77.8%) 7590 (78.1%) 25,676 (77.0%) 7226 (78.9%) 11,126 (77.0%) 1999 (76.4%) 5493 (78.2%) 14,846 (77.2%) 1369 (70.9%) 5634 (76.9%) 8017 (76.8%)
75.4 66.6 63.5 70.7 75.2 70.8 72.1 70.7 77.3 71.6 67.3 76.2
68 68 67 68 68 68 68 68 69 68 67 68
94.2 92.2 95.7 89.5 92.2 97.1 98.3 94.9 91.5 96.5 96.5 95.0
153,371
13.5
132,612
102,286 (77.1%)
72.0
68
93.3
Diagnosis period included 2003–2010. Selected administrative districts: Regierungsbezirk Münster. Selected administrative district: Regierungsbezirk Trier. Selected administrative district: Cities Halle and Dessau-Roßlau. Selected administrative districts: Cities Flensburg, Kiel and Neumünster and districts Nordfriesland, Ostholstein, Plön, Rendsburg-Eckernförde, Schleswig-Flensburg.
Table 2 Distribution of histology by age, period of diagnosis and stage for male and female lung cancer patients, A: male, B: female. A Male
Squamous cell n (%)
Large cell n (%)
6284 (24.1) 9408 (36.0) 8422 (32.3) 1994 (7.6)
5575 (18.2) 11,016 (36.0) 11,450 (37.4) 2570 (8.4)
679 (23.0) 1079 (36.5) 966 (32.7) 235 (7.9)
4153 (24.2) 6340 (37.0) 5372 (31.3) 1285 (7.5)
545 (25.1) 767 (35.3) 724 (33.3) 138 (6.4)
Period of diagnosis 7170 (27.5) 2002–04 8953 (34.3) 2005–07 2008–10 9985 (38.2)
9551 (31.2) 10,879 (35.5) 10,181 (33.3)
1062 (35.9) 920 (31.1) 977 (33.0)
5444 (31.7) 6011 (35.0) 5695 (33.2)
Stage UICC Ia UICC Ib UICC IIa UICC IIb UICC IIIa UICC IIIb UICC IV Missing All cases
1138 (3.7) 2162 (7.1) 232 (0.8) 1619 (5.3) 2499 (8.2) 3926 (12.8) 6869 (22.4) 12,166 (39.7) 30,611 (100)
91 (3.1) 176 (6.0) 17 (0.6) 101 (3.4) 179 (6.1) 299 (10.1) 1066 (36.0) 1030 (34.8) 2959 (100)
137 (0.8) 197 (1.2) 58 (0.3) 165 (1.0) 599 (3.5) 1519 (8.9) 7009 (40.9) 7466 (43.5) 17,150 (100)
Age 15–59 60–69 70–79 80+
Adeno-carcinoma n (%)
1224 (4.7) 1316 (5.0) 180 (0.7) 673 (2.6) 1342 (5.1) 1974 (7.6) 9039 (34.6) 10,360 (39.7) 26,108 (100)
Small cell n (%)
Other specified n (%)
Other unspecified n (%)
Missing n (%)
All cases n (%)
2612 (15.9) 4892 (29.7) 5903 (35.9) 3043 (18.5)
7 (9.1) 15 (19.5) 33 (42.9) 22 (29.3)
19,855 (20.8) 33,517 (35.1) 32,870 (34.4) 9287 (9.7)
553 (25.4) 721 (33.2) 900 (41.4)
5385 (32.7) 5718 (34.8) 5347 (32.5)
17 (22.1) 25 (32.5) 35 (45.5)
29,182 (30.6) 33,227 (34.8) 34,020 (35.6)
198 (9.1) 162 (7.5) 29 (1.3) 99 (4.6) 137 (6.3) 132 (6.1) 444 (20.4) 973 (44.8) 2174 (100)
221 (1.3) 310 (1.9) 36 (0.2) 220 (1.3) 475 (2.9) 847 (5.2) 5023 (30.5) 9318 (56.6) 16,450 (100)
0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 1 (1.3) 1 (1.3) 75 (97.4) 77 (100)
3009 (3.2) 4323 (4.5) 552 (0.6) 2877 (3.0) 5231 (5.5) 8698 (9.1) 29,451 (30.8) 41,388 (43.3) 95,529 (100)
B Female
Age 15–59 60–69 70–79 80+
Adeno-carcinoma
Squamous cell
Large cell
Small cell
Other specified
Other unspecified
Missing
All cases
n (%)
n (%)
n (%)
n (%)
n (%)
n (%)
n (%)
n (%)
4412 (31.6) 4198 (30.1) 3825 (27.4) 1509 (10.8)
1371 (21.6) 2051 (32.4) 2179 (34.4) 737 (11.6)
357 (32.2) 347 (31.4) 300 (27.1) 103 (9.3)
2430 (31.4) 2623 (33.9) 2078 (26.9) 609 (7.9)
511 (35.5) 452 (31.4) 366 (25.4) 111 (7.7)
1120 (17.3) 1461 (22.5) 1918 (29.6) 1986 (30.6)
2 (6.9) 2 (6.9) 11 (37.9) 14 (48.3)
10,203 (27.5) 11,134 (30.0) 10,677 (28.8) 5069 (13.7)
Period of diagnosis 3441 (24.7) 2002–04 4752 (34.1) 2005–07 5751 (41.2) 2008–10
1742 (27.5) 2283 (36.0) 2313 (36.5)
380 (34.3) 349 (31.5) 378 (34.1)
2193 (28.3) 2634 (34.0) 2913 (37.6)
354 (24.6) 502 (34.9) 584 (40.5)
1913 (29.5) 2220 (34.2) 2352 (36.3)
2 (6.9) 12(41.4) 15 (51.7)
10,025 (27.0) 12,752 (34.4) 14,306 (38.6)
Stage UICC Ia UICC Ib UICC IIa UICC IIb UICC IIIa UICC IIIb UICC IV Missing All cases
263 (4.2) 383 (6.0) 50 (0.8) 276 (4.4) 441 (7.0) 683 (10.8) 1541 (24.3) 2701 (42.6) 6338 (100)
31 (2.8) 47 (4.3) 3 (0.3) 32 (2.9) 56 (5.1) 109 (9.9) 415 (37.5) 414 (37.4) 1107 (100)
86 (1.1) 90 (1.2) 37 (0.5) 71 (0.9) 263 (3.4) 721 (9.3) 2834 (36.6) 3638 (47.0) 7740 (100)
212 (14.7) 111 (7.7) 15 (1.0) 37 (2.6) 64 (4.4) 63 (4.4) 208 (14.4) 730 (50.7) 1440 (100)
93 (1.4) 121 (1.9) 10 (0.2) 64 (1.0) 131 (2.0) 268 (4.1) 1972 (30.4) 3826 (59.0) 6485 (100)
29 (100) 29 (100)
1490 (4.0) 1455 (3.9) 231 (0.6) 795 (2.1) 1664 (4.5) 2814 (7.6) 11,661 (31.5) 16,973 (45.8) 37,083 (100)
805 (5.8) 703 (5.0) 116 (0.8) 315 (2.3) 709 (5.1) 970 (7.0) 4691 (33.6) 5635 (40.4) 13,944 (100)
Please cite this article in press as: A. Eberle, et al., Lung cancer survival in Germany: A population-based analysis of 132,612 lung cancer patients, Lung Cancer (2015), http://dx.doi.org/10.1016/j.lungcan.2015.10.007
G Model
ARTICLE IN PRESS
LUNG-4959; No. of Pages 6
A. Eberle et al. / Lung Cancer xxx (2015) xxx–xxx
4
Table 3 Age group-specific and age-standardised 5-year relative survival of lung cancer patients during the period 2007-2010 in Germany. Sex differences are presented as estimates of relative excess risk (RER) of death for women compared to men. Age group
Sex differences by age Male
Female RS, %
N
15–59 60–69 70–79 80+ Overall age adjusted
19,855 33,517 32,870 9287 95,529
18.5 17.0 13.8 8.4 15.5
SE
Adjusted for histology and stage
N
0.4 0.3 0.3 0.6 0.2
RS, %
10,203 11,134 10,677 5069 37,083
SE
23.7 22.9 19.0 10.6 20.3
RER
0.6 0.6 0.6 0.8 0.3
p value
0.86 0.83 0.86 0.98 0.87
<.0001 <.0001 <.0001 0.4383 <.0001
RER
0.68 0.68 0.74 0.78 0.81
p value
<.0001 <.0001 <.0001 <.0001 <.0001
Adj. RS, % Male
Female
19.3 18.6 16.8 12.9 17.5
23.0 23.1 19.7 14.7 21.0
RS: relative survival, SE: standard error of the estimate, RER: relative excess risk of death for women vs. men. Table 4 Five-year relative survival of lung cancer patients during the period 2007-2010 in Germany by histological groups and sex. Sex differences are presented as estimates of relative excess risk (RER) of death for women compared to men. Histological group
Sex differences by histology Male
Female
N
Adenocarcinoma Squamous Cell Large Cell Small Cell Other specified Other unspecified
RS, %
26,108 30,611 2959 17,150 2174 16,450
18.2 19.8 17.0 7.3 34.1 8.4
SE
Adjusted for age and stage
N
0.4 0.4 1.3 0.4 1.7 0.4
RS, %
13,944 6338 1107 7740 1440 6485
SE
24.8 22.7 21.3 10.7 54.1 9.7
0.6 0.9 2.1 0.6 2.2 0.6
RER
p value
0.80 0.95 0.90 0.86 0.54 0.94
<.0001 0.0001 0.0008 <.0001 <.0001 <.0001
RER
0.75 0.94 0.75 0.85 0.64 0.87
p value
<.0001 0.0361 0.0002 <.0001 <.0001 <.0001
Adj. RS, % Male
Female
20.0 22.2 16.7 8.3 38.0 9.6
25.4 23.0 21.8 10.8 45.7 11.5
RS: relative survival, SE: standard error of the estimate, RER: relative excess risk of death for women vs. men. Table 5 Five-year relative survival of lung cancer patients in the period 2007-2010 in Germany by stage and sex. Sex differences are presented as estimates of relative excess risk of death for women compared to men. Stage
Sex differences by stage Male N
UICC Ia UICC Ib UICC IIa UICC IIb UICC IIIa UICC IIIb UICC IV Missing
3009 4323 552 2877 5231 8698 29,451 41,388
Female RS, %
SE
62.9 53.4 39.7 35.6 20.3 12.5 4.6 14.9
1.8 1.5 3.8 1.6 1.1 0.6 0.2 0.3
N
1490 1455 231 795 1664 2814 11,661 16,973
Adjusted for age and histology RS, %
SE
RER
