Abstract
Background
Gastric cancer accounts for 5 % of cancer deaths. Successful implementation of guideline-recommended treatment procedures should result in population-based outcome improvements despite the still poor prognosis. In this context, the objective of this study was to compare the outcome of gastric cancer by different levels of hospital care.
Materials and methods
Total of 8,601 patients with invasive gastric cancer documented between 1998 and 2012 by the Munich Cancer Registry were evaluated. Tumour and therapy characteristics and outcome were analysed in regard to five levels of hospital care: three levels were defined for general hospitals (level I–III), while university hospitals and speciality hospitals were grouped as separate classes. Survival was investigated using the Kaplan–Meier-method, computing relative survival, and by multivariate Cox proportional hazard regression.
Results
The average age differed between 66 years in university hospitals and 75 years in hospitals providing a basic level of care (level I). No survival differences were found for patients treated in different levels of hospital care in 75 % of the patient cohort, namely the M0 patients. A better survival could only be shown for patients with M1 at diagnosis when treated in a university or level III hospital compared to those treated in other hospitals.
Conclusion
The outcome difference of M1 patients is most likely caused by selection effects concerning health status differences and not by processes of health care attributable to level of hospital care. Thus, this study demonstrates and confirms appropriate treatment and care of gastric cancer over all levels of hospital care.
Keywords: Gastric cancer, Stomach cancer, Cancer registry, Comparative effectiveness research, Health care, Level of hospital care
Introduction
Although gastric cancer incidence and mortality in Germany are decreasing over the past three decades, it is still accountable for 3 and 3.7 % of new cancer cases and for 4.6 and 5.1 % of cancer deaths in women and men, respectively (Robert-Koch Institute, Association of Population-Based Cancer Registries 2007). Because of its peak incidence in the seventh decade in men and in the eighth and ninth decade in women (Munich Cancer Registry 2013), the proportion of older patients with gastric cancer will increase with the expanding older population in industrial countries (Nashimoto 2013; Endo et al. 2013).
While early gastric cancer yields an excellent prognosis after curative treatment, survival of advanced gastric cancer remains dismal despite the introduction of perioperative multimodal treatment as standard of care (Mihaljevic et al. 2013). With the knowledge of the poor prognosis of gastric cancer, the National German Guideline ‘Diagnosis and Treatment of Oesophagogastric Cancer’ was published in 2011 (Moehler et al. 2011). The successful implementation of recommended treatment procedures with high levels of evidence should result in adequate quality health care and should improve population-based outcomes, e.g. survival.
In addition, in Germany, the federal states are legally obligated by §6 of the Hospital Financing Act (KHG) to establish regional hospital plans (Krankenhausfinanzierungsgesetz 1991). This should ensure sufficient supply of hospital care and thus health care in adequate quality and economic efficiency. In Bavaria, the State Ministry of the Environment and Public Health is responsible for this plan (Bayerisches Staatsministerium für Umwelt und Gesundheit 2013). It classifies general hospitals into three levels of care: Level I hospitals have to provide the basic inpatient services in midsize towns. Level II hospitals should ensure hospital care with supra-regional supply in diagnostics and therapies. Level III hospitals provide the maximum range of medical knowledge and technical infrastructure in diagnostics and treatment of nearly all diseases. University hospitals count as level III hospitals. A fourth category comprises speciality hospitals which are concerned with limited groups of patients or diseases, e.g. rehabilitation after cancer or cardiac surgery, conservative treatment only or therapy of rheumatic diseases. Gastric cancer is diagnosed and treated in this broad health care structure.
Comparative effectiveness research (CER) has become increasingly important in oncology, because rapid advances in molecular biology and genetics have led to the identification of a virtually limitless number of potential strategic targets for improving the diagnosis and treatment of cancer while rapidly rising health care costs are threatening the economic stability (Lyman 2009). CER has led to a renewed interest in observational studies, which often include large numbers of patients who are more representative of the general population than those in randomised clinical trials (Armstrong 2012). Observational studies can inform medical decision-making and health policy, provided that their opportunities, strengths and limitations are considered (Hershman and Wright 2012).
Against this background, the data of population-based clinical cancer registries such as the Munich Cancer Registry (MCR) allow the investigation of gastric cancer patients’ outcome in routine care including the adjustment for the most important patient and tumour characteristics.
The objective of this study was to compare gastric cancer patients and their survival by the level of care of all hospitals which conduct the diagnosis and treatment at any stage, early, localised or advanced. We specifically asked whether the level of hospital care has an impact on patient outcome in invasive gastric cancer.
Materials and methods
Data collection
The Munich Cancer Registry (MCR) is the population-based clinical cancer registry of Upper Bavaria and parts of Lower Bavaria (Southern Germany). Its catchment area has increased from 2.3 million inhabitants to 3.8 million in 2002 and to 4.5 million in 2007. Pathology reports of solid tumours from all pathology laboratories in this catchment area are available. From these reports, the total number of gastric cancer patients in the region is systematically recorded and the main prognostic factors are ascertained. In parallel, clinicians complete standardised forms concerning patients’ domicile, age, primary disease characteristics such as TNM stage, histology, grade, as well as therapies. Additionally, the life status of patients with a cancer diagnosis is maintained systematically through death certificates. Thus, active follow-up data are available for about 95 % of all cases.
Patient sample
A total of 11,124 patients with residence in the catchment area were diagnosed with a malignant gastric tumour over the 15-year study period from 1998 to 2012 (Fig. 1). Patients with non-invasive carcinoma, neuroendocrine tumour, sarcoma or lymphoma were excluded, as well as cases with death certificates only (DCO, 10.2 %). The analyses of the epidemiological cohort of 8,601 patients provide a current and population-based survey of invasive gastric cancer.
Fig. 1.
Study flow chart
Patients with evidence of another previous or synchronous malignant tumour were excluded from survival analyses to eliminate overlapping tumour effects. Thus, the cohort for all survival analyses comprises 7,196 patients.
Table 1 shows the distribution of cases by years of initial diagnosis: increasing numbers can be seen in 2001/2002 and 2007/2008 when the catchment area was expanded. The numbers of gastric cancers as first non-synchronous malignant tumours are presented as well, because they provide the cohort for survival analyses. 84 % of gastric cancers are the patients’ first malignant tumour, 16 % of patients have had other malignomas before.
Table 1.
