Decrease in Mortality in Lynch Syndrome Families Because of Surveillance

Article Outline

• Abstract
• Materials and Methods
  • HNPCC Registry
  • Selection of the Study Cohort
  • Data Collection
  • Statistical Methods
• Results
• Discussion
• Acknowledgment
• References
• Copyright

Background & Aims: Lynch syndrome family members have a high risk of developing colorectal (CRC), endometrial (EC), and other cancers. A large-scale surveillance program was introduced in The Netherlands in the late 1980s. The aims of the study were to evaluate the effectiveness of this program by assessing mortality because of CRC and EC before and after 1990 and to compare mortality because of all cancers (except CRC/EC) with mortality in the general population. Methods: Family members with at least 50% probability of being a carrier were selected for the study. The standardized mortality ratio (SMR) because of cancer and the absolute excess risk of death (AER) were calculated. Results: In the total cohort (N = 2788), 445 subjects had died because of cancer. The 3 most frequent causes of cancer-related deaths were CRC (50.3%), EC (6.7%), and brain tumors (6.7%). A significant decrease (70%) in SMR for CRC over time was observed (P < .001); the SMR for EC showed no decreasing trend over time. A significantly increased SMR was found for cancer of the small bowel (SMR = 18.3), brain (SMR = 9.1), kidney/ureter (SMR = 5.9), ovarium (SMR = 2.3), pancreas (SMR = 2.2), and stomach (SMR = 2.1). The AER was significantly increased for brain tumors only. Conclusions: Since the introduction of surveillance, the mortality because of CRC has decreased. Except for brain tumors, we did not find a significantly increased AER for tumors other than CRC/EC.

Abbreviations used in this paper:  AER, absolute excess risk , CRC, colorectal cancer , EC, endometrial cancer , HNPCC, hereditary nonpolyposis colorectal cancer , SMR, standardized mortality ratio

Germ-line mutations in any of 4 DNA mismatch repair (MMR) genes, MLH1,¹ MSH2,² MSH6,³, ⁴ and PMS2,⁵ are responsible for Lynch syndrome, also named hereditary nonpolyposis colorectal cancer (HNPCC). Carriers of an MMR gene mutation in hMLH1 or hMSH2 have a lifetime risk of developing any cancer of 85%–90%.⁶ Colorectal cancer (CRC) and endometrial cancer (EC) are the most frequent cancers in Lynch syndrome, with a cumulative risk of 60%–80% and 30%–50%, respectively.⁷, ⁸ Also, significantly increased risks have been reported for cancers of the stomach, small bowel, upper urological tract (ureter and renal pelvis), ovary, biliary tract, and brain.⁹, ¹⁰

Presymptomatic genetic testing of members of Lynch syndrome families allows the identification of carriers of a mutation. To reduce the risk of CRC in mutation carriers, periodic colonoscopic examination has been recommended or, in selected cases, prophylactic colectomy.⁸, ¹¹ Colonoscopic surveillance of a small set of Lynch syndrome families have demonstrated not only a significant reduction in incidence of CRC but also a reduction in overall mortality.¹¹, ¹² Surveillance of the endometrium by transvaginal ultrasound has also been recommended, but the effectiveness of such a program is yet unknown.¹³, ¹⁴ Surveillance for the other carcinomas of the Lynch syndrome tumor spectrum has only been advised if such cancers run in the family.

In the late 1980s, surveillance of Lynch syndrome families was introduced in The Netherlands on a large scale. The aim of the present study was to evaluate the effectiveness of this program in a large set of Lynch syndrome families, by assessing the mortality because of CRC before and after 1990. The second aim was to compare the mortality because of cancers other than CRC and EC with the mortality in the general population.

