Malignant tumors of the kidney account for 2 to 3% of adult cancers worldwide1,2. Kidney cancer occurs about twice as often among men as among women, with average age at diagnosis of 60 to 65 years3. Incidence rates vary more than tenfold between countries, being highest in Scandinavia and other parts of northern Europe, and in North America. Survival of patients diagnosed with kidney cancer has improved over time, with five-year relative survival rate increasing from 30 to 40% in the 1960s to 50 to 60% in the 1990s3. Renal cell cancer (renal parenchymal cancer) represents about 80 to 85% of all adult kidney neoplasms. In Sweden, cancer of the renal parenchyma accounted for 82%, renal pelvis cancer for 9% and cancer at unspecified sub-sites for 9% of the total number in 19974.
Kidney cancer occurs in both sporadic and familial forms. Familial kidney cancer is rare because only a few percent of first-degree relatives share this type of neoplasia5. Familial cancers can be caused by environmental factors and habits shared by family members or by inherited gene defects. It is often impossible to apportion the environmental and heritable components; in twin studies such estimated can be obtained, but for kidney cancer even the largest published twin study was non-informative because of lack of concordant twin pairs6. However, for cancers lacking strong identified environmental risk factors, such as kidney cancer, familial risk factors may primarily reflect heritable risks7,8.
Familial renal cell cancer is manifested as a part of a rare autosomal-dominant cancer syndrome, von Hippel-Lindau (VHL) disease1,9. VHL is a monogenic disease caused by alteration in the von Hippel-Lindau (VHL) gene at chromosome 3p10. Affected individuals are at risk of developing tumors in number of organs, including the kidneys, central nervous system (hemangioblastoma), eye, inner ear, endocrine glands, and pancreas11. Renal pelvis cancer is a manifestation in hereditary nonpolyposis colorectal carcinoma (HNPCC) with about fivefold risk in mutation carriers in mismatch repair genes12. In a limited number of non-VHL families the genetic defects have been characterized to be inherited translocations, many of which include chromosome 313. Renal cancer also is involved in tuberous sclerosis, but most cases represent new mutations; brain tumors occur in a small number of cases14. Another familial form of kidney cancer is hereditary papillary renal carcinoma, a rare autosomal-dominant inherited syndrome, the true incidence of which is unknown. Affected family members develop multiple papillary tumors of varying size in both kidneys15,16. Some, but not all papillary renal cell carcinomas have been shown to be caused by activating mutations in the c-MET proto-oncogene. Some recently characterized renal cancers are Birt-Hogg-Dube syndrome with mainly chromophobe tumors and familial renal oncocytoma13.
Second primary malignancies associated with renal cell carcinoma include those of urinary bladder, prostate, rectal and lung cancer, as well as non-Hodgkin's lymphoma and melanoma17,18,19,20,21,22. The etiology of second cancers is most likely multifactorial and may include effects of radio- and chemotherapy, genetic predisposition, environmental exposures such as ultraviolet light, impaired immunological mechanisms, gender specific and hormonal factors and interactions of these factors.
Here we used the nationwide Swedish Family-Cancer Database, which comprises 10 million individuals and more than 1 million cancers, to assess the familial risks in kidney cancer. Familial risks in kidney cancers have previously been studied among some 20 other cancer sites or between different sites using our Database5,23,24. In this study we investigated the familial relationships of kidney cancer in detail and analyzed the site-specific risk of second primary neoplasm among 23,137 patients diagnosed with a kidney cancer during the years 1961 to 1998.
METHODS
The Swedish Family Cancer Database
The Swedish Family Cancer Database includes all persons born in Sweden after 1931 with their biological parents, totaling over 10.2 million individuals25. The Database is organized into 3.1 million families, with parents and offspring. It has been updated in the beginning of year 2001 to include cancers from the nationwide Swedish Cancer Registry from the years 1958 to 1998. Of the invasive cancers, over 769,000 were diagnosed among parents and 166,000 among offspring.
