The Clinical Features of Ovarian Cancer in Hereditary Nonpolyposis Colorectal Cancer

Gynecologic Oncology 82, 223–228 (2001) doi:10.1006/gyno.2001.6279, available online at http://www.idealibrary.com on

The Clinical Features of Ovarian Cancer in Hereditary Nonpolyposis Colorectal Cancer Patrice Watson, Ph.D.,* Ralf Bu¨tzow, M.D.,† Henry T. Lynch, M.D.,* Jukka-Pekka Mecklin, M.D.,‡ Heikki J. Ja¨rvinen, M.D.,† Hans F. A. Vasen, M.D., Ph.D.,§ Lisa Madlensky, M.Sc., ¶ Paulo Fidalgo, M.D.,㛳 Inge Bernstein, M.D., Ph.D.,** and the International Collaborative Group on HNPCC *Department of Preventive Medicine, Creighton University School of Medicine, Omaha, Nebraska, 68178; †Helsinki University Central Hospital, Helsinki, Finland; ‡Department of Surgery, Jyva¨skyla¨ Central Hospital, Jyva¨skyla¨, Finland; §Foundation for the Detection of Hereditary Tumors, c/o Lieden University Hospital, Leiden, The Netherlands; ¶Familial GI Cancer Registry, Mount Sinai Hospital, Toronto, Ontario, Canada; 㛳Gatroenterology Department, Portuguese Institute of Oncology, Lisboa, Portugal; and **The Danish HNPCC Register, Department of Surgical Gastroenterology, Hvidovre Hospital, Hvidovre, Denmark Received February 6, 2001

predominance; multiple CRC are common; and improved survival after CRC has been observed [1–3]. HNPCC is also associated with an elevated risk for extracolonic cancers, including carcinoma of the endometrium [4], ovary, stomach, small bowel, hepatobiliary tract, and transitional cell carcinoma of the ureter and renal pelvis [5], as well as certain skin tumors [6 – 8] and brain tumors [9]. Heritable mutations in DNA mismatch repair genes cosegregate with disease in HNPCC families [10 –16]. Testing for these mutations now allows the identification of mutation carriers prior to disease development in HNPCC families [17]. CRC and other tumors from HNPCC patients are characterized by widespread microsatellite instability (MSI) consistent with a defect in mismatch repair [18]. It is now clear that, in HNPCC and in a minority of sporadic cases, CRC results from specific defects in DNA repair [19]. This etiologic pathway may account for the differences which have been observed between HNPCC CRC and most sporadic CRC. HNPCC CRC and sporadic CRC with extensive microsatellite instability are more likely to be located in the proximal colon, to be diploid, and to have poorly differentiated histology, mucoid features, peritumoral lymphocyte infiltration, and Crohn-like reaction [1, 20]. Patients with CRC in HNPCC show better survival than age- and stagematched comparison groups [3, 21–23]. No studies have been done to search for unusual clinicopathologic features in HNPCC ovarian cancers or survival differences between HNPCC and non-HNPCC cases. It has been noted that they occur at strikingly early ages compared to what is typical in the general population [5] and that they occur 3.6 –13 times more frequently than expected, based on general population incidence rates [5, 24, 25]. Our aim was to describe the clinicopathologic features of ovarian cancer in HNPCC family members.

