Gastric cancer ranks as the fourth most common cancer and the second most frequent cause of cancer death worldwide.1 The aetiology includes causes such as Helicobacter pylori infection (for distal gastric cancer but not for cardia cancer),2 dietary imbalance, smoking, and genetic factors.2,3 Estimation of the incidence of second primary cancers may provide valuable insight into the aetiology and shared risk factors with the initial cancer. However, as gastric cancer patients have poor survival,4 a study of second malignancies after primary gastric cancer may not be informative. Instead, we examined the occurrence of second gastric cancers following any first cancers, based on the nationwide Swedish Family Cancer Database. This database has been described in detail previously.5 Briefly, it was created by linking information from the Multigenerational Register, censuses, Cancer Registry, and death notifications. The database has an almost complete follow up of registered cancer patients and it provides a unique opportunity to quantify the risks of developing second gastric cancers among all primary cancer patients. Person years at risk were accumulated for each subject from the data of diagnosis of the first malignancy to that of a second gastric cancer, death, emigration, or 31 December 2002, which came first. Standardised incidence ratio (SIR) was used to estimate the risks of second gastric cancers, adjusted for sex, age, period, residence, and socioeconomic level. Confidence intervals were calculated assuming a Poisson distribution. Family history included all first degree relatives (parents, siblings, and children) of the first cancer patients.
A total of 824 465 patients with first cancer were retrieved from this database, of whom 2648 developed a second gastric carcinoma. Table 1 shows the SIRs of second gastric cancers following all primary cancers. The risks were examined further by follow up time (data not shown). Significant increases after cervical, ovarian, and testicular cancers, and after non‐Hodgkin lymphoma and Hodgkin disease were mainly confined to the period >9 years of follow up, which could be partly explained by therapeutic effects because treatment related effects usually occur a decade after the first carcinoma.6 The highest SIR (mainly for cardia cancer) was noted after oesophageal carcinoma, but it was confined to the first year after diagnosis which may have been due to observation bias or increased surveillance. Gastric carcinoma was in excess after skin carcinoma throughout the follow up period, except for the first year, which was in agreement with previous studies.7 Epstein‐Barr virus may be related to the increase because skin cancer patients can be immunocompromised and thus at an increased risk of virally induced disease.7 Also, H pylori may escape from faltering immune surveillance, and the specific increase in corpus cancer would support the role of H pylori.
Table 1 Standardised incidence ratio (SIR) for second gastric cancers following first primary cancers.
| First cancer | Second gastric cancer | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| All | Corpus cancer | Cardia cancer | ||||||||
| Cancer site | Cases | O | SIR | 95% CI | O | SIR | 95% CI | O | SIR | 95% CI |
| Upper aerodigestive tract | 19092 | 105 | 1.16 | 0.95, 1.41 | 61 | 1.33 | 1.01, 1.70 | 9 | 0.83 | 0.38, 1.59 |
| Oesophagus | 6360 | 32 | 3.44 | 2.35, 4.86 | 14 | 3.19 | 1.74, 5.37 | 11 | 10.53 | 5.23, 18.91 |