Since 1958, all new cases of cancer in Sweden have been reported to Swedish Cancer Register. The completeness of cancer registration in the 1970s has been estimated to be over 95%, and is now considered to be close to 100%. This has been achieved by compulsory reporting from clinicians who diagnose a neoplasm and the pathologists/cytologists who must report separately any diagnosis of malignancy made on pathological and cytological specimens.
Patients
Offspring were diagnosed for their first primary cancer at ages 0 to 66 years, while the age of parents at their diagnosis was not limited. Since there is incomplete information in the Database about death among cancer cases between years 1958 and 1960, we used the follow-up period between 1 January 1961 and 31 December 1998. A four-digit diagnostic code according to the 7th revision of the International Classification of Diseases (ICD-7) was used in combination with histology codes. The following ICD-7 codes were pooled: "upper aerodigestive tract" cancer codes 161 (larynx), and 140–148 (lip, mouth, pharynx), except for code 142 (salivary glands); and "leukemia" codes 204–207 (leukemias), 208 (polycythemia vera), and 209 (myelofibrosis). Kidney cancer (ICD-7 code 180) was subdivided to cancer of the renal parenchyma (1800), renal pelvis (1801) and cancer at unspecified sub-sites (1809). The main histopathological types of kidney cancer were adenocarcinoma, papillary type and Wilms tumor with pathological anatomic diagnosis (PAD) codes 096, 116 and 886. Hemangioblastoma was defined as benign tumor of the central nervous system (ICD-7 code 193) with PAD code 501 and 511. Besides hemangioblastoma there are a few other benign tumors of central nervous system and endocrine glands included in our Database to which were generally referred to as "cancer" during our analysis.
The incidence of second cancer was analyzed in all patients with an initial kidney cancer diagnosed between 1 January 1961 and 31 December 1998. During this period of time 1900 patients developed a second primary malignancy among 23,137 kidney cancer patients. Cases of second primary kidney malignancies in the Database were taken into analysis when the diagnosis date of the first and second malignancy differed by at least one month in order to exclude the possibility of two notifications of the same cancer.
Statistical analysis
Standardized incidence ratios (SIRs) were used to measure the cancer risks for offspring according to occurrence of cancers in their family. Cancer risks in offspring who had a parent affected with cancer are referred to as "offspring risk." Cancer risks in offspring who had a sibling affected with the cancer are referred to as "sibling risk." SIRs were calculated as the ratio of observed (O) to expected (E) number of cases. The expected numbers of cancers were obtained by assuming that these persons experienced the same cancer incidence as prevailed in the corresponding general population in the Database. Tumor site-, sex-, period- and age-specific rates were applied to the appropriate person-years at risk26. For first primary cancer, person-years at risk were accumulated for each offspring beginning with the date of birth or January 1, 1961 and ending with the date of diagnosis of a first primary cancer, date of death, date of emigration, or December 31, 1998.
For each cancer pair, that is, offspring-parent (in calculating "offspring risk") and offspring-sibling (in calculating "sibling risk"), we counted cancer cases and person-years at risk only once for each offspring regardless of the number of affected members in the family. Confidence intervals (95% CI) were calculated assuming that the numbers of cancers among offspring follow a Poisson distribution26. In the calculation of 95% CIs for sibling risks, the dependence between the affected pairs was taken into consideration27. The population-attributable proportion of cases with family history of kidney cancer was estimated as follows: (familial SIR - 1)/familial SIR 
proportion of familial cases28.
Standardized incidence ratios were used to estimate the risk for a new primary cancer after kidney cancer. Person-years at risk were accumulated for each patient beginning with the date of diagnosis of the first primary cancer and ending with the date of diagnosis of a second primary cancer, date of death, date of emigration, or December 31, 1998, whichever came first. Family histories SIRs for second cancers were calculated in a similar way, starting at the follow-up from the diagnosis of the kidney cancer. Family history information was collected on all first-degree relatives (parents, siblings, and children). The follow-up time was divided into three periods (<1 year; 1 to 10 years; 11 to 37 years) allowing the assessment of the effect of follow-up time.