Numbers of total epidemiologic cohort and first malignant tumour by year (epidemiologic cohort 1998–2012, n = 8,601)
| Years | University hospital n = 1,990 n (%) |
Level III hospital n = 1,710 n (%) |
Level II hospital n = 2,032 n (%) |
Level I hospital n = 2,281 n (%) |
Speciality hospital n = 343 n (%) |
Hospital n.a. n = 245 n (%) |
Total n = 8,601 n (%) |
Total cumulative n |
First tumour non-synchronous |
|
|---|---|---|---|---|---|---|---|---|---|---|
| n (%) | Total cumulative n |
|||||||||
| 1998 | 104 (21.5) | 151 (31.3) | 44 (9.1) | 128 (26.5) | 41 (8.5) | 15 (3.1) | 483 (100) | 483 | 423 (87.6) | 423 |
| 1999/2000 | 235 (28.0) | 248 (29.6) | 107 (12.8) | 182 (21.7) | 50 (6.0) | 17 (2.0) | 839 (100) | 1,322 | 714 (85.1) | 1,137 |
| 2001/2002 | 289 (26.4) | 240 (22.0) | 203 (18.6) | 289 (26.4) | 41 (3.8) | 31 (2.8) | 1,093 (100) | 2,415 | 957 (87.6) | 2,094 |
| 2003/2004 | 345 (26.2) | 215 (16.3) | 343 (26.0) | 343 (26.0) | 40 (3.0) | 32 (2.4) | 1,318 (100) | 3,733 | 1,108 (84.1) | 3,202 |
| 2005/2006 | 331 (25.9) | 205 (15.9) | 334 (25.8) | 339 (26.2) | 46 (3.6) | 38 (2.9) | 1,293 (100) | 5,026 | 1,075 (83.1) | 4,277 |
| 2007/2008 | 301 (20.2) | 238 (15.9) | 408 (27.3) | 450 (30.1) | 47 (3.2) | 49 (3.3) | 1,493 (100) | 6,519 | 1,223 (81.9) | 5,500 |
| 2009/2010 | 252 (19.5) | 228 (17.7) | 353 (27.3) | 359 (27.8) | 49 (3.8) | 51 (4.0) | 1,292 (100) | 6,519 | 1,044 (80.8) | 6,544 |
| 2011/2012 | 133 (16.8) | 185 (23.4) | 240 (30.4) | 191 (24.2) | 29 (3.7) | 12 (1.5) | 790 (100) | 7,811 | 652 (82.5) | 7,196 |
| Total | 1,990 (23.1) | 1,710 (19.9) | 2,032 (23.6) | 2,281 (26.5) | 343 (4.0) | 245 (2.9) | 8,601 (100) | 8,601 | 7,196 (83.6) | 7,196 |
n.a. not available
Levels of hospital care
Patients were treated in 64 different surgical and 32 medical hospitals and departments, of them 24 surgical departments participated in the documentation of more than 85 % of surgical patients and 21 medical departments participated in the documentation of more than 85 % of the medical patients, respectively. The levels of care were assigned to each hospital or department and gastric cancer according to the hospital plan of the Bavarian State Ministry of Environment and Public Health. Thus, three levels are defined for general hospitals (level I n = 2281 patients, level II n = 2,032 patients, level III n = 1,710 patients) and speciality hospitals (n = 343 patients) apply as a separate class. In this study, the university hospitals (n = 1,990 patients) with their level III maximum care are classified separately to be compared to the non-university level III hospitals, likewise all cancer cases which could not be related to a primary caregiver (n = 245) were considered separately.
Statistics
The MCR organises data in an Oracle database. Statistical analyses were run in SAS (Statistical Analysis System 9.2). Frequency data were analysed using the chi-square test. The percentages of the presented subcategories are related to the sum of each item with available data; missing values are not taken into account.
Observed (unadjusted overall) survival was estimated by the Kaplan–Meier method and tested by the log-rank test. Relative survival was calculated by the ratio of the observed survival rate to the expected survival rate. The expected survival time of age- and gender-matched individuals was calculated using life tables of the general German population. Relative survival can be interpreted as survival from cancer after correction for other causes of death and therefore is used as an estimate for cancer-specific survival. Conditional and cumulative mortality rates are computed by the Cutler–Ederer Method. Independent prognostic factors influencing overall survival were investigated by Cox proportional hazards models. Hazard ratios (HR) and 95 % confidence intervals (95 % CI) are presented. The following prognostic factors were entered simultaneously as independent variables in the multivariate analyses: age and sex as patient characteristics, TNM categories, grade and primary tumour location as tumour characteristics, and levels of hospital care as further independent variables. The significance level α was set at 0.05 in all statistical tests.
Results
Patient and tumour characteristics
The population-based cohort of gastric carcinoma comprises 8,601 cases of invasive gastric cancer from 1998 to 2012. Table 2 shows the patient and tumour characteristics by level of hospital care. Age is distributed very differently. The median age in university hospitals is lowest with 65.7 years (±12.6) compared to 74.5 years (±11.6) in level I hospitals. Accordingly, in university hospitals, 40 % of patients are aged 70 years or older compared to 70 % in level I hospitals. The proportion of female patients is also lowest in university hospitals with 36 % and highest in level I hospitals with 46.5 %.
Table 2.