Back to Article Outline

Materials and Methods 

HNPCC Registry 

In 1987, a registry of families with Lynch syndrome was established in The Netherlands at The Netherlands Foundation for the Detection of Hereditary Tumors. The main objective of the registry is to promote the early detection of cancer in high-risk families. The detailed approach of the registry has been described elsewhere.¹⁵ In brief, clinical specialists or clinical genetic centers from all parts of The Netherlands refer any family suspected for Lynch syndrome to the registry. In the early years, the genealogic studies were performed by genetic field workers associated with the registry; since the establishment of cancer family clinics in the early 1990s at all university centers in The Netherlands, this analysis is performed mainly at these centers. After confirmation of diagnosis of Lynch syndrome, individual family members have been offered registration. Registration is only possible after written, informed consent. To guarantee the continuity of the surveillance program, periodically, the registry sends letters to the physicians to remind them of the date of the planned colonoscopy (surveillance every 1–2 years, from age 20–25 years) and/or the planned gynecologic examination (tranvaginal ultrasound and determination of tumor marker CA125 every year from age 30–35 years). The specialist will inform the registry about the results of surveillance.

Selection of the Study Cohort 

Registered families were eligible for the study if at least 1 family member was identified with a germ-line mutation in 1 of the MMR genes. On April 1, 2004, 140 families with a known mutation were registered at the Dutch HNPCC Registry, 55 with an hMLH1 mutation, 67 with an hMSH2 mutation, and 18 with an hMSH6 mutation. Two families with both an MMR gene defect and a BRCA1 mutation were excluded from the study. To increase the number of families with an hMSH6 mutation, we included 6 families known at the Department of Medical Clinical Genetics, Leiden University Medical Center, and not yet registered at the National Registry.¹⁶

Within these 140 families with a known mutation, we selected all mutation carriers, obligate mutation carriers, and putative carriers and all their first-degree relatives. Putative carriers were subjects with CRC/EC diagnosed before age 60 years who were not tested. We estimate that the likelihood of being a carrier for these subjects is 80%. We included only subjects who were alive at or after January 1, 1960, and who were born before April 1, 1984. Follow-up before 1960 was ignored to minimize misclassification of cancer diagnoses and because reliable population-specific mortality rates were available from that date onward. We also excluded all individuals born before January 1, 1880. In addition, relatives who were found not to carry the mutation (n = 604) were excluded from the cohort analyses. The final cohort comprised 2788 individuals, of whom 882 were (obligate) mutation carriers, 310 were putative carriers, and 1596 individuals were of unknown carrier status (all first-degree relatives).

Data Collection 

For all individuals in the cohort, basic follow-up information was obtained, including date of birth; date of death; date of last contact if not deceased; sex; type of mutation in family; mutation status; cause of death (code from the International Classification of Diseases [ICD], revision 9);¹⁷ and whether or not the cause of death had been confirmed by pathologic review, pathologist’s report, cancer registry record, clinical record, or death certificate. With respect to the causes of death, we included only causes of death confirmed by either medical or pathologic record. Information on the causes of death based on family history was considered as insufficient (category unknown). Using this approach, the cause of death was lacking in 11% of the cases, which is, in our opinion, acceptable. Because of the extensive genealogic information performed by the HNPCC Registry and because of the selection of being alive at or after January 1, 1960, in all cases, we had adequate information about whether the disease was inherited along the maternal or paternal line. Because of this early starting point of the study, information about the date of births, especially of the family members in the older generations, was lacking in 14% of the cases. If the date of birth was not available (391 family members), the date of birth was estimated from the pedigree structure. Individuals with loss of follow-up information (n = 20) and individuals with unknown sex (n = 40) were excluded. Also information was obtained on participation in the surveillance program. Subjects were divided into 3 groups: (1) those who never attended a surveillance colonoscopy, (2) subjects who underwent at least 1 lifetime surveillance colonoscopy, and (3) subjects for which this information was unclear or unknown.