p value
RER
p value
75.2 64.3 53.9 41.5 28.0 15.7 7.0 19.9
2.2 2.1 7.5 2.8 1.8 1.1 0.4 0.5
0.62 0.68 0.66 0.91 0.77 0.89 0.86 0.87
<.0001 <.0001 <.0001 0.0157 <.0001 <.0001 <.0001 <.0001
0.50 0.51 0.56 0.71 0.68 0.82 0.81 0.78
<.0001 <.0001 0.0021 <.0001 <.0001 <.0001 <.0001 <.0001
Adj. RS, % Male
Female
64.9 53.1 39.2 36.2 20.1 13.0 5.5 22.7
73.9 64.2 54.1 40.1 28.5 15.6 8.0 27.0
RS: relative survival, SE: standard error of the estimate, RER: relative excess risk of death for women vs. men.
Across all age groups, 5-year relative survival was higher in women than in men even after adjustment for histology and stage. Survival varied by histology (Table 4). In both sexes, patients with small cell carcinoma had the worst survival (men 7.3% and women 10.7%). Significant sex-related differences were observed for all histological groups. Results of the Poisson regression analysis showed that especially women with adenocarcinoma or “other specified tumours” had the lowest relative excess risk of dying (RER 0.80 and 0.54, respectively) compared to men. The survival advantages of women with squamous cell carcinoma or small cell carcinoma were practically unchanged after the adjustment for age and stage. For patients with other specified tumours, 5-year relative survival was much higher than for all other histological groups in both men (34.1%) and women (54.1%). The group ‘other specified tumours’ particularly included adenosquamous carcinoma (46.8%) and carcinoid tumours (32.8%). Table 5 shows 5-year relative survival by stage and sex. Survival differed strongly according to stage at diagnosis. In stage Ia,
5-year survival was above 60%; however, 5-year relative survival decreased to below 5% in men and 7% in women when patients presented with metastatic disease at diagnosis (stage IV). A substantial survival advantage of women was observed among lung cancer patients diagnosed in stage Ia, Ib and IIa. The risk of dying within five years past diagnosis was significantly lower in women than in men among all stages even after adjustment for age and histology. The sex difference was most evident in patients with earlier stages (UICC Ia, Ib and IIa), where women had an approximately 50% decreased risk of dying.
4. Discussion The present analysis of lung cancer survival was based on a very large population-based cancer database from Germany, which enabled us to present a comprehensive overview of 5-year survival of lung cancer patients. The overall age-adjusted 5-year survival
Please cite this article in press as: A. Eberle, et al., Lung cancer survival in Germany: A population-based analysis of 132,612 lung cancer patients, Lung Cancer (2015), http://dx.doi.org/10.1016/j.lungcan.2015.10.007
G Model LUNG-4959; No. of Pages 6
ARTICLE IN PRESS A. Eberle et al. / Lung Cancer xxx (2015) xxx–xxx
was 16.9% in 2007–2010. Relative survival estimates were significantly higher in women than in men, even after adjustment for histology and stage. Survival differed markedly according to age, histological group and stage at diagnosis. Our calculated overall 5-year survival was only slightly higher than the German lung cancer survival estimate reported in the EUROCARE-5 study for the earlier period 2000–2007 (15.6%) [5]. We observed a strong age gradient in both sexes. Increasing age is generally considered to be an indicator for poor prognosis and we found, as expected, that survival decreased rapidly with increasing age in Germany. Poorer survival in the elderly population has been ascribed to higher prevalence of co-morbidities. These are known to be a major reason why patients and especially older patients do not undergo surgery or other effective treatments [16]. Furthermore, the prognosis strongly varied by sex, histological group and stage. In our study, the prognosis of men was considerably worse than that of women. These results are in agreement with findings from the EUROCARE project, which reported a better survival of women under the age of 65 years compared to men [17]. However, sex differences in the current study were even more pronounced. Our study showed an age-adjusted survival difference of 4.8 percent units between men and women, compared to 2.4 percent units in the EUROCARE-4 database for patients diagnosed in 1995–1999 with smallest difference in UK and Ireland (1.2 percent units). The female survival advantage was most marked in middle age groups. In recent European studies, a female survival disadvantage with older age group was reported [17,18]. Micheli et al. suggested that age at diagnosis might be a proxy for biological factors which changed more markedly in women than in men with increasing age [17]. Variation in tumour cell biology such as expression and mutation rates of multiple regulating gene pathways, differences in DNA repair capacity, and changes in reproductive hormones are possible explanations for the differences between men and women. Furthermore, differences in prevalence of comorbidities especially cardiovascular diseases could have therapeutic and prognostic consequences. In particular, cardiovascular disease seems to influence the survival of lung cancer patients [19]. In order to further elucidate potential underlying reasons for differences in survival, patients should ideally not only be classified by sex, age, histology and stage, but also by the presence of prognostically relevant comorbidity. Additionally, survival estimates differ according to the histological group of the tumours. Further, the distribution of the histological groups differs significantly between men and women. In some European countries, e.g. Italy, Austria, and the Scandinavian countries, the incidence of adenocarcinoma in men has exceeded the incidence of squamous cell carcinoma, which has so far been the most common histological subtype [2]. In women, however, adenocarcinoma is the most common histological group in almost all European countries except Poland and Bulgaria [2]. In our study, patients with adenocarcinoma showed relatively high survival in both sexes. It is noticeable that small cell carcinoma belongs to the most frequent histological groups in Scotland [2]. This could be an explanation for the remarkably low lung cancer survival in this country. In our study, 5-year relative survival was lowest in patients diagnosed with small cell lung cancer. The relatively high survival estimates observed for ‘other specified tumours’ are in accordance with a previous population-based study conducted in Spain [20]. An explanation for this finding is the high proportion of less aggressive tumours in this category. In our study, approximately one third of lung cancers in this heterogeneous group were neuroendocrine (carcinoid) tumours which tend to grow slowly and only rarely spreads beyond the lungs [21]. In our study, other specified tumours had the highest proportion of stage I tumours (men: 30.0%; women: 45.5% of those with stage information).