Patient and tumour characteristics by level of hospital care (1998–2012, n = 8,601)
| Level of hospital care | University hospital | Level III hospital |
Level II hospital |
Level I hospital |
Speciality hospital |
Hospital n.a. |
Total | p value | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| n = 1,990 | n = 1,710 | n = 2,032 | n = 2,281 | n = 343 | n = 245 | n = 8,601 | |||||||||
| Prognostic factors | n | (%)a | n | (%)a | n | (%)a | n | (%)a | n | (%)a | n | (%)a | n | (%)a | |
| Age | |||||||||||||||
| mean ± SD | 65.7 | ±12.6 | 71.2 | ±12.5 | 72.2 | ±12.3 | 74.5 | ±11.6 | 73.8 | ±12.2 | 76.0 | ±12.1 | 71.3 | ±12.7 | <.0001 |
| Age in classes | |||||||||||||||
| <50 years | 240 | (12.1) | 104 | (6.1) | 115 | (5.7) | 94 | (4.1) | 13 | (3.8) | 11 | (4.5) | 577 | (6.7) | <.0001 |
| 50–59 years | 367 | (18.4) | 221 | (12.9) | 204 | (10.0) | 169 | (7.4) | 38 | (11.1) | 15 | (6.1) | 1,014 | (11.8) | |
| 60–69 years | 583 | (29.3) | 373 | (21.8) | 448 | (22.1) | 414 | (18.2) | 70 | (20.4) | 41 | (16.7) | 1,929 | (22.4) | |
| 70–79 years | 581 | (29.2) | 568 | (33.2) | 660 | (32.5) | 789 | (34.6) | 103 | (30.0) | 70 | (28.6) | 2,771 | (32.2) | |
| ≥80 years | 219 | (11.0) | 444 | (26.0) | 605 | (29.8) | 815 | (35.7) | 119 | (34.7) | 108 | (44.1) | 2,310 | (26.9) | |
| Sex | |||||||||||||||
| Male | 1,278 | (64.2) | 947 | (55.4) | 1,123 | (55.3) | 1,221 | (53.5) | 186 | (54.2) | 134 | (54.7) | 4,889 | (56.8) | <.0001 |
| Female | 712 | (35.8) | 763 | (44.6) | 909 | (44.7) | 1,060 | (46.5) | 157 | (45.8) | 111 | (45.3) | 712 | (35.8) | |
| T Category (comb.) | |||||||||||||||
| T1 | 317 | (20.1) | 325 | (27.7) | 347 | (25.0) | 316 | (23.0) | 40 | (23.4) | 19 | (30.2) | 1,364 | (23.7) | <.0001 |
| T2 | 438 | (27.7) | 397 | (33.9) | 494 | (35.5) | 458 | (33.4) | 53 | (31.0) | 20 | (31.8) | 1,860 | (32.4) | |
| T3 | 692 | (43.8) | 330 | (28.2) | 357 | (25.7) | 392 | (28.6) | 55 | (32.2) | 15 | (22.8) | 1,841 | (32.0) | |
| T4 | 133 | (8.4) | 120 | (10.2) | 192 | (13.8) | 207 | (15.1) | 23 | (13.5) | 9 | (14.3) | 684 | (11.9) | |
| Not available | 410 | (20.6) | 538 | (31.5) | 642 | (31.6) | 908 | (39.8) | 172 | (50.2) | 182 | (74.3) | 2,852 | (33.2) | |
| N Category (comb.) | |||||||||||||||
| N0 | 593 | (36.8) | 439 | (37.1) | 533 | (37.3) | 540 | (37.8) | 57 | (32.2) | 27 | (47.4) | 2,189 | (37.2) | <.0001 |
| N+ | 967 | (60.0) | 664 | (56.1) | 822 | (57.6) | 772 | (54.1) | 107 | (60.5) | 27 | (47.4) | 3,359 | (57.1) | |
| NX | 53 | (3.3) | 81 | (6.8) | 73 | (5.1) | 114 | (8.0) | 13 | (7.3) | 3 | (5.3) | 337 | (5.7) | |
| Not available | 377 | (18.9) | 526 | (30.8) | 604 | (29.7) | 855 | (37.5) | 166 | (48.4) | 188 | (76.7) | 2,716 | (31.6) | |
| M Category (primary) | |||||||||||||||
| M1 | 571 | (28.7) | 461 | (27.0) | 533 | (26.2) | 555 | (24.3) | 97 | (28.3) | 41 | (16.7) | 2,258 | (26.3) | 0.0003 |
| Grade | |||||||||||||||
| G1 | 55 | (2.3) | 67 | (4.2) | 129 | (6.9) | 100 | (4.7) | 18 | (6.3) | 15 | (7.7) | 384 | (4.8) | <.0001 |
| G2 | 403 | (21.2) | 472 | (29.8) | 529 | (28.3) | 609 | (28.8) | 72 | (25.3) | 82 | (42.3) | 2,167 | (27.3) | |
| G3/4 | 1,442 | (75.9) | 1,047 | (66.0) | 1,213 | (64.8) | 1,404 | (66.5) | 195 | (68.4) | 97 | (50.0) | 5,398 | (67.9) | |
| Not available | 90 | (4.5) | 124 | (7.3) | 161 | (7.9) | 168 | (7.4) | 58 | (16.9) | 51 | (20.8) | 652 | (7.6) | |
| Lauren classification | |||||||||||||||
| Intestinal | 869 | (55.2) | 774 | (59.4) | 904 | (59.5) | 1,085 | (61.0) | 146 | (64.9) | 90 | (64.8) | 3,868 | (59.2) | <.0001 |
| Diffuse | 601 | (38.2) | 484 | (37.1) | 569 | (37.5) | 639 | (35.9) | 73 | (32.4) | 44 | (31.7) | 2,410 | (36.9) | |
| Mixed | 103 | (6.6) | 46 | (3.5) | 46 | (3.0) | 54 | (3.0) | 6 | (2.7) | 5 | (3.6) | 260 | (4.0) | |
| Not available | 417 | (20.1) | 406 | (23.7) | 513 | (25.3) | 503 | (22.1) | 118 | (34.4) | 106 | (43.3) | 2,063 | (24.0) | |
| Histology | |||||||||||||||
| Adenocarcinoma | 1,362 | (69.4) | 1,220 | (72.6) | 1,454 | (72.9) | 1,691 | (75.5) | 263 | (80.2) | 178 | (78.4) | 6,168 | (73.2) | <.0001 |
| Signet cell carcinoma | 550 | (28.0) | 425 | (25.3) | 489 | (24.5) | 482 | (21.5) | 55 | (16.8) | 41 | (18.1) | 2,042 | (24.2) | |
| Other carcinoma | 51 | (2.6) | 35 | (2.1) | 51 | (2.6) | 66 | (3.0) | 10 | (3.1) | 8 | (3.5) | 221 | (2.6) | |
| Not available | 27 | (1.4) | 30 | (1.8) | 38 | (1.9) | 42 | (1.8) | 15 | (4.4) | 18 | (7.4) | 170 | (2.0) | |
| Location of primary tumour | |||||||||||||||
| GEJ, Cardia | 665 | (36.4) | 287 | (18.8) | 355 | (20.1) | 363 | (18.3) | 65 | (24.0) | 49 | (30.4) | 1,784 | (23.7) | <.0001 |
| Corpus, others | 769 | (42.1) | 731 | (47.9) | 824 | (46.7) | 905 | (45.7) | 127 | (46.9) | 66 | (41.0) | 3,422 | (45.4) | |
| Antrum, Pylorus | 391 | (21.4) | 508 | (33.3) | 586 | (33.2) | 714 | (36.0) | 79 | (29.2) | 46 | (28.6) | 2,324 | (30.9) | |
| Not available | 165 | (8.3) | 184 | (10.8) | 267 | (13.1) | 299 | (13.1) | 72 | (21.0) | 84 | (34.3) | 1,071 | (12.5) | |
| Total | 1,990 | (100) | 1,710 | (100) | 2,032 | (100) | 2,281 | (100) | 343 | (100) | 245 | (100) | 8,601 | (100) | |
n.a., not available; SD, standard deviation; comb., combined cT/pT or cN/pN; GEJ, gastro-oesophageal junction
a Missing values were excluded from calculations of frequency distribution; column percentage can differ slightly from 100 % due to rounding
The proportions of T3 and N+ tumours are the highest in university hospitals, but missing values concerning this are nearly twice as high in all other hospitals. Primary distant metastases (M1) are between 29 % in university hospitals and 24 % in level I hospitals. The proportion of tumours at the gastro-oesophageal junction is the highest in university hospitals with 37 % and about 20 % in the other hospitals. Furthermore, university hospitals treat significantly more G3/4 tumours and less intestinal-type adenocarcinomas than level I–III hospitals.