Statistical Methods 

For all individuals, follow-up started at January 1, 1960, or date of birth, whichever occurred later. Follow-up ended on the date of death, date of last contact, at the age of 80 years, or at the closing date of the study, ie, April 1, 2004, whichever occurred first. The SPSS package 11.1 (SPSS Inc, Chicago, IL) was used for all statistical analyses.

Cancer mortality in the Lynch syndrome cohort was compared with cancer mortality in the general Dutch population by computation of standardized mortality ratio (SMR). The expected number of cancer deaths were computed for each cancer site on the basis of age-, sex-, and calendar period-specific cancer mortality rates of the general population, with the use of a person-years computer program developed at The Netherlands Cancer Institute.¹⁸ The overall ratio of the observed (O) and expected (E) number of deaths in the study population was determined, and the 95% confidence intervals (CI) of the O/E ratio (SMR) were calculated using exact Poisson probabilities of observed numbers. SMR was calculated for all carcinomas and, if significantly elevated and comprising sufficient numbers, stratified by sex and mutated gene. P values for the test of heterogeneity and for tests for trend were calculated according to standard methods.¹⁹ We also calculated the absolute excess risk of death (AER) to evaluate the contribution of specific cancers to excess mortality in Lynch syndrome family members. AER is an appropriate measure because it takes into account the absolute risk of death from a given disease in the general population. From the results of the person-years analysis, the AER was expressed as the observed number of deaths because of a given cancer in our cohort minus the number expected and divided by person-years at risk. AER was calculated per 10,000 person years.

The overall SMR largely depended on the proportion of carriers, putative carriers, and first-degree relatives within this cohort. We also estimated the SMR for mutation carriers using all information of the cohort. We assumed that the probability of being a carrier for putative carriers was 80%. The probability of being a carrier in the first-degree relatives was assumed to be lower than 50% because the tested (non)carriers and the putative carriers were classified in separate groups and therefore were excluded from the remaining group of first-degree relatives. Based on the assumption that all Lynch syndrome family members together (including the noncarriers) have a probability of 50% of being a carrier, we calculated the probability of being a carrier for first-degree relatives with the following formula:

Where N = number of family members, c = carriers, p = putative carriers, nc = noncarriers, FDR = first-degree relatives, and A = probability of being a carrier for the first-degree relatives.

In this formula, A proved to be 35.5%. Next, the observed number of deaths for the different subgroups were multiplied by the weight factor: for carriers with 1, for putative carriers with 1.25 (1/0.8), and for first-degree relatives with 2.82 (1/0.355).

The estimated SMR (SMRest) was calculated with the following formula:

Where SMRest = estimated SMR, E = expected, c = carriers, p = putative carriers, and FDR = first-degree relatives.

Back to Article Outline

Results 

We identified 2788 subjects who fulfilled the selection criteria (Table 1). Thirty-two percent (n = 882, 407 males) of subjects were mutation carriers, 11.1% (n = 310, 143 males) were putative carriers, and 57.2% (n = 1596, 863 males) were first-degree relatives of (putative) carriers. Slightly more females were included in the carrier and putative carrier groups because women with an EC diagnosis might opt for a DNA test and, by definition, EC cases were included in the putative carrier group. Slightly more men were observed in the group of first-degree relatives (54% vs 46%) in comparison with the putative and carrier group. We included 1088 subjects (of 55 families) with an identified hMLH1 mutation in the family, 1351 subjects (of 67 families) with an identified hMSH2 mutation in the family, and 349 subjects (of 24 families) with a known hMSH6 mutation in the family.

Table 1. Descriptive Characteristics of 146 Lynch Families

The exact causes of death were obtained for 89% of the 733 deceased patients. Four hundred forty-five patients died of cancer. Overall, the 5 most frequent causes of cancer-related death were CRC (50.3%), EC (6.7%), brain tumor (6.7%), lung cancer (4.5%), and cancer of the stomach (4.1%). For men (n = 240), the top 3 were CRC (56.3%), lung (7.5%), and brain (7.5%). For women (n = 205), the top 3 were CRC (43.4%), EC (14.6%), and brain (5.9%).