5
We observed significant sex-related differences in survival for all histological groups, with women having survival advantages over men. The lowest survival advantage was observed for women with squamous cell carcinoma and the highest survival advantage for those with adenocarcinoma. The sex-related difference for squamous cell carcinoma was however only marginal. This result is in line with the findings of a Norwegian study [22] in which patients with squamous cell carcinoma showed no sex-related survival difference. The reported survival differences are affected by the time trends of histological groups: In men, the shift from squamous cell carcinoma to adenocarcinoma and in women, the continued increase of adenocarcinoma rates [23]. Although most adenocarcinoma are seen in smokers, they develop more frequently than any other histologic type of lung cancer in individuals who have never smoked, especially in women [24]. Women diagnosed with lung cancer are more likely to be non-smokers than male patients. Hence smoking habits alone are insufficient to explain the differences in histologic distribution and survival. Other genetic and hormonal factors possibly also play a role [25]. Survival differences between countries are partly explained by differences in stage distribution at diagnosis. An international population-based study with data from six national cancer registries showed additionally wide disparities in stage-specific survival [6]. Therefore, other factors, such as differences in treatment seem to play an important role in the prognosis of lung cancer. Like other registry-based studies, no information on applied therapy, comorbidities and further risk factors such as smoking behaviour and performance status were available. More detailed “high resolution studies” including additional information on specific types of treatment and comorbidity are needed to explain differences in survival between populations. The strength of this analysis is the use of a large and representative population-based database. The survival analyses were not limited by selection biases as often encountered in clinical trials which exclude patients with comorbidities and of older age. Variation in the quality of cancer registration may be an additional source of variation of survival rates seen between countries [5,6]. In particular, an incomplete follow-up of patients with under-ascertainment of deaths and insufficient completeness of registration of incident cases (high proportions of cases notified by DCO) could influence the results of survival analyses. In our study, the DCO proportion was 13.5%. The exclusion of these DCO cases could have led to overestimated survival estimates as DCO cases tend to be older and often have very short survival times [26]. However, no relevant differences in 5-year relative survival were found when comparing data from regional cancer registries with a low proportion of DCO cases (Bremen and Saarland) with data from other cancer registries (Bremen 17.1%; Saarland 17.6%; other registries: 15.4–19.4%). In addition, the potential overestimation of 5-year relative survival by exclusion of DCO cases is likely to be limited in our study, as reflected in the narrow plausibility range for the true estimate of 14.6– 16.9% [27]. Furthermore, difficulties in ascertainment of vital status are of minor concern in Germany as most of the cancer registries conduct a comprehensive record linkage with data from residents’ registration offices. In summary, this population-based study from a very large database delivers up-to-date estimates of lung cancer survival in Germany. Women showed higher survival than men independent of age, tumour histology and stage at diagnosis. The reasons for the survival differences between both sexes require further clarification. Although lung cancer survival in Germany appears to be slightly better than in some European countries, overall survival has remained rather poor, as a large number of patients are diagnosed in only late stages when the perspectives for curative treatment are
Please cite this article in press as: A. Eberle, et al., Lung cancer survival in Germany: A population-based analysis of 132,612 lung cancer patients, Lung Cancer (2015), http://dx.doi.org/10.1016/j.lungcan.2015.10.007
G Model LUNG-4959; No. of Pages 6
ARTICLE IN PRESS A. Eberle et al. / Lung Cancer xxx (2015) xxx–xxx
6
rather limited. Recent studies have suggested possible approaches to promote early detection of and to reduce mortality from lung cancer by screening in high risk populations, such as heavy smokers [28]. Efforts to limit the epidemic of smoking, by far the most important risk factor, remain the most important and most effective approaches to limit the burden of lung cancer deaths. Such efforts need to be followed with priority. Sources of support The study was funded by the German Cancer Aid (Deutsche Krebshilfe), grant number 108257 and 110446. Conflict of interest None declared. Acknowledgments We thank the members of the GEKID Cancer Survival Working Group: Karla Geiss, Martin Meyer (Cancer Registry of Bavaria), Andrea Eberle, Sabine Luttmann (Cancer Registry of Bremen), Roland Stabenow (Cancer Registry of Berlin and the New Federal States), Stefan Hentschel, Alice Nennecke (Hamburg Cancer Registry), Joachim Kieschke, Eunice Sirri (Cancer Registry of Lower Saxony), Bernd Holleczek (Saarland Cancer Registry), Katharina Emrich (Cancer Registry of Rhineland-Palatinate), Hiltraud Kajüter, Volkmar Mattauch (Cancer Registry of North RhineWestphalia), Alexander Katalinic, Nora Eisemann (Cancer Registry of Schleswig-Holstein), Klaus Kraywinkel (Robert Koch Institute, Berlin), Hermann Brenner, Felipe Castro, Adam Gondos, Lina Jansen (German Cancer Research Center). References [1] J. Ferlay, I. Soerjomataram., M. Ervik, et al., GLOBOCAN 2012 v1.0, Cancer Incidence and Mortality Worldwide: IARC CancerBase No. 11 [Internet]. Lyon, France: International Agency for Research on Cancer; 2013. Available from: http://globocan.iarc.fr (accessed on 17.02.14). [2] D. Forman, F. Bray, D.H. Brewster, et al. (Eds.). Cancer Incidence in Five Continents, vol. X (electronic version) Lyon, IARC 2013. http://ci5.iarc.fr (last accessed 03.04.14). [3] Robert Koch Institute and Association of Population-based Cancer Registries in Germany, eds. Cancer in Germany 2009/2010. Berlin, 2013. [4] M.G. Kris, S.I. Benowitz, S. Adams, et al., Clinical cancer advances annual report on progress against cancer from the American society of clinical oncology, J. Clin. Oncol. 28 (2010) 5327–5347. [5] R. De Angelis, M. Sant, M.P. Coleman, et al., Cancer survival in Europe 1999–2007 by country and age: results of EUROCARE-5—a population-based study, Lancet Oncol. 15 (2014) 23–34.
[6] S. Walters, C. Maringe, M.P. Coleman, et al., Lung cancer survival and stage at diagnosis in Australia, Canada, Denmark, Norway, Sweden and the UK: a population-based study, 2004–2007, Thorax 68 (2013) 551–564. [7] M.P. Coleman, D. Forman, H. Bryant, et al., Cancer survival in Australia, Canada, Denmark, Norway, Sweden, and the UK 1995-2007 (The International Cancer Benchmarking Partnership): an analysis of population-based cancer registry data, Lancet 377 (2011) 127–138. [8] A. Fritz, C. Percy, A. Jack, et al., WHO International Classification of Disease for Oncology, 3rd ed., World Health Organization, Geneva, 2000. [9] M.P. Curado, B. Edwards, H.R. Shin, et al., Cancer Incidence in Five Continents, IARC Scientific Publications, Lyon: IARC, 2007, vol. IX, No. 160. [10] Union for International Cancer Control (UICC). TNM classification of malignant tumours, 6th ed. Geneva: UICC, 2002. [11] H. Brenner, O. Gefeller, T. Hakulinen, Period analysis for providing more up-to-date cancer survival data: theory, empirical evaluation, computational realization and applications, Eur. J. Cancer 40 (2004) 326–335. [12] F. Ederer, H. Heise, Instructions to IBM 650 Programmers in Processing Survival Computations: Methodological Note No 10, National Cancer Institute, Bethesda, MD, 1959. [13] I. Corazziari, M. Quinn, R. Capocaccia, Standard cancer patient population for age standardizing survival ratios, Eur. J. Cancer 40 (2004) 2307–2316. [14] H. Brenner, T. Hakulinen, Up-to-date and precise estimates of cancer patient survival: model-based period analysis, Am. J. Epidemiol. 164 (2006) 689–696. [15] H. Brenner, O. Gefeller, T. Hakulinen, A computer program for period analysis of cancer patient survival, Eur. J. Cancer 38 (2002) 690–695. [16] C. Aggarwal, C.J. Langer, Older age, poor performance status and major comorbidities: how to treat high-risk patients with advanced nonsmall cell lung cancer, Curr. Opin. Oncol. 24 (2012) 130–136. [17] A. Micheli, R. Ciampichini, W. Oberaigner, et al., The advantage of women in cancer survival: an analysis of EUROCARE-4 data, Eur. J. Cancer 45 (2009) 1017–1027. [18] W. Oberaigner, U. Siebert, Do women with cancer have better survival as compared to men after adjusting for staging distribution? Eur. J. Public Health 21 (2010) 387–391. [19] V. Ambrogi, E. Pompeo, S. Elia, et al., The impact of cardiovascular comorbidity on the outcome of surgery stage I and II non-small-cell lung cancer, Eur. J. Cardiothorac. Surg. 23 (2003) 811–817. [20] D. Salmerón, M.D. Chirlaque, M.I. Izsrzugaza, et al., Lung cancer prognosis in Spain: the role of histology, age and sex, Respir. Med. 106 (2012) 1301–1308. [21] J.C. Yao, M. Hassan, A. Phan, et al., One hundred years after carcinoid: epidemiology of and prognostic factors for neuroendocrine tumors in 35,825 cases in the United States, J. Clin. Oncol. 26 (2008) 3063–3072. [22] C.M.T. Sagerup, M. Småstuen, T.B. Johannesen, et al., Sex-specific trends in lung cancer incidence and survival: a population study of 40118 cases, Thorax 66 (2011) 301–307. [23] J. Lortet-Tieulent, I. Soerjmataram, J. Ferlay, et al., International trends in lung cancer incidence by histological subtype: adenocarcinoma stabilizing in men but still increasing in women, Lung Cancer 84 (2014) 13–22. [24] C.S. Cruz Dela, L.T. Tanoue, R.A. Matthay, Lung cancer: epidemiology, etiology, and prevention, Clin. Chest Med. 32 (2011) 605–644. [25] J.S. Donington, Y.L. Colson, Sex and gender differences in non-small cell lung cancer, Semin. Thorac. Cardiovasc. Surg. 23 (2011) 137–145. [26] B. Holleczek, H. Brenner, Reduction of population-based cancer survival estimates by trace back of death certificate notifications: an empirical illustration, Eur. J. Cancer 48 (2012) 797–804. [27] H. Brenner, B. Holleczek, Deriving valid population-based cancer survival estimates in the presence of nonnegligible proportions of cancers notified by death certificate only, Cancer Epidemiol. Biomarkers Prev. 20 (2011) 2480–2486. [28] J.K. Field, D.M. Hansell, CT screening for lung cancer: countdown to implementation, Lancet Oncol. 14 (2013) 591–600.