Therapy
Table 3 presents the therapy of gastric cancer as provided. In university hospitals, 75 % of the patients were treated by surgery. In all other hospitals, this percentage was lower than 70 %.
Table 3.
Therapy by level of hospital care (1998–2012, n = 8,601)
| Level of hospital care | University hospital | Level III hospital | Level II hospital | Level I hospital | Speciality hospital | Hospital n.a. |
Total | p value | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| n = 1,990 | n = 1,710 | n = 2,032 | n = 2,281 | n = 343 | n = 245 | n = 8,601 | |||||||||
| Therapy characteristics | n | (%)a | n | (%)a | n | (%)a | n | (%)a | n | (%)a | n | (%)a | n | (%)a | |
| Therapy | |||||||||||||||
| Surgery only | 932 | (52.8) | 936 | (71.4) | 1,118 | (72.5) | 1,151 | (77.0) | 148 | (70.1) | 52 | (68.4) | 4,337 | (67.8) | <.0001 |
| Surgery + CTX/RCTX (neoadj.) | 394 | (22.3) | 145 | (11.1) | 95 | (6.2) | 87 | (5.8) | 9 | (4.3) | 7 | (9.2) | 737 | (11.5) | |
| Surgery + Chemo (adj.) | 97 | (5.5) | 62 | (4.7) | 102 | (6.6) | 92 | (6.2) | 11 | (5.2) | 4 | (5.3) | 368 | (5.8) | |
| Surgery + RTX/RCTX (adj.) | 65 | (3.7) | 21 | (1.6) | 57 | (3.7) | 31 | (2.1) | 10 | (4.7) | 184 | (2.9) | |||
| CTX only | 232 | (13.2) | 128 | (9.8) | 151 | (9.8) | 118 | (7.9) | 23 | (10.9) | 11 | (14.5) | 663 | (10.4) | |
| RTX/RCTX only | 44 | (2.5) | 19 | (1.5) | 19 | (1.2) | 15 | (1.0) | 10 | (4.7) | 2 | (2.6) | 109 | (1.7) | |
| Palliation/not available | 226 | (11.4) | 399 | (23.3) | 490 | (24.1) | 787 | (34.5) | 132 | (38.5) | 169 | (69.0) | 2,203 | (25.6) | |
| Therapy | n = 1,990 | n = 1,710 | n = 2,023 | n = 2,281 | n = 343 | n = 245 | n = 8,601 | ||||||||
| Surgery ± CTX/RTX/RCTX | 1,488 | (74.8) | 1,164 | (68.1) | 1,372 | (67.5) | 1,361 | (59.7) | 178 | (51.9) | 63 | (25.7) | 5,626 | (65.4) | <.0001 |
|
No surgery/palliation/ not available |
502 | (25.2) | 546 | (31.9) | 660 | (32.5) | 920 | (40.3) | 165 | (48.1) | 182 | (74.3) | 2,975 | (34.6) | |
| Surgery | n = 1,488 | n = 1,164 | n = 1,372 | n = 1,361 | n = 178 | n = 63 | n = 5,626 | ||||||||
| Oncologic resection | 1,387 | (93.2) | 1,058 | (90.9) | 1,229 | (89.6) | 1,174 | (86.3) | 138 | (77.5) | 11 | (17.5) | 4,997 | (88.8) | <.0001 |
| n = 1,387 | n = 1,058 | n = 1,229 | n = 1,174 | n = 138 | n = 11 | n = 4,997 | |||||||||
| Gastrectomy | 848 | (61.1) | 417 | (39.4) | 647 | (52.6) | 586 | (49.9) | 70 | (50.7) | 3 | (27.3) | 2,571 | (51.5) | <.0001 |
| Partial gastric resection | 267 | (19.3) | 414 | (39.1) | 373 | (30.4) | 412 | (35.1) | 48 | (34.8) | 4 | (36.4) | 1,518 | (30.4) | |
| Surgery incl. oesophageal res. | 190 | (13.7) | 107 | (10.1) | 87 | (7.1) | 46 | (3.9) | 2 | (1.5) | 1 | (9.1) | 433 | (8.7) | |
| Other tumour resection | 82 | (5.9) | 120 | (11.3) | 122 | (9.9) | 130 | (11.1) | 18 | (13.0) | 3 | (27.3) | 475 | (9.5) | |
| Surgery | n = 1,488 | n = 1,164 | n = 1,372 | n = 1,361 | n = 178 | n = 63 | n = 5,626 | ||||||||
| Residual-status | |||||||||||||||
| R0 | 689 | (81.2) | 737 | (79.4) | 931 | (84.2) | 879 | (81.9) | 101 | (79.5) | 8 | (61.5) | 3,345 | (81.7) | 0.0363 |
| R1/R2/RX | 160 | (18.8) | 191 | (20.6) | 175 | (15.8) | 194 | (18.9) | 26 | (20.5) | 5 | (38.5) | 751 | (18.3) | |
| Not available | 639 | (42.9) | 236 | (20.3) | 266 | (19.4) | 288 | (21.2) | 51 | (28.7) | 50 | (79.4) | 1,530 | (27.2) | |
| Surgery | n = 1,488 | n = 1,164 | n = 1,372 | n = 1,361 | n = 178 | n = 63 | n = 5,626 | ||||||||
| Lymphonodectomy | |||||||||||||||
| Performed | 1,028 | (69.1) | 586 | (50.3) | 686 | (50.0) | 673 | (49.5) | 68 | (38.2) | 7 | (11.1) | 3,048 | (54.2) | <.0001 |
| Thereof D1 | 20 | (2.0) | 80 | (13.7) | 56 | (8.2) | 88 | (13.1) | 12 | (17.7) | 1 | (14.3) | 257 | (8.4) | <.0001 |
| Thereof D2 | 789 | (76.8) | 329 | (56.1) | 416 | (60.6) | 422 | (62.7) | 38 | (55.9) | 2 | (28.6) | 1,996 | (65.5) | |
n.a. not available, CTX Chemotherapy, RCTX combined radiochemotherapy, RTX radiation therapy
aMissing values were excluded from calculations of frequency distribution; column percentage can differ slightly from 100 % due to rounding
If surgery was performed, the proportions of oncologic resections were 93 % in university hospitals, 91 % in other level III hospitals, 90 % in level II and 86 % in level I hospitals. Between 79 and 84 %, R0 resections were reported. Performed lymphadenectomy was reported in 69 % of surgically treated patients of university hospitals and about 50 % in level I to other level III hospitals.