We compared the SMR for all cancers among 3 periods of time: 1960–1974, 1975–1989, and 1990–2004. The SMR for CRC (Table 2) significantly decreased in successive time periods. In contrast, there was no significant difference in the SMR because of EC between the years 1960–1974 and the years 1990–2004. No significant difference was observed for all individual other cancers as well. When comparing the subjects who did (n = 897) or did not have (n = 1073) surveillance colonoscopies, also a significant difference was observed in SMR for CRC. The O/E ratio for the surveillance group was 14/2.2; SMR = 6.5, the O/E ratio for the nonsurveillance group was 200/8.4; SMR = 23.9 (P < .001).

Table 2. Standard Mortality Ratio for Colorectal and Endometrial Cancer in 3 Successive Time Periods

Obs., observed; Exp., expected; NS, not significant; SMR, standardized mortality ratio; CRC, colorectal cancer; EC, endometrial cancer.

a Difference between periods 1960–1975 and 1990–2004.

b Except nonspecified cancers.

When we compared the mortality because of cancer among the 3 types of mutation in the families, only a significant difference was found for CRC. The SMR for CRC was significantly lower in hMSH6 family members than in family members with an hMLH1 or an hMSH2 mutation (Table 3). No statistical difference was observed for mortality because of CRC between hMLH1 (SMR = 17.3) and hMSH2 (SMR = 22.0) family members. A nonstatistical difference was observed in SMR for mortality because of stomach cancer between hMLH1 (SMR = 1.2) and hMSH2 (SMR = 3.5) family members (P = .09).

Table 3. Standard Mortality Ratio for Colorectal and Endometrial Cancer According to Type of MMR-Gene Mutation

Obs., observed; Exp, expected; SMR, standardized mortality ratio; MMR, mismatch repair; NS, not significant; CRC, colorectal cancer; EC, endometrial cancer.

a Except nonspecified cancers.

Comparison of the mortality between males and females showed for females a higher SMR for all cancers (P = .04). The SMR for CRC was lower in females than in males (SMR = 15.3 vs 20.5; P = .04). When taking EC and CRC together, the SMR is higher for females, but this difference was not significant (Table 4). No significant sex differences were found for all other causes of death.

Table 4. Standard Mortality Ratio for Colorectal and Endometrial Cancer According to Sex

Obs, observed; Exp, expected; NS, not significant; SMR, standardized mortality ratio; CRC, colorectal cancer; EC, endometrial cancer.

a Except nonspecified cancers.

Table 5 shows the observed and expected numbers of cancers with the SMRs and AERs. Significantly elevated SMRs were found for cancer of the small bowel (SMR = 18.3), liver/bile duct (SMR = 15.3), brain (SMR = 9.1), kidney/ureter (SMR = 5.9), ovarium (SMR = 2.3), pancreas (SMR = 2.2), and stomach (SMR = 2.1). Because of the low background risk of death due to small bowel cancer, the AER was below 1. Only mortality because of brain tumors (2.9 per 10,000 person-years) reached a substantial absolute excess risk.

Table 5. Observed and Expected Numbers of Cancers, Standardized Mortality Ratios, and Absolute Excess Risks in Members of 146 Families With Lynch Syndrome

NOTE. AER, per 100,000 cancers.

Obs., observed; Exp., expected; SMR, standardized mortality ratio; AER, absolute excess risks.