RESULTS
The annual incidence rates for male and female kidney cancer in Sweden for the period 1961 to 1998 (standardized according to the European standard population) are shown in Figure 1a. The male incidence was about 1.7 times higher than the female incidence throughout the analyzed period. In both genders the incidence rates slightly decreased after 1985 and reached 6.80 for women and 11.30 for men per 100,000 person-years in year 1998. The age dependence of the incidence rates of kidney cancer is shown in Figure 1b. After the age of 50 years the incidence of kidney cancer increased steadily both in males and females.
Figure 1.
Incidence of kidney cancer in males and females according to (A) age as standardized according to the European population standard, and (B) by five-year age ranges. Symbols are: (
, 
) males; (
, 
) females.
Table 1 reports the characteristics of kidney cancer malignancies in the Family-Cancer Database by sub-sites and histological type. There were 23,137 cases of kidney cancer during the years 1961 through 1998. Cancer of the renal parenchyma accounted for 83.7%, renal pelvis cancer for 8.8%, and cancer at unspecified sub-sites for 7.5%. The most common histological type was adenocarcinoma, which accounted for 81% of all kidney cancers. Among 23,137 kidney cancer cases in Database, 3268 cases were observed in the offspring (1969 in men and 1299 in women) and 19,869 cases were observed in parents (11,824 in men and 8045 in women).
The risks of kidney cancer in offspring in association with any cancer in parent are shown in Table 2. There were 71 parent-offspring pairs concordant for kidney cancer in the Database. The risk of kidney cancer was increased in offspring giving a SIR of 1.56 (range 1.22 to 1.95). Risk was similar for both genders: 1.52 (1.10 to 2.01) in sons and 1.64 (1.09 to 2.30) in daughters. At a parental age of diagnosis 
66 years (the same age limit as for offspring), the risk of kidney cancer in offspring was 1.94 (1.36 to 2.61), while at higher parental age it was lower with SIR of 1.29 (0.90 to 1.77).
A discordant tumor site that associated with kidney cancer between the two generations was hemangioblastoma of the central nervous system. The SIRs for hemangioblastoma by parental kidney cancer and age at diagnosis are shown in Table 3. SIR by parental kidney cancer was 8.51 (3.06 to 16.69) for offspring diagnosed before the age of 25 years and 2.52 (1.08 to 4.57) for offspring diagnosed later. The highest risk for hemangioblastoma by parental kidney cancer of 17.39 (5.49 to 35.98) was observed for sons of parent with kidney cancer when diagnosed before age 25.
Table 3 - SIR for hemangioblastoma of central nervous system in offspring by parental kidney cancer.
Full table The sibling risk from a kidney cancer proband was 4.72 (2.28 to 9.20) Table 4. Among 32 sibs with kidney cancer 18 were males, giving a SIR of 4.34 (1.81 to 9.29) for males and 5.34 (2.06 to 12.02) for females. Discordant sites that associated with kidney cancer among sibs were ovaries with a SIR of 1.77 (1.03 to 2.71) and non-thyroid endocrine glands that increased risk for kidney cancer in males with a SIR 2.37 (1.22 to 3.90). In addition, a SIR of 3.53 (1.11 to 7.30) was observed when young age of onset (<50 years) was combined with pancreatic cancer of the sibling. Among the 32 affected siblings with kidney cancer only one had an affected parent with kidney cancer and 15 had a parent with other cancer type; 4 cases of parental colon cancer resulted in SIR of 17.57 (3.23 to 55.15), 4 cases of parental urinary bladder cancer in SIR of 32.35 (5.95 to 101.58) and 2 cases of parental melanoma in SIR of 42.87 (2.86 to 173.77).