| Stomach | 34152 | 36 | 0.43 | 0.30, 0.59 | 15 | 0.36 | 0.20, 0.60 | 6 | 0.77 | 0.28, 1.69 |
| Colon | 58767 | 218 | 1.19 | 1.03, 1.36 | 86 | 1.02 | 0.81, 1.26 | 27 | 1.27 | 0.84, 1.85 |
| Rectum | 33806 | 128 | 1.09 | 0.91, 1.30 | 51 | 0.94 | 0.70, 1.24 | 16 | 1.16 | 0.66, 1.88 |
| Lung | 59598 | 76 | 0.87 | 0.68, 1.09 | 31 | 0.73 | 0.50, 1.04 | 8 | 0.80 | 0.34, 1.58 |
| Breast | 114333 | 379 | 1.44 | 1.29, 1.59 | 192 | 1.59 | 1.38, 1.84 | 20 | 0.94 | 0.58, 1.46 |
| Cervix | 19425 | 74 | 1.37 | 1.08, 1.73 | 30 | 1.11 | 0.75, 1.58 | 4 | 1.07 | 0.28, 2.76 |
| Endometrium | 23351 | 66 | 0.85 | 0.66, 1.08 | 29 | 0.82 | 0.55, 1.17 | 8 | 1.29 | 0.55, 2.55 |
| Ovary | 22617 | 54 | 1.33 | 1.00, 1.73 | 30 | 1.54 | 1.04, 2.20 | 6 | 2.00 | 0.72, 4.38 |
| Prostate | 107644 | 510 | 1.02 | 0.93, 1.11 | 202 | 0.91 | 0.79, 1.04 | 79 | 1.13 | 0.90, 1.41 |
| Testis | 6065 | 20 | 1.92 | 1.17, 2.97 | 10 | 2.41 | 1.15, 4.45 | 5 | 2.64 | 0.83, 6.22 |
| Kidney | 25948 | 67 | 0.90 | 0.70, 1.14 | 28 | 0.78 | 0.52, 1.12 | 12 | 1.38 | 0.71, 2.42 |
| Urinary bladder | 38107 | 202 | 1.13 | 0.98, 1.29 | 73 | 0.89 | 0.70, 1.12 | 28 | 1.17 | 0.78, 1.69 |
| Melanoma | 30561 | 66 | 0.79 | 0.61, 1.01 | 29 | 0.88 | 0.59, 1.26 | 9 | 0.77 | 0.35, 1.48 |
| Skin, squamous cell | 27090 | 178 | 1.48 | 1.27, 1.71 | 89 | 1.70 | 1.36, 2.09 | 13 | 0.84 | 0.44, 1.44 |
| Nervous system | 32287 | 49 | 0.84 | 0.62, 1.11 | 22 | 0.82 | 0.51, 1.24 | 4 | 0.62 | 0.16, 1.62 |
| Thyroid gland | 8512 | 29 | 1.23 | 0.82, 1.76 | 8 | 0.73 | 0.31, 1.44 | 4 | 1.62 | 0.42, 4.20 |
| Endocrine glands | 15871 | 45 | 0.83 | 0.60, 1.11 | 19 | 0.79 | 0.48, 1.24 | 3 | 0.48 | 0.09, 1.42 |
| Connective tissue | 6438 | 23 | 1.33 | 0.84, 2.00 | 12 | 1.51 | 0.78, 2.65 | 4 | 1.90 | 0.49, 4.90 |
| Non‐Hodgkin's lymphoma | 25874 | 80 | 1.36 | 1.08, 1.70 | 28 | 1.12 | 0.74, 1.62 | 11 | 1.44 | 0.71, 2.58 |
| Hodgkin's disease | 6264 | 21 | 1.90 | 1.17, 2.91 | 8 | 1.49 | 0.64, 2.96 | 2 | 1.58 | 0.15, 5.83 |
| Myeloma | 11482 | 41 | 1.52 | 1.09, 2.06 | 25 | 1.90 | 1.23, 2.81 | 1 | 0.33 | 0.00, 1.90 |
| Leukaemia | 26161 | 49 | 0.85 | 0.63, 1.13 | 19 | 0.70 | 0.42, 1.10 | 5 | 0.77 | 0.24, 1.82 |
| All | 824465 | 2648 | 1.11 | 1.07, 1.15 | 1152 | 1.05 | 0.99, 1.11 | 304 | 1.10 | 0.98, 1.23 |
Bold type, 95% confidence interval (CI) does not include 1.00; underline type, 99% CI does not include 1.00.
O, observed.
Patients with a family history of the first cancer were at a higher risk than all patients (table 2). SIR was increased after familial colon carcinoma (3.31) and was very high when colon cancer patients were diagnosed at a young age (10.61). Inherited predisposition, such as hereditary non‐polyposis colorectal carcinoma (HNPCC), may contribute towards the occurrence of second gastric carcinomas.8 However, gastric cancers were not increased after other HNPCC related cancers (rectal, endometrial, and ovarian cancers). Gastric cancer was also increased among patients with familial breast cancer (2.32), particularly when diagnosed at a young age (3.43). Breast carcinomas are increased in BRCA1and BRCA2 mutation carriers who also have an increased risk of gastric carcinoma.9,10 Another link between breast and gastric cancer may be germline mutations of the E‐cadherin gene, which have been observed in patients with diffuse gastric cancer and breast carcinoma.11 This study suggested that shared genetic factors may be a contributor towards the development of second gastric cancers.