The SMRest for mutation carriers was markedly increased for liver cancer, brain tumor, small bowel cancer, and carcinoma of the kidney/ureter (Table 6). Of the 30 subjects who died from a brain tumor, only 20 had primary tumors confirmed by a pathology report. The mean age of death was 38.3 years, range 1.9–78.6 years. Results of pathology were gliomas (n = 9), astrocytomas (n = 8), oligogendrogliomas (n = 2), and ependymoma (n = 1). The remaining 10 carcinomas were confirmed by medical record or death certificate, but it was unknown whether the tumors were primaries or distant metastases. The mean age of death of these 10 patients was 50.2 years, range 19.0–76.1. The subjects who died from liver cancer deceased at an average age of 57.6 years (range, 29.5–77.8 years). In 6 of the 15 subjects, the liver/intrahepatic bile duct was established to be the primary location of cancer. In the other cases, metastases could not be excluded.

Table 6. Estimated Standardized Mortality Ratios for Various Types of Cancer in Mutation Carriers

SMRest, estimated standardized mortality ratio.

Back to Article Outline

Discussion 

The present report is a study on cancer mortality in a large series of families with Lynch syndrome, harboring a mutation in one of the mismatch repair genes. The most frequent causes of death in the total cohort were CRC, EC, brain tumor, lung cancer, and cancer of the stomach. A significant decrease in SMR because of CRC was observed in the period 1990–2004 compared with the period 1960–1975 and for those adhering to a colonoscopic surveillance program. Over all periods, the SMR was significantly increased for cancer of the small bowel, brain, kidney, ovarium, pancreas, and stomach.

A concern in this study is that the observed excess cancer mortality risk in carriers may be the result of selection of families for the occurrence of cancers. Most families were selected on the basis of clustering of CRC (Amsterdam I Criteria)²⁰ and, since 1999, also on the basis of clustering of CRC, EC, and other cancers (Amsterdam II criteria).²¹ Therefore, rather than examine the magnitude of the elevated CRC and EC mortality as such, we compared the CRC and EC mortality among subgroups according to time period, sex, and type of mutation. On the other hand, we expect that the number of families that were selected on the basis of other Lynch syndrome-associated tumors is very low. Therefore, we assume that bias does not play a significant role in the observed mortality because of these (non-CRC/EC) cancers.

Another possible selection bias arises when affected (in contrast to nonaffected) subjects would preferably decide to undergo genetic testing. To address this issue, we evaluated whether there was a difference in the proportion of individuals who underwent genetic testing among affected (other than CRC/EC <age 60 years) and nonaffected individuals (67 vs 2045 individuals). No significant difference was found between both groups.

In the current study, we also included patients who died from a second tumor. Because it is not excluded that second tumors are caused by treatment of the first tumor, inclusion of these cases might result in a slight overestimation of the genetic effect.

Already in the late 1980s, a national registry for Lynch syndrome families was established in The Netherlands. The main objective of this registry is to promote surveillance in these families. The finding that the SMR because of CRC significantly decreased since 1990 might suggest that this program is effective. This assumption is confirmed by the observed difference in CRC mortality between subjects who attended the surveillance program and subjects who never did. A concern in this study is that the observed decreased CRC mortality over time periods may not be the result of screening only. Assuming that a Lynch syndrome family has a mean family-specific elevated CRC risk with some random fluctuation over time, it is possible that the family was selected for screening in a period with higher than the mean elevated risk. If the risk returns to the mean risk during follow-up, this regression to the mean effect should not be misinterpreted as the real screening effect. Thus, the real screening effect may be smaller than the declining trend in mortality over the 3 time periods would suggest. One study investigated whether successive generations of HNPCC families had earlier onset of disease.²² A trend was observed toward a higher incidence in successive generations. With this theory, the regression-to-the-mean effect is negligible, and the screening effect might even be underestimated.

There was a slight decrease in EC mortality in the time period 1990–2004 compared with the time period 1960–1975, but the difference was not significant. This might suggest that the surveillance procedures for EC are of limited value. Another reason might be the relatively low numbers of EC-related deaths. It is known that, despite its high incidence in Lynch syndrome, this tumor is an infrequent cause of death.²³, ²⁴ The mortality because of other cancers did not show significant differences among the 3 time-periods. Despite the large series of Lynch syndrome families in the present study, the mortality because of these cancers is probably too low to observe differences.