Using the familial SIRs and the proportion of familial cases of kidney cancer, we calculated the population-attributable proportions for family history. The attributable proportion by parental kidney cancer (using familial SIR data from Table 2) was 0.78% [that is, (1.56 - 1)/1.56 (71/3268); the last term is the number of offspring with an affected parent, divided by the number of all offspring with kidney cancer]. The attributable proportion by sibling's kidney cancer was 0.77% (using familial SIR values from Table 4).
Table 5 reports on the risk of second primary cancers following kidney cancer by follow-up time. Since no apparent sex-specific differences were noted (except for second primary cancer of urinary bladder) when the analysis was carried out separately for males and females, the results in Table 5 are presented for combined genders. For patients with kidney cancer, a significantly elevated overall SIR for second primary cancer was found: colon (1.39), pancreas (1.81), lung (1.36), breast (1.25), prostate (1.70), kidney (2.00), urinary bladder (4.56), nervous system (4.12), endocrine glands (5.25), Non-Hodgkin lymphoma (1.86), and leukemia (2.45). The highest risk of second cancer was found within less than one year of follow-up, probably due to intensive medical surveillance. In the 1 to 10 year interval a higher SIR was observed as compared to 11 to 37 year interval for cancers at urinary bladder (4.01) and nervous system (2.80). The highest risk of second primary tumor was found for hemangioblastoma of central nervous system with a SIR of 18.99 in the 1 to 10 year interval and of 21.19 overall. In addition, the risk of hemangioblastoma following familial kidney cancer was 1206 (114 to 3456).
DISCUSSION
Few analytical epidemiological studies have presented quantitative risk estimates for family history of kidney cancer23,29. While some authors found a 60% increased risk for kidney cancer when a first-degree relative was affected with the disease29, no association was found in a smaller study30. In a study on the Utah population, analysis of familial relationships in kidney cancer (among many other cancer sites) has shown familial risk of 1.5331. In agreement with that analysis, we found that the risk to offspring when their parent had cancer was 1.56 (1.22 to 1.95). Interestingly, sibling risk with a SIR of 4.72 (2.28–9.20) was significantly higher when compared to offspring risk for kidney cancer. One possible reason for higher sibling risks could be the younger age distribution. However, a statistically significant ratio of sibling risk to offspring risk of 2.43 (1.51 to 3.90) was shown when the age of parents in our analysis was limited to 66 years (the same as for offspring). The population-attributable proportion by parental cancer (using familial SIR data from Table 2) was 0.78% and the attributable proportion by sibling's cancer was 0.77% (using familial SIR values from Table 4). These values are somewhat lower compared to estimates from case-control studies that suggest that up to 4% of kidney carcinomas may be hereditary1,11.
The genetic interpretation of the familial risks is that dominant effects are reflected in offspring risks (and in sibling risks when a parent is affected) whereas recessive effects are signaled by elevated sibling risks with no parent affected. The risk to offspring from an affected parent was of moderate magnitude in our study and can be at least partly explained by a dominant susceptibility due to the VHL gene. Although a high ratio of sibling risk to offspring risk is consistent with an involvement of recessive susceptibility, a high sibling risk of kidney cancer in this study could be explained partly by a shared childhood environment. Still, little is known about the relative importance of genes and environment in familial clustering of cancer, and in twin studies the effect of a shared environment is a minor cause of cancer6.
Our analysis revealed a few discordant sites that associated with familial kidney cancer. A VHL related tumor site that associated with an increased kidney cancer risk between the two generations was hemangioblastoma of central nervous system Table 3. VHL related cancers that associated with kidney cancer among siblings were those of endocrine glands with SIR of 2.37 (1.22 to 3.90) in males and pancreas with SIR of 3.53 (only for age of onset <50 years). We also showed an association of kidney cancer with ovaries, colon and urinary bladder cancers that were most probably due to hereditary non-polyposis colorectal cancer (HNPCC) syndrome12,32,33. Ovaries were associated with sibling risk for kidney cancer resulting in a SIR of 1.77 (1.03 to 2.71). Colon or urinary bladder cancer in parent in families with siblings affected with kidney cancer resulted in high familial risks: four cases of parental colon cancer in SIR of 17.57 (3.23 to 55.15) and four cases of parental bladder cancer in SIR of 32.35 (5.95 to 101.58). Multiple comparisons by cancer site in parent/sibling could possibly result in significant findings on the basis of chance. However, biological plausibility and previous epidemiological and molecular genetic data support the associations found in our study.