Objective. Hereditary nonpolyposis colorectal cancer (HNPCC) is a hereditary cancer susceptibility disorder associated with a very high risk for carcinoma of the colon and an elevated risk for certain extracolonic cancers including ovarian cancer. Our aim in this study was to describe the clinicopathologic features of ovarian cancer in HNPCC family members. Methods. Members of the International Collaborative Group on HNPCC collected retrospective data on 80 ovarian cancer patients who were members of HNPCC families, including 31 known mutation carriers, 35 presumptive carriers (by colorectal/endometrial cancer status), and 14 at-risk family members. Results. Mean age at diagnosis of ovarian cancer was 42.7. Nonepithelial tumors made up only 6.4% of the cancers, and borderline tumors comprised just 4.1% of the epithelial cancers. Among frankly malignant epithelial cases, most cancers were well or moderately differentiated, and 85% were FIGO stage I or II at diagnosis. Synchronous endometrial cancer was reported in 21.5% of cases. Conclusions. Ovarian cancer in HNPCC differs from ovarian cancer in the general population in several clinically important respects. It occurs at a markedly earlier age. It is more likely to be epithelial. If it is a frankly invasive epithelial cancer, it is more likely to be well or moderately differentiated. HNPCC patients with ovarian cancer are more likely to have a synchronous endometrial cancer than other ovarian cancer patients and are more likely to be diagnosed at an early stage. © 2001 Academic Press Key Words: ovarian cancer; cancer genetics; hereditary nonpolyposis colorectal cancer.

INTRODUCTION Hereditary nonpolyposis colorectal cancer (HNPCC), also referred to as the Lynch syndrome, is one of the more common hereditary cancer syndromes. It is associated with a very high risk for development of colorectal carcinoma (CRC), which is typically diagnosed at an early age and with a proximal colonic 223

0090-8258/01 $35.00 Copyright © 2001 by Academic Press All rights of reproduction in any form reserved.

224

WATSON ET AL.

MATERIALS AND METHODS The International Collaborative Group on HNPCC (ICGHNPCC) is an association of centers involved in the study of, and provision of clinical services to, members of HNPCC families. Questionnaires were mailed to all members of the ICG-HNPCC, requesting information on ovarian cancer patients who were members of HNPCC families. Regarding the diagnosis, the information requested included the date at diagnosis, stage at diagnosis, whether the cancer was histologically verified, histologic type, grade, and the initial treatment. Regarding the patient, it included the date of birth, status at last follow-up, date of last follow-up/death, history of other cancer diagnoses, degree of relationship to closest relative with endometrial/colorectal cancer, and whether the patient had both descendants and sibs/ancestors/collateral relatives with endometrial/colorectal cancer. Regarding the family, the information requested included whether it met the Amsterdam criteria [26], whether an HNPCC-associated mutation had been identified, and, if so, which gene was affected. In cases where the family was known to have a specific HNPCC-associated mutation, the information requested on the patient also included test results for the patient, the degree of relationship to the closest relative with a known mutation by test, and whether the patient had both descendants and sibs/ancestors/collateral relatives with known mutations by test. Most questionnaire items (except dates and sites of cancer other than ovary) were multiple-choice to minimize coding errors. Returned questionnaires were evaluated for eligibility. For this study, eligible cases were those with a histologically verified ovarian tumor and with a family history of HNPCC evidenced by findings of an HNPCC-associated mutation in family members or a family history of cancer meeting the Amsterdam criteria [26]. Cases eligible by these criteria were still excluded if the ovarian tumor was of a benign histologic type or if the patient had been shown by test not to carry the mutation carried by other affected family members in a family with a known HNPCC-associated mutation. Descriptive statistics, including means and proportions and the associated 95% confidence intervals (95% CI), were calculated. Survival estimates were calculated using life table methods. All statistical analysis was performed using SAS statistical software (SAS Institute, Cary, NC). Borderline and nonepithelial ovarian cancers were excluded from the survival analysis. In cases where year rather than date (of diagnosis, birth, last follow-up or death) was available, estimates were made by these rules: dates of birth, last follow-up, and death were estimated as the first day of the given year, and dates of diagnosis were estimated as the last day of the given year. These rules were chosen to bias our estimate of age at diagnosis upward and our estimate of survival time downward (both biases in the opposite direction from the expected trend of younger age at diagnosis and the hypothesized trend toward longer survival). Parameter estimates based on our sample