Mortality
Post-operative 30-days mortality or mortality after initial diagnosis (in cases of non-surgical treatment or missing information about surgery) differed between 4.5 % in university hospitals and 14.5 % in speciality hospitals as presented in Table 4. In university hospitals, the 30-days mortality was about a third of that in level I and a half of that in level II and other level III hospitals.
Table 4.
Mortality rates (post-surgery or since initial diagnosis) by level of hospital care
| Level of hospital care | University hospital | Level III hospital | Level II hospital | Level I hospital | Speciality hospital | Hospital n.a. |
Total | ||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| n = 1,990 | n = 1,710 | n = 2,032 | n = 2,281 | n = 343 | n = 245 | n = 8,601 | |||||||||||||||
| Mortality rates | n | % cond. | % cum. | n | % cond. | % cum. | n | % cond. | % cum. | n | % cond. | % cum. | n | % cond. | % cum. | n | % cond. | % cum. | n | % cond. | % cum. |
| Died within | |||||||||||||||||||||
| ≤30 days | 90 | 4.5 | 4.5 | 166 | 9.7 | 9.7 | 186 | 9.2 | 9.2 | 307 | 13.5 | 13.5 | 50 | 14.5 | 14.5 | 36 | 14.6 | 14.6 | 835 | 9.7 | 9.7 |
| >30 to ≤60 days | 86 | 4.5 | 8.8 | 104 | 6.7 | 15.8 | 142 | 7.7 | 16.2 | 179 | 9.1 | 21.3 | 29 | 9.9 | 23.0 | 25 | 11.9 | 24.8 | 565 | 7.3 | 16.3 |
| >60 to ≤90 days | 81 | 4.5 | 12.9 | 66 | 4.6 | 19.7 | 89 | 5.2 | 20.5 | 123 | 6.9 | 26.7 | 13 | 4.9 | 26.7 | 20 | 10.8 | 32.9 | 392 | 5.5 | 20.8 |
| >90 to ≤1 year | 475 | 27.5 | 36.9 | 402 | 29.6 | 43.5 | 510 | 31.9 | 45.9 | 553 | 33.2 | 51.0 | 97 | 39.0 | 55.3 | 72 | 43.8 | 62.3 | 2,109 | 31.2 | 45.6 |
| >1 to ≤2 years | 306 | 25.1 | 52.7 | 195 | 21.5 | 55.6 | 263 | 25.4 | 59.6 | 281 | 26.2 | 63.9 | 43 | 30.1 | 68.7 | 23 | 25.3 | 71.8 | 1,111 | 24.9 | 59.1 |
| >2 to ≤ 5 years | 243 | 30.1 | 66.9 | 180 | 29.6 | 68.7 | 202 | 31.3 | 72.3 | 229 | 33.6 | 76.0 | 41 | 47.7 | 83.6 | 13 | 24.5 | 78.7 | 908 | 31.5 | 72.0 |
| >5 to ≤10 years | 94 | 25.4 | 75.3 | 76 | 26.8 | 77.1 | 79 | 31.6 | 81.0 | 83 | 29.4 | 83.0 | 8 | 24.2 | 87.6 | 5 | 30.3 | 85.2 | 345 | 28.0 | 79.9 |
| >10 to ≤15 years | 14 | 16.3 | 79.4 | 14 | 19.1 | 81.5 | 9 | 22.5 | 85.3 | 13 | 21.3 | 86.7 | 6 | 42.9 | 92.9 | 0 | 0.0 | 85.2 | 56 | 20.7 | 84.0 |
| Alive/lost to follow-up | 601 | 507 | 552 | 513 | 56 | 51 | 2,280 | ||||||||||||||
Survival
The survival analyses include 7,196 patients with gastric cancer as their first malignant tumour from the years 1998–2012. The average and median observation time of the survivors was 64.7 and 56.8 months, respectively. The median relative survival of all the total population was about 1.6 years, and the relative 5-year survival was 33 %.
Figure 2 presents the observed and relative survival of all invasive gastric cancer stratified by level of hospital care. The greatest differences were between university hospitals and speciality hospitals or those without classification to a hospital care level with a median relative survival of 2.5 years and 6 months, respectively. The relative 10-year survival was between 31 and 20 %.
Fig. 2.
Overall (OS) and relative survival (RS) by level of hospital care (n = 7,196, log-rank test for OS p < 0.0001)
Figure 3 shows the relative survival in the primary M0 cohort with even greater differences. The median survival was about 5.5 years in those with university hospital care and 8 months in those without classification to a hospital. The relative 10-year survival was 43 % with university hospital care and 29 % with care in a speciality hospital.
Fig. 3.
Relative survival in primary M0 by level of hospital care (n = 5,260, log-rank test for OS p < 0.0001)
As presented in Fig. 4, the differences were smaller in the primary M1 cohort with its bad prognosis, and university hospitals as well had a better median survival: 9.5 versus 3.5 months. The relative 3-year survival was less than 10 % for all patients.
Fig. 4.
Relative survival in primary M1 by level of hospital care (n = 1,936, log-rank test for OS p < 0.0001)
Table 5 shows multivariate Cox proportional hazard regression analyses for the total cohort and for primary M0 and M1 separately. Age was implemented in the models meticulously with 5-year steps from the age of 70 to maximise the adjustment for comorbidities. Sex, TN, grading and primary tumour site were considered as well. In the model for the total cohort, the level of hospital care is statistically significant with hazard ratios lower than 1 for university and level III hospital care. In the model for patients with primary M0, however, level of hospital care is statistically not significant. Only in the model for patients with primary M1, level of hospital care is statistically significant for university and level III hospital care.
Table 5.