Even though other renal cancer syndromes have been identified, as discussed in the introductory paragraphs, the present epidemiological data do not allow their identification, probably because of their rarity and the lack of established associated malignancies. Brain cancer is a manifestation in tuberous sclerosis, but we found no association between kidney and brain cancer in any age group.
Second primary cancers may be a result of the same genetic factors as the first cancers and therefore have been of interest in genetic epidemiology. On the other hand, second primary cancers also may arise due to the treatment of the first cancer, or they may be diagnosed earlier because of the medical follow-up of the first cancer. Moreover, the second primary cancer may be a recurrence of the first cancer. To be able to distinguish between these factors, we subdivided the follow-up period Table 5. The increase in risk of the second cancers was largest during the first year of follow-up and most likely is due to the intensive medical surveillance after the first diagnosis. It is probable that urologists treating patients for kidney cancer found an increase in another urologic malignancy, for example, prostate cancer or urinary bladder cancer.
Treatment-related effects in solid tumors would be expected to manifest after about a decade from the diagnosis of the first cancer34,35,36. However, there was no significant increase in risks 11 to 37 years after the first diagnosis, compared to 1 to 10 years for most second cancers, excluding any large effects by treatment. Indeed, because the primary treatment modality for kidney cancer is surgery, the basis for the association between kidney cancer and second cancers is unlikely to be treatment related, but possibly due to common etiologic factors, either environmental or genetic.
Higher SIRs were seen in the 1 to 10 year interval compared to 11 to 37 year interval for cancers at urinary bladder (4.01) and nervous system (2.80). As an increase of risk for second cancer in the 1 to 10 year interval is likely to signal inherited or acquired susceptibility to cancer, it would be interesting to compare the patterns of multiple cancers in familial and sporadic cancer. However, only the nervous system had more than two familial cases of tumors Table 5. The highest risk of second primary tumor was seen for hemangioblastoma, a benign tumor of central nervous system, with a SIR of 18.99. Among three cases in the 1 to 10 year interval, there were two cases of familial hemangioblastoma following kidney cancer, giving a SIR value of 1668 (157 to 4780). The association between hemangioblastoma and kidney cancer in Table 5 (and also in Table 3) is likely to cover all VHL kidney cancer cases.
While familial risks associated with the VHL syndrome were high according to our analysis, they can explain only a small proportion of kidney cancers since it is a rare syndrome. The high ratio of sibling risk to offspring risk in kidney cancer may be indicative of a recessive susceptibility gene. The high risk for second primary cancer in the patients without a family history seems to be consistent with a polygenic model with a variable degree of environmental modification. We believe that our data should motivate a search for recessive candidate genes for kidney cancer susceptibility.