Please cite this article in press as: A. Eberle, et al., Lung cancer survival in Germany: A population-based analysis of 132,612 lung cancer patients, Lung Cancer (2015), http://dx.doi.org/10.1016/j.lungcan.2015.10.007
ARTICLE IN PRESS Lung Cancer xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Lung Cancer journal homepage: www.elsevier.com/locate/lungcan
Lung cancer survival in Germany: A population-based analysis of 132,612 lung cancer patients Andrea Eberle a,∗ , Lina Jansen b , Felipe Castro b , Agne Krilaviciute b , Sabine Luttmann a , Katharina Emrich c , Bernd Holleczek d , Alice Nennecke e , Alexander Katalinic f , Hermann Brenner b,g , the GEKID Survival Working Group a
Bremen Cancer Registry, Leibniz-Institute for Prevention Research and Epidemiology—BIPS, Bremen, Germany Division of Clinical Epidemiology and Aging Research, German Cancer Research Center (DKFZ), Heidelberg, Germany c Cancer Registry of Rhineland-Palatinate, Institute for Medical Biostatistics, Epidemiology and Informatics, University Medical Center, Johannes Gutenberg University Mainz, Mainz, Germany d Saarland Cancer Registry, Saarbrücken, Germany e Hamburg Cancer Registry, Authority for Health and Consumer Protection, Hamburg, Germany f Cancer Registry of Schleswig-Holstein, Institute of Cancer Epidemiology, University of Lübeck, Lübeck, Germany g German Cancer Consortium (DKTK), Heidelberg, Germany b
a r t i c l e
i n f o
Article history: Received 18 August 2014 Received in revised form 18 May 2015 Accepted 4 October 2015 Keywords: Lung cancer survival Population-based study Cancer registry Germany Period analysis
a b s t r a c t Objectives: Lung cancer is the most common cancer-related death worldwide. In Germany it accounts for 25% of cancer deaths in men, and 14% in women. The aim of this study is to provide an overview of 5-year relative survival by sex, age, histology, and tumour stage in Germany representing a population of 26.7 million people. Materials and methods: The study is based on a pooled German dataset including data from 12 populationbased cancer registries covering around one third of the German population. A total of 132,612 patients diagnosed with lung cancer from 2002 to 2010 were included in the analysis. Survival estimates for the time period 2007–2010 were calculated using period analysis. Differences in survival between sexes were tested for statistical significance by model-based period analysis (poisson regression model). The relative excess risk (RER) of death (women vs. men) was extracted from the model with the p value for the difference in RER. Results: The overall age adjusted 5-year relative survival was 15.5% (standard error (SE) 0.2) for men and 20.3% (SE 0.3) in women. Survival differed markedly according to age (men: <60 years 18.5% vs. 80+ years 8.4% and women 23.7% vs. 10.6%, respectively), histology (largest difference between histological groups: men 25.7 and women 44.4% points) and stage (men: UICC Ia 62.9%, vs. UICC IV 4.6% and women 75.2% vs. 7.0%, respectively). Our study showed survival advantages for women compared to men, most notably in younger aged patients (RER 0.83, p < 0.0001), patients with adenocarcinoma (RER 0.80, p < 0.0001), and patients with lower stage cancer (RER 0.62, p < 0.0001). Conclusions: This study presents up-to-date survival estimates for lung cancer in Germany. Compared to other European countries survival was relatively high. Women showed higher survival than men independent of age, histology and stage. The reasons for the survival differences require further clarification. © 2015 Elsevier Ireland Ltd. All rights reserved.
1. Introduction
∗ Corresponding author at: Leibniz-Institute for Prevention Research and Epidemiology—BIPS, Department Biometry and Data Management, Unit Statistical Modelling of Secondary and Registry Data, Achterstrasse 30, D-28359 Bremen, Germany. Fax: +49 421 218 56941. E-mail address: [email protected] (A. Eberle).
In Germany as well as in other European industrialised countries, lung cancer is among the three leading forms of cancer both in terms of incidence and mortality. Since the end of the 1990s age-standardised incidence and mortality rates are showing an opposing trend between men and women: Incidence and mortality have decreased steadily for men by around 20% but have increased
http://dx.doi.org/10.1016/j.lungcan.2015.10.007 0169-5002/© 2015 Elsevier Ireland Ltd. All rights reserved.