Cox proportional hazard regression analyses of the association between level of hospital care and survival (total cohort n = 7,196, primary M0 n = 5,260, primary M1 n = 1,936)
| Total n = 7,196 5,198 events HR (95 % CI) |
M0 n = 5,260 3,398 events HR (95 % CI) |
M1 n = 1,936 1,800 events HR (95 % CI) |
|
|---|---|---|---|
| Age (years) | p < 0.0001 a | p < .0001 a | p < .0001 a |
| <50 | Ref. | Ref. | Ref. |
| 50–59 | 1.03 (0.90–1.18) | 1.14 (0.94–1.39) | 0.95 (0.78–1.16) |
| 60–69 | 1.27 (1.12–1.43) | 1.41 (1.18–1.69) | 1.15 (0.97–1.38) |
| 70–74 | 1.59 (1.39–1.81) | 1.87 (1.56–2.26) | 1.34 (1.11–1.61) |
| 75–79 | 2.10 (1.85–2.39) | 2.58 (2.15–3.08) | 1.58 (1.30–1.94) |
| 80–84 | 2.70 (2.37–3.08) | 3.27 (2.72–3.93) | 2.10 (1.71–2.59) |
| 85–89 | 3.00 (2.59–3.46) | 3.65 (3.01–4.44) | 2.23 (1.73–2.89) |
| ≥90 | 3.27 (2.76–3.88) | 3.79 (3.05–4.71) | 3.05 (2.13–4.37) |
| Sex | p = 0.0499 a | p = 0.1570 a | p = 0.1438 a |
| Male | Ref. | Ref. | Ref. |
| Female | 0.95 (0.89–1.00) | 0.95 (0.89–1.02) | 0.93 (0.85–1.03) |
| T category | p < 0.0001 a | p < .0001 a | p = 0.1235 a |
| T1 | Ref. | Ref. | Ref. |
| T2 | 1.78 (1.57–2.01) | 1.81 (1.59–2.07) | 1.05 (0.66–1.68) |
| T3 | 2.52 (2.21–2.86) | 2.67 (2.32–3.08) | 1.25 (0.80–1.97) |
| T4 | 2.88 (2.48–3.34) | 3.66 (3.04–4.40) | 1.27 (0.80–2.01) |
| T n.a. | 2.67 (2.33–3.07) | 2.67 (2.27–3.15) | 1.37 (0.87–2.17) |
| N category | p < 0.0001 a | p < .0001 a | p < .0001 a |
| N0 | Ref. | Ref. | Ref. |
| N+ | 1.94 (1.78–2.13) | 1.93 (1.75–2.13) | 1.59 (1.22–2.08) |
| NX | 2.09 (1.78–2.46) | 2.16 (1.70–2.76) | 1.70 (1.23–2.34) |
| N n.a. | 2.93 (2.59–3.32) | 3.12 (2.66–3.65) | 2.27 (1.69–3.03) |
| M category | p < .0001 a | ||
| M0 | Ref. | – | – |
| M1 | 2.42 (2.27-2.59) | – | – |
| Grade | p < 0.0001 a | p < .0001 a | p < .0001 a |
| G1 | Ref. | Ref. | Ref. |
| G2 | 1.05 (0.89–1.23) | 1.07 (0.89–1.27) | 1.02 (0.66–1.57) |
| G3/4 | 1.35 (1.15–1.58) | 1.37 (1.15–1.63) | 1.36 (0.89–2.07) |
| G n.a. | 1.08 (0.90–1.29) | 0.94 (0.76–1.17) | 1.40 (0.92–2.17) |
| Primary tumour location | p = <.0001 a | p < .0001 a | p = 0.6889 a |
| GEJ, Cardia | Ref. | Ref. | Ref. |
| Corpus, others | 0.86 (0.80–0.93) | 0.80 (0.72–0.88) | 1.01 (0.88–1.45) |
| Antrum, Pylorus | 0.98 (0.90–1.07) | 0.94 (0.84–1.05) | 1.06 (0.92–1.23) |
| n.a. | 1.05 (0.95–1.16) | 1.03 (0.91–1.17) | 1.08 (0.91–1.27) |
| Level of hospital care | p = 0.0006 a | p = 0.2550 a | p = 0.0009 a |
| University hospital | 0.84 (0.78–0.92) | 0.90 (0.81–1.00) | 0.74 (0.64–0.85) |
| Level III | 0.89 (0.82–0.96) | 0.91 (0.82–1.01) | 0.85 (0.74–0.98) |
| Level II | 0.96 (0.88–1.04) | 0.99 (0.90–1.08) | 0.89 (0.78–1.02) |
| Level I | Ref. | Ref. | Ref. |
| Speciality hospital | 1.01 (0.88–1.16) | 0.97 (0.82–1.16) | 1.06 (0.83–1.35) |
| n.a. | 0.96 (0.82–1.12) | 0.96 (0.80–1.15) | 0.88 (0.62–1.25) |
HR hazard ratio, CI confidence interval, Ref. referent, GEJ gastro-oesophageal junction, n.a. not available
aWald test
Discussion
This study presents analyses of a population-based cohort of gastric cancer patients diagnosed between 1998 and 2012 with descriptions and outcomes that are as near to being representative for the German population. The relative 5-year survival of 33 % is comparable to 31 % for the whole of Germany. In comparison to other western industrialised countries, the relative 5-year survival of 33 % is somewhat better: 5-year relative survival in the USA is 28 % which is certainly explainable with a higher proportion of primary M1 cases (34 % in the USA vs. 26 % in the MCR) (Howlader et al. 1975); in Scandinavian countries, 5-year relative survival is between 23 and 29 % (Engholm et al. 2009), in Italy 29 % (AIRTUM Working Group 2011), in the UK 18 % (Cancer Research UK 2012).
At first view, it appears that the level of hospital care has an effect on survival and that being treated in a university or other level III hospital prolongs life. Level of hospital care remains a statistically significant independent factor in the multiple Cox regression of the outcome of the total cohort after adjustment for age, sex and the relevant prognostic tumour-specific variables. In the Cox model of the primary M0 cohort, which accounts for 74 % of all gastric cancer patients, however, a significant influence of level of hospital care cannot be seen. Only the multiple regression of the primary M1 cohort (a subgroup of 26 % of the whole cohort) demonstrates a significant effect of the level of care. This result leads initially to the question about the maximal possible extent of outcome improvement in the metastasised situation. Clinical studies in the 1990s found a median survival of 2–3 months in M1 patients with best supportive care (Pyrhönen et al. 1995; Scheithauer et al. 1996). Under treatment with 5-FU containing chemotherapy regimens, the median survival improved to 6–10 months. Recent trials report median survival times of 9–13 months.(Kang and Kauh 2011). A population-based study from the Netherlands, however, found no survival improvement of median survival in M1 patients despite the increased use of palliative chemotherapy (Bernards et al. 2013).
An explanation for better results in university hospitals for this subgroup could be the lower average age of the patients, but this should have an effect in both the M1 and M0 cohorts. Otherwise, the more the age is precisely modelled (e.g. in 5-year classes above 70 years), the weaker the estimates of level of care become. This should be a clue to differences in patient collectives by the different levels of hospital care but not to differences in care.
A further explanation for better results in M1 cohorts of university hospitals could be the more frequent and ‘better’ systemic treatment. The proportion of patients with chemotherapy is highest in university hospitals. But it may be assumed that improved survival is more likely an effect of a selected patient collective. Patients are younger, in better general condition, more capable and willing to cover greater distances for treatment and therefore get more effective (and exhausting) therapies. Additionally, the better survival of the M1 collective in university hospitals already is detectable shortly after initial diagnosis; if this was a consequence of better systemic therapy, it would be observed later on in disease course.