statistics were judged to indicate significant differences between HNPCC and non-HNPCC ovarian cancer, when the 95% CI calculated from our data excluded published estimates based on the general population. RESULTS Data on 100 patients were submitted from 14 registries in 11 countries. Eighty were eligible for inclusion. The 80 eligible cases included 31 known mutation carriers (by test or inference from tests of offspring and other relatives); among the remaining 49 cases, 35 were presumed to be mutation carriers because either they or their offspring were affected with colorectal or endometrial cancer (4 of these also had a first-degree relative that was a known mutation carrier by test) and 9 had a sibling or parent with colorectal or endometrial cancer (1 of these also had a first-degree relative that was a known mutation carrier by test). In the remaining 5 cases, the closest relative with colorectal or endometrial cancer was a second- or third-degree relative. Forty-four (55%) of the 80 cases were from families with a known mutation, 27 in hMLH1 and 17 in hMSH2. The 20 ineligible cases included 1 benign tumor of the ovary (mature teratoma), 2 reported cancers which were not histologically verified, 1 case which was a proven noncarrier of the HNPCC-associated mutation affecting other members of her family, and 16 cases who were members of families which neither met the Amsterdam criteria nor were known to carry a specific HNPCC-associated mutation. In one case, information on the histology of the ovarian tumor was unavailable. The distribution of histologic types in the remaining cases is given in Table 1. Grade was available on 35 of the frankly invasive epithelial cancers; these data are shown in Table 1. Note that these data are based on diagnoses abstracted from medical records, including pathology reports, from a variety of hospitals and over a wide range of years (1936 –1997; the median year of diagnosis was 1981); many cases were classified as epithelial carcinoma, type unspecified. The mean age at diagnosis of the 80 cases was 42.7 (95% confidence interval [CI] ⫽ 40.7– 44.6). Approximately onethird of the patients were younger than 40 at diagnosis; 85% were younger than 50. The majority of the borderline epithelial carcinoma and nonepithelial carcinoma cases were younger than the overall mean age at diagnosis, but excluding these cases did not markedly affect the mean age; when only frankly invasive epithelial cancers were included, the mean age was 43.4 (95% CI ⫽ 41.38 – 45.46). The range in age of the epithelial carcinomas was 25–76; for the nonepithelial carcinomas it was 19 – 44. Stage information (FIGO stage I, II, III, or IV) was available for 64 of the frankly invasive epithelial cancers. The stage distribution for these cases is shown in Table 2. Lifetable methods were used to calculate proportion surviving after 5 years within each stage category. The percentage surviving was 85% in 39 stage I cases, 52% in 15 stage II cases, and 42%

225

CLINICAL FEATURES OF OVARIAN CANCER IN HNPCC

TABLE 1 Histologic Types in Malignant Ovarian Tumors in HNPCC, N ⴝ 79 (1 Missing) 95% Confidence interval

Comparison cases, percentage

93.7 6.3

88.3–99.1

83.6

71 3

95.9 4.1

91.4–1.0

23 17 7 13 7 4

32.4 23.9 9.9 18.3 9.9 5.6

21.5–43.3 14.0–33.8 3.0–16.8 9.3–27.3 3.0–16.8 0.3–10.9

35.4 29.8 13.4 9.6 3.6 8.2

11 15 10

30.6 41.7 27.8

15.5–45.7 25.6–57.8 13.1–42.4

16.1 25.0 58.9

Histology

Frequency

Percentage

Epithelial Nonepithelial Among epithelial cases Frankly invasive Borderline Among frankly invasive epithelial cases Pap/epithelial nos Serous Mucinous Endometrioid Clear cell Mixed (other) Among frankly invasive epithelial cases Well differentiated Moderately differentiated Poorly and undifferentiated Among nonepithelial cases Granulosa cell Dysgerminoma Sex cord tumor Endodermal sinus tumor

74 5

2 1 1 1

Note. The comparison cases are 21240 cases of invasive ovarian cancer from the US SEER Program, including 10,040 graded cases [28].