A striking finding in the present study was that there was no difference in EC mortality between relatives from hMLH1/hMSH2 and hMSH6 families despite the reported higher risk of EC in MSH6 mutation carriers compared with the risk in MLH1 and MSH2 mutation carriers.¹⁶ This might be explained by the relatively good prognosis of patients with EC. A large difference was observed for CRC mortality between MLH1/MSH2 and MSH6 relatives, which is in agreement with the higher cumulative CRC risk reported in hMLH1/MSH2 mutation carriers.¹⁶

The SMR because of CRC was significantly higher in males than in females. This finding is in agreement with the higher incidence of CRC in males than in females in Lynch syndrome families.⁶, ¹⁶ The SMR for all cancers is higher in females than in males. This indicates that EC and the other associated cancers play a significant role in cancer mortality in female mutation carriers.

Among all tumors other than CRC and EC, the SMR for small bowel tumors was the highest. However, the clinical relevance of this observed/expected ratio is limited because small bowel tumors are a very rare cause of death in the general population. A better measure is AER because it takes into account the absolute risk of death from a given disease in the population. The AER for small bowel cancer was relatively low, as expected.

A significant increased SMR as well as a high AER was observed for brain tumors. The SMR might be overestimated because it is not excluded that some of the tumors are metastases of other primaries. However, in 20 of the 30 tumors, we were able to confirm the presence of a primary brain tumor by pathologic reports. In the other patients, the diagnosis was confirmed by medical report or death certificate. In these cases, it is not excluded that the brain tumor was a metastasis of colorectal cancer. However, metastasis of colorectal cancer to the brain is rare (<5%),²⁵ and, in only 3 of the 10 cases, there was a personal history of CRC. An intriguing question is whether surveillance for brain tumors would be effective in reducing the mortality from this tumor.

A remarkable finding was the significantly increased SMR for liver cancer. Although part of the tumors might be because of metastases of a primary colorectal tumor, at least 6 of the 15 tumors were proven primary tumors, 2 of the intrahepatic bile duct and 4 of the liver. If we take only these cancers into account, the SMR would still be strongly increased.⁶, ¹³

Interestingly, significant low SMRs for lung and breast cancer were observed. One explanation for the lower SMR for lung cancer could be a difference in smoking habits. It is possible that Lynch syndrome family members, because they are aware of their higher risk of cancer, are more health conscious and smoke less than the general population. One Dutch study reported indeed a lower frequency (approximately 20%) of smokers in the Lynch syndrome study population²⁶ compared with the frequency (30.8%) observed in the general Dutch population (www.cbs.nl). Previous studies on the incidence of breast cancer showed no differences between Lynch syndrome families and the general population.¹⁰, ²⁷ The explanation for the lower SMR for breast cancer is probably due to concurrent mortality.

In conclusion, although the observed decrease in CRC mortality since the start of large-scale surveillance programs is encouraging, the mortality because of this cancer is still relatively high. Long-term studies are needed to show that the intensive surveillance program will lead to a further decrease of the CRC mortality.

Unfortunately, we did not demonstrate a significant decrease in EC mortality over time. However, because the expected difference in mortality would be small, we probably need a longer observation time to show a positive effect of surveillance.

With respect to the other tumors, both a high SMR and AER were observed for brain tumors. For the other tumors with an increased SMR, we did not find a significant AER, providing no reason to extend the current surveillance program.

Back to Article Outline

 

The authors thank Willem Klokman for his help with the person-years computer program.