Another explanation for better results of the M1 cohort could be the better surgical treatment. The proportion of patients with oncologic resection is highest in university hospitals, and this could be a result of better expertise in university hospitals. Maybe university hospitals have better surgeons, better anaesthesiologists, better cardiologists and better intensive care facilities. The 30-day mortality would also match this. But this assumed better treatment should also be seen if not even more in the M0 collective, in which optimal primary therapy really can lead to cure. Why is this better therapy not represented in the M0 collective?
It is very plausible that university hospitals do not only attract younger patients but also those in better general condition with more agility and without noteworthy handicaps. These patients make their informed decisions and often are referred to university hospitals from hospitals with lower level of care.
Although the study cohort seems to be representative, it bears some limitations. First of all, the missing values for some items (for residual status for example) are very high and thus the results run the risk of bias and confounding. Nevertheless, with a better quality of data, the results and their interpretation would probably be confirmed. Another important aspect of the current study is the lack of information about comorbidities and patients’ general condition. In addition to age, this would lead to a better adjustment in multiple regression analyses, as would information in socioeconomic circumstances. A further point is that there is no possibility to obtain statements about hospitalisation routines and flow of patients, and therefore, the plausible considerations about patient selection in different levels of hospital care remain a bit uncertain.
Conclusion
The results of this study demonstrate that the level of hospital care has no influence on the outcome in gastric cancer in 74 % of the patients, namely the M0 cohort. Younger patients and probably patients in better general condition are treated more vigorously in hospitals of higher level of care, especially in university hospitals, while the aged and probably weaker patients appropriately undergo more conservative and milder therapies in nearby lower level of care hospitals. This induces highly selected patient collectives, especially highly selected M1 collectives, in university and level III hospitals and can explain survival differences by level of hospital care in this subgroup. There are no observable signs of deficits in patient care according to levels of hospital care. On the contrary, this study puts in a good word for appropriate treatment and care of gastric cancer patients in the MCR catchment area over all levels of hospital care.
Acknowledgments
We thank all the hospitals, departments and practitioners that participated in the documentation of the data. In particular, we thank the following 24 surgical hospitals and departments that participated in the documentation of 85 % of the presented surgically treated cohort: Klinikum rechts der Isar der TU, Chirurgische Klinik und Poliklinik: Prof. J. Kleeff; Klinikum der Ludwig-Maximilians-Universität-Großhadern, Chirurgische Klinik und Poliklinik: Prof. J. Werner, formerly Prof. K.-W. Jauch; Städt. Klinikum München-Neuperlach, Klinik für Allgemein- und Viszeralchirurgie, endokrine Chirurgie und Coloproktologie: Prof. N. Nüssler; Städt. Klinikum München-Bogenhausen, Klinik für Allgemein-, Viszeral-, Gefäß- und Thoraxchirurgie: Dr. R. Schmid, formerly Prof. W. Heitland; Städt. Klinikum München-Harlaching, Klinik für Allgemein- und Viszeralchirurgie und Coloproktologie: Prof. N. Nüssler; Klinikum Traunstein, Abteilung für Allgemein-, Viszeral- und Minimalinvasive Chirurgie: PD Dr. R. J. Schauer; Klinikum Dritter Orden, Klinik für Allgemein-, Viszeral-, Gefäß- und Thoraxchirurgie: Dr. D. Krenz; Klinikum Dachau, Abteilung für Viszeral- und Thoraxchirurgie: Prof. H.-G. Rau; Klinikum Rosenheim, Klinik für Allgemein-, Gefäß- und Thoraxchirurgie: Prof. P. Wagner; Kreisklinikum Altötting, Abteilung für Allgemein-, Viszeral- und Thoraxchirurgie: Prof. J. Roder; Krankenhaus Barmherzige Brüder München, Allgemein- und Viszeralchirurgie: Dr. C. Reuter; Städt. Klinikum München-Schwabing, Klinik für Viszeralmedizin: Prof. H. Waldner; Klinikum Landshut, Chirurgische Klinik I: Prof. F. Löhe; Klinikum Landkreis Erding, Abteilung für Viszeral- und Thoraxchirurgie: Dr. H. Bödecker; Klinikum der Ludwig-Maximilians-Universität-Innenstadt, Chirurgische Klinik und Poliklinik: Prof. W. Mutschler; Klinikum München-Pasing, Chirurgische Klinik: Dr. D. Laqua; Klinikum Freising, Abteilung für Allgemein-, Thorax- und Gefäßchirurgie: Dr. F. Zeller; Rotkreuzkrankenhaus, Chirurgische Abteilung: Prof. H. H. Schoenberg; Klinikum Ingolstadt, Chirurgische Klinik I: Prof. S. B. Hosch; Maria-Theresia-Klinik, Chirurgische Abteilung: Prof. T. F. Hoffmann; Klinikum Garmisch-Partenkirchen, Fachabteilung Allgemein-, Viszeral-, Thorax- und endokrine Chirurgie: PD Dr. H. Vogelsang; Klinikum Fürstenfeldbruck, Fachbereich Allgemein- und Visceralchirurgie: Dr. P. Knapp; Kreisklinik Trostberg, Allgemein- und Visceralchirurgie: Dr. J. Deuble; Kreisklinik Prien am Chiemsee, Chirurgische Abteilung: Prof J. Stadler. We also thank the following 21 medical hospitals and departments that participated in the documentation of 85 % of the presented oncologically treated cohort: Klinik Bad Trissl, Abteilung für Innere Medizin I: Dr. B. Weber; Städt. Klinikum München-Bogenhausen, Klinik für Gastroenterologie, Hepatologie und gastrointestinale Onkologie: Prof. W. Schepp; Städt. Klinikum München-Neuperlach, Klinik für Gastroenterologie und Hepatologie: Prof. W. Schmitt; Klinikum rechts der Isar der TU, II. Medizinische Klinik: Prof. R. Schmid; Städt. Klinikum München-Harlaching, Klinik für Hämatologie, Onkologie und Palliativmedizin: Prof. M. Karthaus; Klinikum Dritter Orden, Klinik für Innere Medizin I: Prof. A. Eigler; Klinikum Rosenheim, Medizinische Klinik II: Dr. H. Huff; Städt. Klinikum München-Neuperlach, Klinik für Hämatologie und Onkologie: Prof. M. Karthaus; Klinikum Landshut, Medizinische Klinik I: Prof. A. Holstege; Klinikum Garmisch-Partenkirchen, Zentrum Innere Medizin: Prof. H.-D. Allescher, Dr. H. Lambertz; Klinikum Traunstein, Abteilung für Gastroenterologie und Hepatologie: Dr. A. Kreuzmayr; Klinikum der Ludwig-Maximilians-Universität-Großhadern, Medizinische Klinik und Poliklinik II: Prof. B. Göke; Klinikum Freising, Abteilung für Gastroenterologie: Prof. E. Schulte-Frohlinde; Klinikum der Ludwig-Maximilians-Universität-Großhadern, Medizinische Klinik und Poliklinik III: Prof. W. Hiddemann; Klinikum Fürstenfeldbruck, Fachbereich Gastroenterologie und Onkologie: Prof. R. Eissele; Kreisklinik Ebersberg, Abteilung für Innere Medizin: Prof. T. Bernatik, Dr. H. L. Schneider; Klinikum Traunstein, Abteilung für Hämatologie, Onkologie und Palliativmedizin: Dr. T. W. Kubin; Klinikum München-Pasing, Medizinische Klinik II: Prof. B. Mayinger; Krankenhaus Barmherzige Brüder München, Innere Medizin I: PD Dr. C. Rust; Kreisklinik Trostberg, Innere Abteilung: Prof. T. Glück; Klinikum Ingolstadt, Medizinische Klinik II: Prof. J. Menzel. The Munich Cancer Registry (MCR) is part of the Munich Cancer Centre (MCC), an institution of the Medical Faculties of the Ludwig-Maximilians-University (LMU) and the Technical University (TU) Munich. Additionally, funding is given by the Bavarian Ministry of Health. In addition, analyses of the presented manuscript were funded by the Bundesministerium für Bildung und Forschung, BMBF (CHIR-Net, BMBF-Funding No. 01GH1001C and MSZ, BMBF-Funding No.01KN1104).