in 9 stage III cases. The single stage IV case survived less than 1 year. For 1 of the borderline ovarian cancer cases, survival time was unavailable; the second case is alive after 1 year, and the third case, who had a synchronous stage II endometrial cancer, died after 1.7 years. Two of the nonepithelial ovarian tumors (1 endodermal sinus, 1 granulosa cell) were stage IV, and both patients died within 2 months; the other 3 cases (stage 1 sex cord tumor, stage 1 dysgerminoma, and stage 2 granulosa cell tumor) each survived at least 8 years. Synchronous and/or metachronous (extraovarian primary) cancers were noted in 44 ovarian cancer patients (55%). Eleven

TABLE 2 Stage Distribution in Frankly Invasive Epithelial Ovarian Cancer in HNPCC, N ⴝ 64 FIGO stage Frequency Percentage I II III IV

39 15 9 1

60.9 23.4 14.1 1.6

95% Confidence Comparison cases, interval percentage 48.8–72.9 13.0–33.8 5.6–22.6 0–4.7

25.7 4.0 11.5 58.9

Note. Comparison cases: 21,240 cases of invasive ovarian cancer from the US SEER Program [28]. The proportion of advanced stage cases in HNPCC, 15.6% (95% CI: 6.7–24.5), differs significantly from the value of 70% typical of the general population.

patients had another cancer diagnosis prior to their ovarian cancer (1–20 years prior): 9 of these were colorectal cancers, 1 was an endometrial cancer (13 years prior to the ovarian cancer), and 1 was a breast cancer. Eighteen of the 80 patients (22.5%; 95% CI: 13–32%) had a synchronous endometrial cancer. One patient had synchronous ovarian and colorectal cancer. Two patients had an endometrial cancer diagnosed after the ovarian cancer (3 and 5 years after). Eighteen patients had a subsequent diagnosis of colorectal cancer, and 6 patients had subsequent diagnoses of gastric, small bowel, or urologic tract cancer; the first such cancer per patient occurred from 3 to 22 years after the ovarian cancer. Information about the presentation of the ovarian cancer was available in 71 of the 80 cases. Two cases were reportedly discovered in a screening examination, 7 were found in asymptomatic women having hysterectomy with bilateral oophorectomy, and the remainder (62, 87.3%) were symptomatic. Among the staged, frankly invasive epithelial carcinoma cases, 52 of 58 or 89.7% were symptomatic. The type of symptom was available in 40 cases. The most common symptoms were abdominal pain and abnormal uterine bleeding. Among the cases without a synchronous endometrial carcinoma, pain was cited in 17 of 29 cases (57%) while bleeding was cited in 3 (10%). Among cases with a synchronous endometrial carcinoma, bleeding was cited in 5 of 11 cases (45%) while pain was cited in 3 (27%).

226

WATSON ET AL.

DISCUSSION The ovarian cancer cases included in this study were remarkable in several respects. The mean age at diagnosis, approximately 43, was at least 16 years earlier than the mean age in the general population (U.S.) [27], and since the upper limit of the 95% CI for the mean age was less than 45, this difference is clearly not attributable to sampling errors. The distribution of histologic types differed from the expected distribution in several ways. Nonepithelial tumors made up only 6.4% of all cancers, and 6.7% of all frankly invasive tumors, in our series, which was significantly less that the 16.4% observed in a large series of frankly invasive ovarian tumors [28]. We believe that this reflects a real paucity, relative to epithelial tumors, but these data cannot reveal whether this paucity reflects a smaller elevation in incidence (relative to the general population) for these tumors in HNPCC or no elevation in incidence. Borderline tumors comprised just 4.1% of the epithelial cancers, which is significantly lower than the 15% which is widely cited as the proportion in the general population [29]. It is possible that the paucity of borderline tumors is attributable to misclassification. Cases in this series were diagnosed as early as 1936, and borderline epithelial tumors were not consistently recognized until the early 1970s. In fact, when pre-1975 cancers are excluded, 6.5% of the remaining 46 epithelial malignancies are borderline, which is not significantly lower than 15%. Furthermore, the large proportion of frankly invasive tumors which are well differentiated raises additional concern about misdiagnosis of borderline tumors. Most of the frankly invasive epithelial cases were well or moderately differentiated; only 28% were poorly or undifferentiated, which differs significantly from the 59% observed in the general population. Based on these concerns, we cannot conclude that there is a paucity of borderline tumors, relative to frankly invasive tumors, in HNPCC. However, our data do indicate that borderline tumors are not typical in HNPCC. Even allowing for the possibility that some of the well-differentiated carcinomas are misclassified borderline tumors, we conclude that HNPCC ovarian cancers are less likely to be poorly differentiated cancers than ovarian cancer in the general population. Among the frankly invasive epithelial cases, there was no marked deviation from the distribution of histologic type seen in the general population [28], but there was a modest, marginally significant excess of the endometrioid subtype. It may be that the development of mismatch repair deficiency in the ovarian epithelium favors the development of endometrioid carcinoma. In sporadic ovarian carcinomas of the endometrioid subtype microsatellite instability has been reported at a frequency of 50% [30], whereas only 8% of serous carcinomas showed this phenotype [31]. It is notable that endometrioid carcinoma of the uterus is common in female HNPCC patients and that up to 28% of sporadic uterine endometrioid carcinomas display MSI [32]. Taken together these findings suggest