References
| 1. | Linehan W & Klausner R. Renal carcinoma. inThe Genetic Basis of Human Cancer 1998; edited by Vogelstein B, Kinzler B New York, McGraw-Hill pp 455–473. |
| 2. | Unwin RD, Knowles MA, Selby PJ & Banks RE. Urological malignancies and the proteomic-genomic interface. Electrophoresis 1999; 20: 3629–3637. | Article | PubMed | ISI | ChemPort | |
| 3. | McLaughlin J & Lipworth L. Epidemiologic aspects of renal cell cancer. Semin Oncol 2000; 27: 115–123. | PubMed | ISI | ChemPort | |
| 4. | Epidemiology Centre for Cancer Incidence in Sweden 1997 1999; Stockholm. |
| 5. | Hemminki K & Vaittinen P. Familial cancers in a nation-wide family-cancer database: Age distribution and prevalence. Eur J Cancer 1999; 35: 1109–1111. | Article | PubMed | ISI | ChemPort | |
| 6. | Lichtenstein P, Holm N & Verkasalo P et al. Environmental and heritable factors in the causation of cancer. N Engl J Med 2000; 343: 78–85. | Article | PubMed | ISI | ChemPort | |
| 7. | Hemminki K, Dong C & Vaittinen P. Cancer risks to spouses and offspring in the Family-Cancer Database. Genet Epidemiol 2001; 20: 247–257 10.1002/1098-2272(200102)20:2<247::aid-gepi7>3.0.co;2-u. | Article | PubMed | ISI | ChemPort | |
| 8. | Risch N. The genetic epidemiology of cancer: Interpreting family and twin studies and their implications for molecular genetic approaches. Cancer Epidemiol Biomarkers Prev 2001; 10: 733–741. | PubMed | ISI | ChemPort | |
| 9. | Friedrich C. Von Hippel-Lindau syndrome. A pleomorphic condition. Cancer 1999; 86: 2478–2482 10.1002/(sici)1097-0142(19991201)86:11+<2478::aid-cncr4>3.3.co;2-x. | Article | PubMed | ISI | ChemPort | |
| 10. | Latif F, Tory K & Gnarra J et al. Identification of the von Hippel-Lindau disease tumor suppressor gene. Science 1993; 2603: 17–20. |
| 11. | Linehan WM, Lerman MI & Zbar B et al. Identification of the von Hippel-Lindau (VHL) gene. Its role in renal cancer. JAMA 1995; 273: 564–570. | Article | PubMed | ISI | ChemPort | |
| 12. | Aarnio M, Sankila R & Pukkala E. Cancer risk in mutation carriers of DNA-mismatch-repair genes. Int J Cancer 1999; 81: 214–218. | Article | PubMed | ISI | ChemPort | |
| 13. | Phillips JL, Pavlovich CP & Walther M et al. The genetic basis of renal epithelial tumors: Advances in research and its impact on prognosis and therapy. Curr Opin Urol 2001; 11: 463–469 10.1097/00042307-200109000-00003. | Article | PubMed | ChemPort | |
| 14. | Connor M & Ferguson-Smith M. Essential Medical Genetics 1997; Oxford, Blackwell Science. |
| 15. | Zbar B, Tory K & Merino M et al. Hereditary papillary renal cell carcinoma. J Urol 1994; 151: 561–566. | PubMed | ISI | ChemPort | |
| 16. | Schmidt L, Duh FM & Chen F et al. Germline and somatic mutations in the tyrosine kinase domain of the MET proto-oncogene in papillary renal carcinomas. Nat Genet 1997; 16: 68–73. | Article | PubMed | ISI | ChemPort | |
| 17. | Begg CB, Zhang ZF & Sun M et al. Methodology for evaluating the incidence of second primary cancers with application to smoking-related cancers from the Surveillance, Epidemiology, and End Results (SEER) program. Am J Epidemiol 1995; 142: 653–665. | PubMed | ISI | ChemPort | |
| 18. | Rabbani F, Grimaldi G & Russo P. Multiple primary malignancies in renal cell carcinoma. J Urol 1998; 160: 1255–1259. | PubMed | ISI | ChemPort | |
| 19. | Rabbani F, Reuter VE, Katz J & Russo P. Second primary malignancies associated with renal cell carcinoma: Influence of histologic type. Urology 2000; 56: 399–403. | Article | PubMed | ISI | ChemPort | |