Table 2 Standardised incidence ratio (SIR) for second gastric cancers in patients with a family history in first degree relatives.
| Relatives with cancers (all ages) | Relatives with cancers (age <45 y) | |||||||
|---|---|---|---|---|---|---|---|---|
| Second cancers | Second cancers | |||||||
| Cases | O | SIR | 95% CI | Cases | O | SIR | 95% CI | |
| Cancer site | ||||||||
| Colon | 1268 | 9 | 3.37 | 1.53, 6.42 | 233 | 6 | 10.61 | 3.82, 23.25 |
| Breast | 7737 | 31 | 2.32 | 1.58, 3.30 | 1766 | 12 | 3.43 | 1.76, 6.01 |
| Prostate | 3540 | 12 | 1.00 | 0.52, 1.76 | 12 | 0 | ||
| Skin, squamous cell | 275 | 3 | 3.75 | 0.71, 11.09 | 31 | 0 | ||
| All | 19517 | 67 | 1.65 | 1.28, 2.10 | 3840 | 20 | 2.70 | 1.64, 4.17 |
Bold type, 95% confidence interval (CI) does not include 1.00; underline type, 99% CI does not include 1.00.
O, observed.
Acknowledgements
Supported by Deutsche Krebshilfe, the Swedish Cancer Society, the EU, LSHC‐CT‐2004‐503465, and the Swedish Council for Working Life and Social Research. The Family‐Cancer Database was created by linking registers maintained at Statistics Sweden and the Swedish Cancer Registry.
Footnotes
Conflict of interest: None declared.
References
- 1.Parkin D M. International variation. Oncogene 2004236329–6340. [DOI] [PubMed] [Google Scholar]
- 2.Huang J Q, Sridhar S, Chen Y.et al Meta‐analysis of the relationship between Helicobacter pylori seropositivity and gastric cancer. Gastroenterology 19981141169–1179. [DOI] [PubMed] [Google Scholar]
- 3.Crew K D, Neugut A I. Epidemiology of upper gastrointestinal malignancies. Semin Oncol 200431450–464. [DOI] [PubMed] [Google Scholar]
- 4.Coleman M P, Gatta G, Verdecchia A.et al EUROCARE‐3 summary: cancer survival in Europe at the end of the 20th century. Ann Oncol 200314(suppl 5)v128–v149. [DOI] [PubMed] [Google Scholar]
- 5.Hemminki K, Li X, Plna K.et al The nation‐wide Swedish family‐cancer database—updated structure and familial rates. Acta Oncol 200140772–777. [DOI] [PubMed] [Google Scholar]
- 6.Bhatia S, Sklar C. Second cancers in survivors of childhood cancer. Nat Rev Cancer 20022124–132. [DOI] [PubMed] [Google Scholar]
- 7.Hemminki K, Dong C. Primary cancers following squamous cell carcinoma of the skin suggest involvement of Epstein‐Barr virus. Epidemiology 20001194. [DOI] [PubMed] [Google Scholar]
- 8.Lindor N M, Greene M H. The concise handbook of family cancer syndromes. Mayo Familial Cancer Program. J Natl Cancer Inst 1998901039–1071. [DOI] [PubMed] [Google Scholar]
- 9. Cancer risks in BRCA2 mutation carriers. The Breast Cancer Linkage Consortium. J Natl Cancer Inst 1999911310–1316. [DOI] [PubMed] [Google Scholar]
- 10.Brose M S, Rebbeck T R, Calzone K A.et al Cancer risk estimates for BRCA1 mutation carriers identified in a risk evaluation program. J Natl Cancer Inst 2002941365–1372. [DOI] [PubMed] [Google Scholar]
- 11.Pharoah P D, Guilford P, Caldas C. Incidence of gastric cancer and breast cancer in CDH1 (E‐cadherin) mutation carriers from hereditary diffuse gastric cancer families. Gastroenterology 20011211348–1353. [DOI] [PubMed] [Google Scholar]