Conflict of interest
All authors declare that they have no conflict of interest.
References
- AIRTUM Working Group (2011) Italian cancer figures, report 2011: survival of cancer patients in Italy. Epidemiol Prev 35(Suppl 3):1–200 [PubMed] [Google Scholar]
- Armstrong K (2012) Methods in comparative effectiveness research. J Clin Oncol 30:4208–4214. doi:10.1200/JCO.2012.42.2659 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Bayerisches Staatsministerium für Umwelt und Gesundheit. Krankenhausplan des Freistaates Bayern (2013) http://www.gesetze-im-internet.de/khg/BJNR010090972.html
- Bernards N, Creemers G-J, Nieuwenhuijzen GA, Bosscha K, Pruijt H, Lemmens VE (2013) P-0020: No improvement in median survival for patients with metastasized gastric cancer despite increased use of chemotherapy. Ann Oncol 24:iv38–iv121. doi:10.1093/annonc/mdt203.19 [DOI] [PubMed] [Google Scholar]
- Cancer Research UK. Stomach Cancer Survival Statistics (2012) http://www.cancerresearchuk.org/cancer-info/cancerstats/types/stomach/survival/stomach-cancer-survival-statistics#One
- Endo S, Dousei T, Yoshikawa Y, Hatanaka N, Kamiike W, Nishijima J (2013) Prognosis of gastric carcinoma patients aged 85 years or older who underwent surgery or who received best supportive care only. Int J Clin Oncol 18:1014–1019. doi:10.1007/s10147-012-0482-9 [DOI] [PubMed] [Google Scholar]
- Engholm G, Ferlay J, Christensen N, Johannesen TB, Klint Å, Køtlum JE et al (2009) NORDCAN: cancer incidence, mortality, prevalence and survival in the Nordic countries version 5.3
- Hershman DL, Wright JD (2012) Comparative effectiveness research in oncology methodology: observational data. J Clin Oncol 30:4215–4222. doi:10.1200/JCO.2012.41.6701 [DOI] [PubMed] [Google Scholar]
- Howlader N, Noone AM, Krapcho M, Neyman N, Aminou R, Altekruse SF et al eds. (2012) SEER Cancer Statistics Review 1975–2009 (Vintage 2009 Populations), National Cancer Institute. Bethesda, MD. Posted to the SEER website Apr 2012: http://seer.cancer.gov/csr/1975_2009_pops09/
- Kang H, Kauh JS (2011) Chemotherapy in the treatment of metastatic gastric cancer: is there a global standard? Curr Treat Options Oncol 12:96–106. doi:10.1007/s11864-010-0135-z [DOI] [PubMed] [Google Scholar]
- Krankenhausfinanzierungsgesetz in der Fassung der Bekanntmachung vom 10. April (1991) (BGBl. I S. 886), das durch Artikel 5c des Gesetzes vom 15. Juli 2013 (BGBl. I S. 2423) geändert worden ist. 1991/2013
- Lyman GH (2009) Comparative effectiveness research in oncology: the need for clarity, transparency and vision. Cancer Invest 27:593–597. doi:10.1080/07357900903109952 [DOI] [PubMed] [Google Scholar]
- Mihaljevic AL, Friess H, Schuhmacher C (2013) Clinical trials in gastric cancer and the future. J Surg Oncol 107:289–297. doi:10.1002/jso.23120 [DOI] [PubMed] [Google Scholar]
- Moehler M, Al-Batran SE, Andus T, Anthuber M, Arends J, Arnold D et al (2011) [German S3-guideline “Diagnosis and treatment of esophagogastric cancer”]. Z Gastroenterol 49:461–531 [DOI] [PubMed] [Google Scholar]
- Munich Cancer Registry (Basic Statistics: C16 Gastric Cancer) (2013) http://www.tumorregister-muenchen.de/facts/base/base_C16__G.pdf. Accessed 2 Apr 2013
- Nashimoto A (2013) Current status of treatment strategy for elderly patients with gastric cancer. Int J Clin Oncol 18:969–970. doi:10.1007/s10147-012-0482-9 [DOI] [PubMed] [Google Scholar]
- Pyrhönen S, Kuitunen T, Nyandoto P, Kouri M (1995) Randomised comparison of fluorouracil, epidoxorubicin and methotrexate (FEMTX) plus supportive care with supportive care alone in patients with non-resectable gastric cancer. Br J Cancer 71:587–591 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Robert-Koch Institute, Association of Population-Based Cancer Registries (eds.) (2012) Cancer in Germany 2007/2008. 8th ed. Contributions to federal health reporting. Berlin
- Scheithauer W, Kornek G, Hejna M, Depisch D, Raderer M, Huber H (1996) Palliative chemotherapy versus best supportive care in patients with metastatic gastric cancer: a randomized trial. Ann Hematol 73(Suppl 2):A181 [Google Scholar]