Please cite this article in press as: A. Eberle, et al., Lung cancer survival in Germany: A population-based analysis of 132,612 lung cancer patients, Lung Cancer (2015), http://dx.doi.org/10.1016/j.lungcan.2015.10.007
G Model LUNG-4959; No. of Pages 6
ARTICLE IN PRESS A. Eberle et al. / Lung Cancer xxx (2015) xxx–xxx
2
in women by approximately 30% [1,2]. This development is caused by the changing behaviour with respect to smoking. Nonetheless, currently there are still two-times more men than women diagnosed with lung cancer in Germany [3]. The prognosis for lung tumours is unfavourable and improvements in survival in recent decades have been minimal. Between the mid-1970s and the beginning of the millennium, 5-year relative survival from lung cancer in the USA increased only modestly from 13% to 16% [4]. For Europe, the EUROCARE-5 study (European Registry Based Study on Survival and Care of Cancer Patients) estimated a mean age-standardised 5-year lung cancer survival of 13% for patients diagnosed in 2000–2007. Relative survival varied nearly twofold between European countries, with the lowest survival being observed in Bulgaria (6.2%), UK with Wales (8.6%) and Scotland (8.7%), and the highest survival in Germany (15.6%) and Austria (16.7%) [5]. Differences in health care systems as well as in the distribution of diagnostic and treatment practices in the whole population are possible explanations for geographical differences in survival [6]. Further explanations are differences in the used data base especially regarding the quality of cancer registration which includes completeness of data and ascertainment of vital status [7]. In this study, using population-based cancer registry data, we analysed lung cancer survival in Germany with respect to sex, age, stage at diagnosis and histology. The aim of this investigation is the identification of factors which affect the relatively high survival rates for lung cancer in Germany. 2. Materials and methods 2.1. Study population The study is based on a pooled national dataset including data from cancer registries in 12 out of 16 federal states in Germany, covering a population of 26.7 million people. Only regions with sufficient data quality were included in the analysis data set. Data quality was assessed by proportion of death certificate only notified and autopsy only (DCO) cases over all malignant cancers. Cancer registries were included if the overall proportion of DCO cases was below 13% in 2002–2010, or decreased steadily over time and was overall below 14%. A cut-off at 13% was chosen, as the same cut-off was used in the European Cancer Registry based study on survival and care of cancer patients (EUROCARE-5 study) [5]. If these criteria were not fulfilled for the whole region covered by a cancer registry, only those administrative districts that fulfilled these criteria were included. The analysis included patients with primary invasive lung cancer (ICD-10: C33-34) diagnosed in 2002–2010 who were at least 15 years old at the time of diagnosis. Patients were followed up with respect to vital status until December 31, 2010. Cases notified by DCO were excluded. Topography and histology of tumours were coded according to ICD-O-3 [8]. The histology codes were grouped into the following six categories: adenocarcinoma, squamous cell carcinoma, large cell carcinoma, small cell carcinoma, other specified tumours and other unspecified tumours [9]. Tumour stages were coded in accordance with the tumour–node–metastasis (TNM) classification of malignant tumour [10]. For the analyses, patients were categorized by sex, age at diagnosis (15–59, 60–69, 70–79 and 80+ years), tumour histology and stage at diagnosis. 2.2. Statistical methods Five-year relative survival estimates for the time period 2007–2010 were calculated using period analysis, providing more
up-to-date survival estimates than traditional cohort based analysis. The period analysis included only survival experience during the period from 2007 to 2010 [11]. Relative survival was derived as a ratio of the observed survival of the included patients divided by the expected survival of the underlying population. Expected survival was computed according to the Ederer II method [12] using national life tables stratified by age, sex, and calendar year obtained from the German Federal Statistical Office. We calculated 5-year relative survival stratified by sex, age group, stage and histological type. Age-standardised survival was estimated using weights defined by the International Cancer Survival Standard [13]. Model-based period analysis was used to investigate differences in survival between men and women and to perform statistical significance testing [14]. Poisson regression models were fit to the number of excess deaths, using year of follow-up and sex as explanatory variables (of categorical type) and included the logarithm of the person-years at risk as offset. The relative excess risk (RER) of death for women compared to men was extracted from the model together with the corresponding p value for the difference in RER between the sexes. Analyses were repeated with additional adjustment for age, histology group, and stage at diagnosis. All analyses were performed with the SAS software (version 9.2), using special macros for period analysis [15] and their adaptation for model-based analysis [14]. Statistical significance was tested two-sided using a significance level of 0.05.
3. Results After exclusion of DCO cases (13.5%) a total of 132,612 lung cancer cases were included in the analysis (Table 1). The percentage of DCO cases varied between registries, ranging from 8.0% in Saarland to 16.4% in Lower Saxony. The proportion of cancers with microscopic confirmation was 93.3%. Nearly three-quarters of lung cancer tumours (72%) were diagnosed in men. The basic characteristics of the patients included in the analyses according to sex, histological group, age, period of diagnosis, and stage are presented in Table 2. Among men, the most common histological category was squamous cell carcinoma (32.0%), with a substantially higher proportion in men than in women (17.1%). Adenocarcinoma was the second most common histological group (27.3%). Among women, the most common histological groups are adenocarcinoma (37.6%), other unspecified tumours (17.5%), and squamous cell carcinoma (17.1%). Large cell carcinoma was the least frequent histological group in both sexes (3.1%). In both sexes, lung cancer patients in the oldest age group (80+ years) had the highest proportion of unspecified tumours (men 32.8%, women 39.2%) whereas patients in the youngest age group (15–69 years) had the highest proportion of adenocarcinoma (men 31.7%, women 43.7%). In both sexes, the number of lung cancer patients rose steadily from 39,207 cases in 2002–2004 to 48,326 cases in 2008–2010. Staging of cancer at the time of diagnosis is the most important predictor of survival, and treatment options are based on stage. In our study, information on stage of tumours at diagnosis was often missing (44%). Because lung cancer is asymptomatic in early stages, only 14% of cases with stage information were classified in stage I. There was no difference in the proportion of tumours with earlier stages (Ia,b) between men and women. In both sexes, small cell carcinoma patients had the highest proportion of stage IV tumours (men: 72.4%; women: 69.1% of those with stage information). The overall age adjusted 5-year relative survival for the time period 2007–2010 was 16.9%, with substantial variation between men (15.5%) and women (20.3%; Table 3). Survival varied by age, with increasing age being associated with reduced relative survival.
Please cite this article in press as: A. Eberle, et al., Lung cancer survival in Germany: A population-based analysis of 132,612 lung cancer patients, Lung Cancer (2015), http://dx.doi.org/10.1016/j.lungcan.2015.10.007
G Model
ARTICLE IN PRESS
LUNG-4959; No. of Pages 6
A. Eberle et al. / Lung Cancer xxx (2015) xxx–xxx
3
Table 1 Description of participating cancer registries and of the dataset used in period survival analysis of lung cancer patients diagnosed between 2002 and 2010 in Germany. Registry
Underlying population 2010 (million)
Brandenburg Bremen Hamburg Lower Saxonya Mecklenburg-Vorpommern North Rhine-Westphaliab Rhineland-Palatinatec Saarland Saxony Saxony-Anhaltd Schleswig-Holsteine Thüringen
a b c d e
2.50 0.66 1.79 7.91 1.64 2.59 0.51 1.02 4.15 0.32 1.38 2.24 26.7
Total
Cases
Exclusion based on death certificate or autopsy only notification (%)
Cases in analysis
Number of deaths (%)
Proportion male %
Median age at diagnosis
Microscopically confirmed cases (%)
14,815 5075 11,339 39,895 10,245 16,489 3081 7632 21,971 2169 8741 11,919
14.0 8.8 14.3 16.4 10.6 12.4 15.0 8.0 12.5 11.0 16.2 12.4
12,741 4630 9721 33,348 9157 14,450 2618 7025 19,224 1930 7323 10,445
9710 (76.2%) 3600 (77.8%) 7590 (78.1%) 25,676 (77.0%) 7226 (78.9%) 11,126 (77.0%) 1999 (76.4%) 5493 (78.2%) 14,846 (77.2%) 1369 (70.9%) 5634 (76.9%) 8017 (76.8%)
75.4 66.6 63.5 70.7 75.2 70.8 72.1 70.7 77.3 71.6 67.3 76.2
68 68 67 68 68 68 68 68 69 68 67 68
94.2 92.2 95.7 89.5 92.2 97.1 98.3 94.9 91.5 96.5 96.5 95.0
153,371
13.5
132,612
102,286 (77.1%)
72.0
68
93.3
Diagnosis period included 2003–2010. Selected administrative districts: Regierungsbezirk Münster. Selected administrative district: Regierungsbezirk Trier. Selected administrative district: Cities Halle and Dessau-Roßlau. Selected administrative districts: Cities Flensburg, Kiel and Neumünster and districts Nordfriesland, Ostholstein, Plön, Rendsburg-Eckernförde, Schleswig-Flensburg.