Back to Article Outline

References 

1. Bronner CE, Baker SM, Morrison PT, Warren G, Smith LG, Lescoe MK, et al. Mutation in the DNA mismatch repair gene homologue hMLH1 is associated with hereditary non-polyposis colon cancer . Nature . 1994;368:258–261
   • View In Article
   • MEDLINE
   • CrossRef
2. Fishel R, Lescoe MK, Rao MR, Copeland NG, Jenkins NA, Garber J, et al. The human mutator gene homolog MSH2 and its association with hereditary nonpolyposis colon cancer . Cell . 1993;75:1027–1038
   • View In Article
   • MEDLINE
   • CrossRef
3. Miyaki M, Konishi M, Tanaka K, Kikuchi-Yanoshita R, Muraoka M, Yasuno M, et al. Germline mutation of MSH6 as the cause of hereditary nonpolyposis colorectal cancer . Nat Genet . 1997;17:271–272
   • View In Article
   • MEDLINE
   • CrossRef
4. Akiyama Y, Sato H, Yamada T, Nagasaki H, Tsuchiya A, Abe R, et al. Germ-line mutation of the hMSH6/GTBP gene in an atypical hereditary nonpolyposis colorectal cancer kindred . Cancer Res . 1997;57:3920–3923
   • View In Article
   • MEDLINE
5. Nicolaides NC, Papadopoulos N, Liu B, Wei YF, Carter KC, Ruben SM, et al. Mutations of two PMS homologues in hereditary nonpolyposis colon cancer . Nature . 1994;371:75–80
   • View In Article
   • MEDLINE
   • CrossRef
6. Vasen HF, Stormorken A, Menko FH, Nagengast FM, Kleibeuker JH, Griffioen G, et al. MSH2 mutation carriers are at higher risk of cancer than MLH1 mutation carriers (a study of hereditary nonpolyposis colorectal cancer families) . J Clin Oncol . 2001;19:4074–4080
   • View In Article
   • MEDLINE
7. Aarnio M, Sankila R, Pukkala E, Salovaara R, Aaltonen LA, De La CA, et al. Cancer risk in mutation carriers of DNA-mismatch-repair genes . Int J Cancer . 1999;81:214–218
   • View In Article
   • MEDLINE
   • CrossRef
8. Vasen HF, Wijnen JT, Menko FH, Kleibeuker JH, Taal BG, Griffioen G, et al. Cancer risk in families with hereditary nonpolyposis colorectal cancer diagnosed by mutation analysis . Gastroenterology . 1996;110:1020–1027
   • View In Article
   • Abstract
   • Full-Text PDF (93 KB)
   • CrossRef
9. Vasen HF, Sanders EA, Taal BG, Nagengast FM, Griffioen G, Menko FH, et al. The risk of brain tumours in hereditary non-polyposis colorectal cancer (HNPCC) . Int J Cancer . 1996;65:422–425
   • View In Article
   • MEDLINE
   • CrossRef
10. Watson P , Lynch HT . Extracolonic cancer in hereditary nonpolyposis colorectal cancer . Cancer . 1993;71:677–685
    • View In Article
    • MEDLINE
    • CrossRef
11. Jarvinen HJ , Mecklin JP , Sistonen P . Screening reduces colorectal cancer rate in families with hereditary nonpolyposis colorectal cancer . Gastroenterology . 1995;108:1405–1411
    • View In Article
    • Abstract
    • Full-Text PDF (674 KB)
    • CrossRef
12. Jarvinen HJ, Aarnio M, Mustonen H, Aktan-Collan K, Aaltonen LA, Peltomaki P, et al. Controlled 15-year trial on screening for colorectal cancer in families with hereditary nonpolyposis colorectal cancer . Gastroenterology . 2000;118:829–834
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (109 KB)
    • CrossRef
13. Rijcken FE , Mourits MJ , Kleibeuker JH , Hollema H , van der Zee AG . Gynecologic screening in hereditary nonpolyposis colorectal cancer . Gynecol Oncol . 2003;91:74–80
    • View In Article
    • MEDLINE
    • CrossRef
14. Dove-Edwin I, Boks D, Goff S, Kenter GG, Carpenter R, Vasen HF, et al. The outcome of endometrial carcinoma surveillance by ultrasound scan in women at risk of hereditary nonpolyposis colorectal carcinoma and familial colorectal carcinoma . Cancer . 2002;94:1708–1712