| 20. | Greenberg RS, Rustin ED & Clark WS. Risk of genitourinary malignancies after cancer of the prostate. Cancer 1988; 61: 396–401. | PubMed | ISI | ChemPort | |
| 21. | Tihan T & Filippa DA. Coexistence of renal cell carcinoma and malignant lymphoma. A causal relationship or coincidental occurrence? Cancer 1996; 77: 2325–2331. | Article | PubMed | ISI | ChemPort | |
| 22. | Anderson CM, Pusztai L & Palmer JL et al. Coincident renal cell carcinoma and nonHodgkin's lymphoma: The M. D. Anderson experience and review of the literature. J Urol 1998; 159: 714–717. | PubMed | ISI | ChemPort | |
| 23. | Dong C & Hemminki K. Modification of cancer risks in offspring by sibling and parental cancers from 2,112,616 nuclear families. Int J Cancer 2001; 91: 144–150. |
| 24. | Hemminki K, Li X & Collins P. Population-based study of familial central nervous system hemangioblastomas. Neuroepidemiology 2001; 20: 257–261. | Article | PubMed | ISI | ChemPort | |
| 25. | Hemminki K, Li X & Plna K et al. The nation-wide Swedish Family-Cancer Database: Updated structure and familial rates. Acta Oncol 2001; 40: 772–777. | Article | PubMed | ISI | ChemPort | |
| 26. | Esteve J, Benhamou E & Raymond L. Statistical Methods in Cancer Research 1994; Lyon, IARC. |
| 27. | Hemminki K, Vaittinen P, Dong C & Easton D. Sibling risks in cancer: Clues to recessive or X-linked genes? Br J Cancer 2001; 84: 388–391 10.1054/bjoc.2000.1585. | Article | PubMed | ISI | ChemPort | |
| 28. | Rothman K & Greenland S. Modern Epidemiology 1998; Philadelphia, Lippincott-Raven. |
| 29. | Schlehofer B, Pommer W & Mellemgaard A et al. International renal-cell-cancer study. VI. The role of medical and family history. Int J Cancer 1996; 66: 723–726 10.1002/(sici)1097-0215(19960611)66:6<723::aid-ijc2>3.0.co;2-1. | Article | PubMed | ISI | ChemPort | |
| 30. | Kreiger N, Marrett LD & Dodds L et al. Risk factors for renal cell carcinoma: Results of a population-based case-control study. Cancer Causes Control 1993; 4: 101–110. | Article | PubMed | ISI | ChemPort | |
| 31. | Goldgar DE, Easton DF, Cannon-Albright LA & Skolnick MH. Systematic population-based assessment of cancer risk in first-degree relatives of cancer probands. J Natl Cancer Inst 1994; 86: 1600–1607. | PubMed | ChemPort | |
| 32. | Lynch H & Smyrk T. Hereditary nonpolyposis colorectal cancer (Lynch syndrome). Cancer 1996; 78: 1149–1167 10.1002/(sici)1097-0142(19960915)78:6<1149::aid-cncr1>3.3.co;2-q. | Article | PubMed | ISI | ChemPort | |
| 33. | Lindor N, Greene M & The Mayo Familial Cancer Program. A concise handbook of family cancer syndromes. J Natl Cancer Inst 1998; 90: 1039–1071. | Article | PubMed | ChemPort | |
| 34. | Bergfeldt K, Einhorn S, Rosendahl I & Hall P. Increased risk of second primary malignancies in patients with gynecological cancer. Acta Oncol 1995; 34: 771–777. | PubMed | ISI | ChemPort | |
| 35. | Metayer C, Lynch C & Clarke E et al. Second cancers among long-term survivors of Hodgkin's disease diagnosed in childhood and adolescence. J Clin Oncol 2000; 18: 2435–2443. | PubMed | ISI | ChemPort | |
| 36. | Swerdlow A, Barber J & Hudson G et al. Risk of second malignancy after Hodgkin's disease in a collaborative British cohort: The relation to age at treatment. J Clin Oncol 2000; 18: 498–509. | PubMed | ISI | ChemPort | |