Table 2 Distribution of histology by age, period of diagnosis and stage for male and female lung cancer patients, A: male, B: female. A Male
Squamous cell n (%)
Large cell n (%)
6284 (24.1) 9408 (36.0) 8422 (32.3) 1994 (7.6)
5575 (18.2) 11,016 (36.0) 11,450 (37.4) 2570 (8.4)
679 (23.0) 1079 (36.5) 966 (32.7) 235 (7.9)
4153 (24.2) 6340 (37.0) 5372 (31.3) 1285 (7.5)
545 (25.1) 767 (35.3) 724 (33.3) 138 (6.4)
Period of diagnosis 7170 (27.5) 2002–04 8953 (34.3) 2005–07 2008–10 9985 (38.2)
9551 (31.2) 10,879 (35.5) 10,181 (33.3)
1062 (35.9) 920 (31.1) 977 (33.0)
5444 (31.7) 6011 (35.0) 5695 (33.2)
Stage UICC Ia UICC Ib UICC IIa UICC IIb UICC IIIa UICC IIIb UICC IV Missing All cases
1138 (3.7) 2162 (7.1) 232 (0.8) 1619 (5.3) 2499 (8.2) 3926 (12.8) 6869 (22.4) 12,166 (39.7) 30,611 (100)
91 (3.1) 176 (6.0) 17 (0.6) 101 (3.4) 179 (6.1) 299 (10.1) 1066 (36.0) 1030 (34.8) 2959 (100)
137 (0.8) 197 (1.2) 58 (0.3) 165 (1.0) 599 (3.5) 1519 (8.9) 7009 (40.9) 7466 (43.5) 17,150 (100)
Age 15–59 60–69 70–79 80+
Adeno-carcinoma n (%)
1224 (4.7) 1316 (5.0) 180 (0.7) 673 (2.6) 1342 (5.1) 1974 (7.6) 9039 (34.6) 10,360 (39.7) 26,108 (100)
Small cell n (%)
Other specified n (%)
Other unspecified n (%)
Missing n (%)
All cases n (%)
2612 (15.9) 4892 (29.7) 5903 (35.9) 3043 (18.5)
7 (9.1) 15 (19.5) 33 (42.9) 22 (29.3)
19,855 (20.8) 33,517 (35.1) 32,870 (34.4) 9287 (9.7)
553 (25.4) 721 (33.2) 900 (41.4)
5385 (32.7) 5718 (34.8) 5347 (32.5)
17 (22.1) 25 (32.5) 35 (45.5)
29,182 (30.6) 33,227 (34.8) 34,020 (35.6)
198 (9.1) 162 (7.5) 29 (1.3) 99 (4.6) 137 (6.3) 132 (6.1) 444 (20.4) 973 (44.8) 2174 (100)
221 (1.3) 310 (1.9) 36 (0.2) 220 (1.3) 475 (2.9) 847 (5.2) 5023 (30.5) 9318 (56.6) 16,450 (100)
0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 1 (1.3) 1 (1.3) 75 (97.4) 77 (100)
3009 (3.2) 4323 (4.5) 552 (0.6) 2877 (3.0) 5231 (5.5) 8698 (9.1) 29,451 (30.8) 41,388 (43.3) 95,529 (100)
B Female
Age 15–59 60–69 70–79 80+
Adeno-carcinoma
Squamous cell
Large cell
Small cell
Other specified
Other unspecified
Missing
All cases
n (%)
n (%)
n (%)
n (%)
n (%)
n (%)
n (%)
n (%)
4412 (31.6) 4198 (30.1) 3825 (27.4) 1509 (10.8)
1371 (21.6) 2051 (32.4) 2179 (34.4) 737 (11.6)
357 (32.2) 347 (31.4) 300 (27.1) 103 (9.3)
2430 (31.4) 2623 (33.9) 2078 (26.9) 609 (7.9)
511 (35.5) 452 (31.4) 366 (25.4) 111 (7.7)
1120 (17.3) 1461 (22.5) 1918 (29.6) 1986 (30.6)
2 (6.9) 2 (6.9) 11 (37.9) 14 (48.3)
10,203 (27.5) 11,134 (30.0) 10,677 (28.8) 5069 (13.7)
Period of diagnosis 3441 (24.7) 2002–04 4752 (34.1) 2005–07 5751 (41.2) 2008–10
1742 (27.5) 2283 (36.0) 2313 (36.5)
380 (34.3) 349 (31.5) 378 (34.1)
2193 (28.3) 2634 (34.0) 2913 (37.6)
354 (24.6) 502 (34.9) 584 (40.5)
1913 (29.5) 2220 (34.2) 2352 (36.3)
2 (6.9) 12(41.4) 15 (51.7)
10,025 (27.0) 12,752 (34.4) 14,306 (38.6)
Stage UICC Ia UICC Ib UICC IIa UICC IIb UICC IIIa UICC IIIb UICC IV Missing All cases
263 (4.2) 383 (6.0) 50 (0.8) 276 (4.4) 441 (7.0) 683 (10.8) 1541 (24.3) 2701 (42.6) 6338 (100)
31 (2.8) 47 (4.3) 3 (0.3) 32 (2.9) 56 (5.1) 109 (9.9) 415 (37.5) 414 (37.4) 1107 (100)
86 (1.1) 90 (1.2) 37 (0.5) 71 (0.9) 263 (3.4) 721 (9.3) 2834 (36.6) 3638 (47.0) 7740 (100)
212 (14.7) 111 (7.7) 15 (1.0) 37 (2.6) 64 (4.4) 63 (4.4) 208 (14.4) 730 (50.7) 1440 (100)
93 (1.4) 121 (1.9) 10 (0.2) 64 (1.0) 131 (2.0) 268 (4.1) 1972 (30.4) 3826 (59.0) 6485 (100)
29 (100) 29 (100)
1490 (4.0) 1455 (3.9) 231 (0.6) 795 (2.1) 1664 (4.5) 2814 (7.6) 11,661 (31.5) 16,973 (45.8) 37,083 (100)
805 (5.8) 703 (5.0) 116 (0.8) 315 (2.3) 709 (5.1) 970 (7.0) 4691 (33.6) 5635 (40.4) 13,944 (100)
Please cite this article in press as: A. Eberle, et al., Lung cancer survival in Germany: A population-based analysis of 132,612 lung cancer patients, Lung Cancer (2015), http://dx.doi.org/10.1016/j.lungcan.2015.10.007
G Model
ARTICLE IN PRESS
LUNG-4959; No. of Pages 6
A. Eberle et al. / Lung Cancer xxx (2015) xxx–xxx
4
Table 3 Age group-specific and age-standardised 5-year relative survival of lung cancer patients during the period 2007-2010 in Germany. Sex differences are presented as estimates of relative excess risk (RER) of death for women compared to men. Age group
Sex differences by age Male
Female RS, %
N
15–59 60–69 70–79 80+ Overall age adjusted
19,855 33,517 32,870 9287 95,529
18.5 17.0 13.8 8.4 15.5
SE
Adjusted for histology and stage
N
0.4 0.3 0.3 0.6 0.2
RS, %
10,203 11,134 10,677 5069 37,083
SE
23.7 22.9 19.0 10.6 20.3
RER
0.6 0.6 0.6 0.8 0.3
p value
0.86 0.83 0.86 0.98 0.87
<.0001 <.0001 <.0001 0.4383 <.0001
RER
0.68 0.68 0.74 0.78 0.81
p value
<.0001 <.0001 <.0001 <.0001 <.0001
Adj. RS, % Male
Female
19.3 18.6 16.8 12.9 17.5
23.0 23.1 19.7 14.7 21.0
RS: relative survival, SE: standard error of the estimate, RER: relative excess risk of death for women vs. men. Table 4 Five-year relative survival of lung cancer patients during the period 2007-2010 in Germany by histological groups and sex. Sex differences are presented as estimates of relative excess risk (RER) of death for women compared to men. Histological group
Sex differences by histology Male
Female
N
Adenocarcinoma Squamous Cell Large Cell Small Cell Other specified Other unspecified
RS, %
26,108 30,611 2959 17,150 2174 16,450
18.2 19.8 17.0 7.3 34.1 8.4
SE
Adjusted for age and stage
N
0.4 0.4 1.3 0.4 1.7 0.4
RS, %
13,944 6338 1107 7740 1440 6485
SE
24.8 22.7 21.3 10.7 54.1 9.7
0.6 0.9 2.1 0.6 2.2 0.6
RER
p value
0.80 0.95 0.90 0.86 0.54 0.94
<.0001 0.0001 0.0008 <.0001 <.0001 <.0001
RER
0.75 0.94 0.75 0.85 0.64 0.87
p value
<.0001 0.0361 0.0002 <.0001 <.0001 <.0001
Adj. RS, % Male
Female
20.0 22.2 16.7 8.3 38.0 9.6
25.4 23.0 21.8 10.8 45.7 11.5
RS: relative survival, SE: standard error of the estimate, RER: relative excess risk of death for women vs. men. Table 5 Five-year relative survival of lung cancer patients in the period 2007-2010 in Germany by stage and sex. Sex differences are presented as estimates of relative excess risk of death for women compared to men. Stage
Sex differences by stage Male N
UICC Ia UICC Ib UICC IIa UICC IIb UICC IIIa UICC IIIb UICC IV Missing
3009 4323 552 2877 5231 8698 29,451 41,388
Female RS, %
SE
62.9 53.4 39.7 35.6 20.3 12.5 4.6 14.9
1.8 1.5 3.8 1.6 1.1 0.6 0.2 0.3
N
1490 1455 231 795 1664 2814 11,661 16,973
Adjusted for age and histology RS, %
SE
RER
p value
RER
p value
75.2 64.3 53.9 41.5 28.0 15.7 7.0 19.9
2.2 2.1 7.5 2.8 1.8 1.1 0.4 0.5
0.62 0.68 0.66 0.91 0.77 0.89 0.86 0.87
<.0001 <.0001 <.0001 0.0157 <.0001 <.0001 <.0001 <.0001
0.50 0.51 0.56 0.71 0.68 0.82 0.81 0.78
<.0001 <.0001 0.0021 <.0001 <.0001 <.0001 <.0001 <.0001
Adj. RS, % Male
Female
64.9 53.1 39.2 36.2 20.1 13.0 5.5 22.7
73.9 64.2 54.1 40.1 28.5 15.6 8.0 27.0
RS: relative survival, SE: standard error of the estimate, RER: relative excess risk of death for women vs. men.