    • View In Article
    • MEDLINE
    • CrossRef
15. Vasen HF , Hartog Jager FC , Menko FH , Nagengast FM . Screening for hereditary non-polyposis colorectal cancer (a study of 22 kindreds in The Netherlands) . Am J Med . 1989;86:278–281
    • View In Article
    • Abstract
    • Full-Text PDF (513 KB)
    • CrossRef
16. Hendriks YM, Wagner A, Morreau H, Menko F, Stormorken A, Quehenberger F, et al. Cancer risk in hereditary nonpolyposis colorectal cancer due to MSH6 mutations (impact on counseling and surveillance) . Gastroenterology . 2004;127:17–25
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (146 KB)
    • CrossRef
17. World Health Organization (WHO). International Classification of Diseases; manual of the international statistical classification of diseases, injuries and causes of death. Volume I. Geneva, Switzerland.
    • View In Article
18. van Leeuwen FE, Klokman WJ, Hagenbeek A, Noyon R, van den Belt-Dusebout AW, van Kerkhoff EH, et al. Second cancer risk following Hodgkin’s disease (a 20-year follow-up study) . J Clin Oncol . 1994;12:312–325
    • View In Article
    • MEDLINE
19. Breslow NE , Day NE . Statistical methods in cancer research. Volume II—The design and analysis of cohort studies . IARC Sci Publ . 1987;1–406
    • View In Article
20. Vasen HF , Mecklin JP , Khan PM , Lynch HT . The International Collaborative Group on HNPCC . Anticancer Res . 1994;14:1661–1664
    • View In Article
    • MEDLINE
21. Vasen HF , Watson P , Mecklin JP , Lynch HT . New clinical criteria for hereditary nonpolyposis colorectal cancer (HNPCC, Lynch syndrome) proposed by the International Collaborative Group on HNPCC . Gastroenterology . 1999;116:1453–1456
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (32 KB)
    • CrossRef
22. Voskuil DW , Vasen HF , Kampman E , van’t Veer P . Colorectal cancer risk in HNPCC families (development during lifetime and in successive generations. National Collaborative Group on HNPCC) . Int J Cancer . 1997;72:205–209
    • View In Article
    • MEDLINE
    • CrossRef
23. Vasen HF, Watson P, Mecklin JP, Jass JR, Green JS, Nomizu T, et al. The epidemiology of endometrial cancer in hereditary nonpolyposis colorectal cancer . Anticancer Res . 1994;14:1675–1678
    • View In Article
    • MEDLINE
24. Boks DE , Trujillo AP , Voogd AC , Morreau H , Kenter GG , Vasen HF . Survival analysis of endometrial carcinoma associated with hereditary nonpolyposis colorectal cancer . Int J Cancer . 2002;102:198–200
    • View In Article
    • MEDLINE
    • CrossRef
25. Sundermeyer ML , Meropol NJ , Rogatko A , Wang H , Cohen SJ . Changing patterns of bone and brain metastases in patients with colorectal cancer . Clin Colorectal Cancer . 2005;5:108–113
    • View In Article
    • Abstract
    • Full-Text PDF (478 KB)
    • CrossRef
26. Diergaarde B , Vrieling A , van Kraats AA , van Muijen GNP , Kok FJ , Kampman E . Environmental factors, truncating APC mutations, and HNPCC/associated colorectal tumors . In: Diet, lifestyle, and molecular alterations that drive colorectal carcinogenesis. Thesis . 2004;p. 91–112
    • View In Article
27. Vasen HF , Morreau H , Nortier JW . Is breast cancer part of the tumor spectrum of hereditary nonpolyposis colorectal cancer? . Am J Hum Genet . 2001;68:1533–1535
    • View In Article
    • MEDLINE
    • CrossRef