Across all age groups, 5-year relative survival was higher in women than in men even after adjustment for histology and stage. Survival varied by histology (Table 4). In both sexes, patients with small cell carcinoma had the worst survival (men 7.3% and women 10.7%). Significant sex-related differences were observed for all histological groups. Results of the Poisson regression analysis showed that especially women with adenocarcinoma or “other specified tumours” had the lowest relative excess risk of dying (RER 0.80 and 0.54, respectively) compared to men. The survival advantages of women with squamous cell carcinoma or small cell carcinoma were practically unchanged after the adjustment for age and stage. For patients with other specified tumours, 5-year relative survival was much higher than for all other histological groups in both men (34.1%) and women (54.1%). The group ‘other specified tumours’ particularly included adenosquamous carcinoma (46.8%) and carcinoid tumours (32.8%). Table 5 shows 5-year relative survival by stage and sex. Survival differed strongly according to stage at diagnosis. In stage Ia,
5-year survival was above 60%; however, 5-year relative survival decreased to below 5% in men and 7% in women when patients presented with metastatic disease at diagnosis (stage IV). A substantial survival advantage of women was observed among lung cancer patients diagnosed in stage Ia, Ib and IIa. The risk of dying within five years past diagnosis was significantly lower in women than in men among all stages even after adjustment for age and histology. The sex difference was most evident in patients with earlier stages (UICC Ia, Ib and IIa), where women had an approximately 50% decreased risk of dying.
4. Discussion The present analysis of lung cancer survival was based on a very large population-based cancer database from Germany, which enabled us to present a comprehensive overview of 5-year survival of lung cancer patients. The overall age-adjusted 5-year survival
Please cite this article in press as: A. Eberle, et al., Lung cancer survival in Germany: A population-based analysis of 132,612 lung cancer patients, Lung Cancer (2015), http://dx.doi.org/10.1016/j.lungcan.2015.10.007
G Model LUNG-4959; No. of Pages 6
ARTICLE IN PRESS A. Eberle et al. / Lung Cancer xxx (2015) xxx–xxx
was 16.9% in 2007–2010. Relative survival estimates were significantly higher in women than in men, even after adjustment for histology and stage. Survival differed markedly according to age, histological group and stage at diagnosis. Our calculated overall 5-year survival was only slightly higher than the German lung cancer survival estimate reported in the EUROCARE-5 study for the earlier period 2000–2007 (15.6%) [5]. We observed a strong age gradient in both sexes. Increasing age is generally considered to be an indicator for poor prognosis and we found, as expected, that survival decreased rapidly with increasing age in Germany. Poorer survival in the elderly population has been ascribed to higher prevalence of co-morbidities. These are known to be a major reason why patients and especially older patients do not undergo surgery or other effective treatments [16]. Furthermore, the prognosis strongly varied by sex, histological group and stage. In our study, the prognosis of men was considerably worse than that of women. These results are in agreement with findings from the EUROCARE project, which reported a better survival of women under the age of 65 years compared to men [17]. However, sex differences in the current study were even more pronounced. Our study showed an age-adjusted survival difference of 4.8 percent units between men and women, compared to 2.4 percent units in the EUROCARE-4 database for patients diagnosed in 1995–1999 with smallest difference in UK and Ireland (1.2 percent units). The female survival advantage was most marked in middle age groups. In recent European studies, a female survival disadvantage with older age group was reported [17,18]. Micheli et al. suggested that age at diagnosis might be a proxy for biological factors which changed more markedly in women than in men with increasing age [17]. Variation in tumour cell biology such as expression and mutation rates of multiple regulating gene pathways, differences in DNA repair capacity, and changes in reproductive hormones are possible explanations for the differences between men and women. Furthermore, differences in prevalence of comorbidities especially cardiovascular diseases could have therapeutic and prognostic consequences. In particular, cardiovascular disease seems to influence the survival of lung cancer patients [19]. In order to further elucidate potential underlying reasons for differences in survival, patients should ideally not only be classified by sex, age, histology and stage, but also by the presence of prognostically relevant comorbidity. Additionally, survival estimates differ according to the histological group of the tumours. Further, the distribution of the histological groups differs significantly between men and women. In some European countries, e.g. Italy, Austria, and the Scandinavian countries, the incidence of adenocarcinoma in men has exceeded the incidence of squamous cell carcinoma, which has so far been the most common histological subtype [2]. In women, however, adenocarcinoma is the most common histological group in almost all European countries except Poland and Bulgaria [2]. In our study, patients with adenocarcinoma showed relatively high survival in both sexes. It is noticeable that small cell carcinoma belongs to the most frequent histological groups in Scotland [2]. This could be an explanation for the remarkably low lung cancer survival in this country. In our study, 5-year relative survival was lowest in patients diagnosed with small cell lung cancer. The relatively high survival estimates observed for ‘other specified tumours’ are in accordance with a previous population-based study conducted in Spain [20]. An explanation for this finding is the high proportion of less aggressive tumours in this category. In our study, approximately one third of lung cancers in this heterogeneous group were neuroendocrine (carcinoid) tumours which tend to grow slowly and only rarely spreads beyond the lungs [21]. In our study, other specified tumours had the highest proportion of stage I tumours (men: 30.0%; women: 45.5% of those with stage information).
5
We observed significant sex-related differences in survival for all histological groups, with women having survival advantages over men. The lowest survival advantage was observed for women with squamous cell carcinoma and the highest survival advantage for those with adenocarcinoma. The sex-related difference for squamous cell carcinoma was however only marginal. This result is in line with the findings of a Norwegian study [22] in which patients with squamous cell carcinoma showed no sex-related survival difference. The reported survival differences are affected by the time trends of histological groups: In men, the shift from squamous cell carcinoma to adenocarcinoma and in women, the continued increase of adenocarcinoma rates [23]. Although most adenocarcinoma are seen in smokers, they develop more frequently than any other histologic type of lung cancer in individuals who have never smoked, especially in women [24]. Women diagnosed with lung cancer are more likely to be non-smokers than male patients. Hence smoking habits alone are insufficient to explain the differences in histologic distribution and survival. Other genetic and hormonal factors possibly also play a role [25]. Survival differences between countries are partly explained by differences in stage distribution at diagnosis. An international population-based study with data from six national cancer registries showed additionally wide disparities in stage-specific survival [6]. Therefore, other factors, such as differences in treatment seem to play an important role in the prognosis of lung cancer. Like other registry-based studies, no information on applied therapy, comorbidities and further risk factors such as smoking behaviour and performance status were available. More detailed “high resolution studies” including additional information on specific types of treatment and comorbidity are needed to explain differences in survival between populations. The strength of this analysis is the use of a large and representative population-based database. The survival analyses were not limited by selection biases as often encountered in clinical trials which exclude patients with comorbidities and of older age. Variation in the quality of cancer registration may be an additional source of variation of survival rates seen between countries [5,6]. In particular, an incomplete follow-up of patients with under-ascertainment of deaths and insufficient completeness of registration of incident cases (high proportions of cases notified by DCO) could influence the results of survival analyses. In our study, the DCO proportion was 13.5%. The exclusion of these DCO cases could have led to overestimated survival estimates as DCO cases tend to be older and often have very short survival times [26]. However, no relevant differences in 5-year relative survival were found when comparing data from regional cancer registries with a low proportion of DCO cases (Bremen and Saarland) with data from other cancer registries (Bremen 17.1%; Saarland 17.6%; other registries: 15.4–19.4%). In addition, the potential overestimation of 5-year relative survival by exclusion of DCO cases is likely to be limited in our study, as reflected in the narrow plausibility range for the true estimate of 14.6– 16.9% [27]. Furthermore, difficulties in ascertainment of vital status are of minor concern in Germany as most of the cancer registries conduct a comprehensive record linkage with data from residents’ registration offices. In summary, this population-based study from a very large database delivers up-to-date estimates of lung cancer survival in Germany. Women showed higher survival than men independent of age, tumour histology and stage at diagnosis. The reasons for the survival differences between both sexes require further clarification. Although lung cancer survival in Germany appears to be slightly better than in some European countries, overall survival has remained rather poor, as a large number of patients are diagnosed in only late stages when the perspectives for curative treatment are
Please cite this article in press as: A. Eberle, et al., Lung cancer survival in Germany: A population-based analysis of 132,612 lung cancer patients, Lung Cancer (2015), http://dx.doi.org/10.1016/j.lungcan.2015.10.007
G Model LUNG-4959; No. of Pages 6
ARTICLE IN PRESS A. Eberle et al. / Lung Cancer xxx (2015) xxx–xxx
6
rather limited. Recent studies have suggested possible approaches to promote early detection of and to reduce mortality from lung cancer by screening in high risk populations, such as heavy smokers [28]. Efforts to limit the epidemic of smoking, by far the most important risk factor, remain the most important and most effective approaches to limit the burden of lung cancer deaths. Such efforts need to be followed with priority. Sources of support The study was funded by the German Cancer Aid (Deutsche Krebshilfe), grant number 108257 and 110446. Conflict of interest None declared. Acknowledgments We thank the members of the GEKID Cancer Survival Working Group: Karla Geiss, Martin Meyer (Cancer Registry of Bavaria), Andrea Eberle, Sabine Luttmann (Cancer Registry of Bremen), Roland Stabenow (Cancer Registry of Berlin and the New Federal States), Stefan Hentschel, Alice Nennecke (Hamburg Cancer Registry), Joachim Kieschke, Eunice Sirri (Cancer Registry of Lower Saxony), Bernd Holleczek (Saarland Cancer Registry), Katharina Emrich (Cancer Registry of Rhineland-Palatinate), Hiltraud Kajüter, Volkmar Mattauch (Cancer Registry of North RhineWestphalia), Alexander Katalinic, Nora Eisemann (Cancer Registry of Schleswig-Holstein), Klaus Kraywinkel (Robert Koch Institute, Berlin), Hermann Brenner, Felipe Castro, Adam Gondos, Lina Jansen (German Cancer Research Center). References [1] J. Ferlay, I. Soerjomataram., M. Ervik, et al., GLOBOCAN 2012 v1.0, Cancer Incidence and Mortality Worldwide: IARC CancerBase No. 11 [Internet]. Lyon, France: International Agency for Research on Cancer; 2013. Available from: http://globocan.iarc.fr (accessed on 17.02.14). [2] D. Forman, F. Bray, D.H. Brewster, et al. (Eds.). Cancer Incidence in Five Continents, vol. X (electronic version) Lyon, IARC 2013. http://ci5.iarc.fr (last accessed 03.04.14). [3] Robert Koch Institute and Association of Population-based Cancer Registries in Germany, eds. Cancer in Germany 2009/2010. Berlin, 2013. [4] M.G. Kris, S.I. Benowitz, S. Adams, et al., Clinical cancer advances annual report on progress against cancer from the American society of clinical oncology, J. Clin. Oncol. 28 (2010) 5327–5347. [5] R. De Angelis, M. Sant, M.P. Coleman, et al., Cancer survival in Europe 1999–2007 by country and age: results of EUROCARE-5—a population-based study, Lancet Oncol. 15 (2014) 23–34.
[6] S. Walters, C. Maringe, M.P. Coleman, et al., Lung cancer survival and stage at diagnosis in Australia, Canada, Denmark, Norway, Sweden and the UK: a population-based study, 2004–2007, Thorax 68 (2013) 551–564. [7] M.P. Coleman, D. Forman, H. Bryant, et al., Cancer survival in Australia, Canada, Denmark, Norway, Sweden, and the UK 1995-2007 (The International Cancer Benchmarking Partnership): an analysis of population-based cancer registry data, Lancet 377 (2011) 127–138. [8] A. Fritz, C. Percy, A. Jack, et al., WHO International Classification of Disease for Oncology, 3rd ed., World Health Organization, Geneva, 2000. [9] M.P. Curado, B. Edwards, H.R. Shin, et al., Cancer Incidence in Five Continents, IARC Scientific Publications, Lyon: IARC, 2007, vol. IX, No. 160. [10] Union for International Cancer Control (UICC). TNM classification of malignant tumours, 6th ed. Geneva: UICC, 2002. [11] H. Brenner, O. Gefeller, T. Hakulinen, Period analysis for providing more up-to-date cancer survival data: theory, empirical evaluation, computational realization and applications, Eur. J. Cancer 40 (2004) 326–335. [12] F. Ederer, H. Heise, Instructions to IBM 650 Programmers in Processing Survival Computations: Methodological Note No 10, National Cancer Institute, Bethesda, MD, 1959. [13] I. Corazziari, M. Quinn, R. Capocaccia, Standard cancer patient population for age standardizing survival ratios, Eur. J. Cancer 40 (2004) 2307–2316. [14] H. Brenner, T. Hakulinen, Up-to-date and precise estimates of cancer patient survival: model-based period analysis, Am. J. Epidemiol. 164 (2006) 689–696. [15] H. Brenner, O. Gefeller, T. Hakulinen, A computer program for period analysis of cancer patient survival, Eur. J. Cancer 38 (2002) 690–695. [16] C. Aggarwal, C.J. Langer, Older age, poor performance status and major comorbidities: how to treat high-risk patients with advanced nonsmall cell lung cancer, Curr. Opin. Oncol. 24 (2012) 130–136. [17] A. Micheli, R. Ciampichini, W. Oberaigner, et al., The advantage of women in cancer survival: an analysis of EUROCARE-4 data, Eur. J. Cancer 45 (2009) 1017–1027. [18] W. Oberaigner, U. Siebert, Do women with cancer have better survival as compared to men after adjusting for staging distribution? Eur. J. Public Health 21 (2010) 387–391. [19] V. Ambrogi, E. Pompeo, S. Elia, et al., The impact of cardiovascular comorbidity on the outcome of surgery stage I and II non-small-cell lung cancer, Eur. J. Cardiothorac. Surg. 23 (2003) 811–817. [20] D. Salmerón, M.D. Chirlaque, M.I. Izsrzugaza, et al., Lung cancer prognosis in Spain: the role of histology, age and sex, Respir. Med. 106 (2012) 1301–1308. [21] J.C. Yao, M. Hassan, A. Phan, et al., One hundred years after carcinoid: epidemiology of and prognostic factors for neuroendocrine tumors in 35,825 cases in the United States, J. Clin. Oncol. 26 (2008) 3063–3072. [22] C.M.T. Sagerup, M. Småstuen, T.B. Johannesen, et al., Sex-specific trends in lung cancer incidence and survival: a population study of 40118 cases, Thorax 66 (2011) 301–307. [23] J. Lortet-Tieulent, I. Soerjmataram, J. Ferlay, et al., International trends in lung cancer incidence by histological subtype: adenocarcinoma stabilizing in men but still increasing in women, Lung Cancer 84 (2014) 13–22. [24] C.S. Cruz Dela, L.T. Tanoue, R.A. Matthay, Lung cancer: epidemiology, etiology, and prevention, Clin. Chest Med. 32 (2011) 605–644. [25] J.S. Donington, Y.L. Colson, Sex and gender differences in non-small cell lung cancer, Semin. Thorac. Cardiovasc. Surg. 23 (2011) 137–145. [26] B. Holleczek, H. Brenner, Reduction of population-based cancer survival estimates by trace back of death certificate notifications: an empirical illustration, Eur. J. Cancer 48 (2012) 797–804. [27] H. Brenner, B. Holleczek, Deriving valid population-based cancer survival estimates in the presence of nonnegligible proportions of cancers notified by death certificate only, Cancer Epidemiol. Biomarkers Prev. 20 (2011) 2480–2486. [28] J.K. Field, D.M. Hansell, CT screening for lung cancer: countdown to implementation, Lancet Oncol. 14 (2013) 591–600.
Please cite this article in press as: A. Eberle, et al., Lung cancer survival in Germany: A population-based analysis of 132,612 lung cancer patients, Lung Cancer (2015), http://dx.doi.org/10.1016/j.lungcan.2015.10.007