that genes that are rate-limiting in carcinomas of the mu¨llerian derivatives with endometrioid phenotype are particularly vulnerable to replication errors. Seventeen, or 21.5%, of the ovarian cancer cases were reported to have had a synchronous endometrial cancer. The difficulty of differentiating tumor metastases and separate primary tumors [33] suggests caution in interpreting this figure. Our observations are based on diagnoses by local pathology departments; in most cases we have no information about the basis for the diagnosis. Neither do we have stage, histologic type, or grade information on the synchronous endometrial cancer. We note that if these were actually metastatic tumors rather than separate primary cancers, many would have been understaged, especially since the majority were assigned stage I or II; we would then expect to see poorer survival than that expected for the early stage cancers. In fact, survival by stage was consistent with general population expectations. This makes the possibility that these were erroneously diagnosed as separate primaries seem less likely. It is also difficult to arrive at an appropriate comparison figure for the general population of ovarian cancers. Sheu and colleagues [34] reviewed 421 consecutive ovarian cancer patients and found 11 with synchronous endometrial cancer (2.6%); however, 5 of the 11 were identified as metastatic cancer after further review. If the cases in our series reported to be diagnosed with a synchronous, separate primary endometrial cancer are accurately classified, and if this phenomenon is rare in the general population (less than 5% of ovarian cancers), the marked excess in our series is very unlikely to be due to sampling error. It may reflect something significant about pathogenesis in the female genital tract in HNPCC. The distribution of stage at diagnosis among frankly invasive epithelial ovarian cancer cases, with over 84% of cases classified as FIGO stage I or II, differed significantly from that observed in the general population, where approximately 30% of cases are stage I/II at diagnosis. Given the long time period during which these cases were diagnosed, some of these cases may have been understaged. The practice of performing extended surgical staging with complete examination of the pelvis and abdominal cavity and lymph node biopsies was established in the mid-1980s. Before that time, a significant proportion of patients were understaged. In one large study, approximately one-third of the Stage I cases actually had a more advanced stage [35]. The HNPCC cases we studied came from 1936 to 1997 (median 1981) and the comparison series from 1973 to 1987. Both are likely to contain understaged cases. However, even if one-third of the HNPCC stage I cases were upgraded to stage III, the difference between the groups would still be significant. Overall survival after diagnosis was approximately 69% at 5 years. Within stage groups, observed time to death is not strikingly different from that expected based survival in the general population, where 5-year survivals of 79, 59, and 22% for stages I, II, and III, respectively, have been reported in a

CLINICAL FEATURES OF OVARIAN CANCER IN HNPCC

large recent study [29]. Thus, it appears that the good overall survival in the HNPCC cases was partly or wholly attributable to the favorable stage distribution of these cases. The favorable stage distribution may be related to the young age in most cases, since in the general population older women are more likely to have advanced disease at initial diagnosis than younger women [29]. It might also be attributed to the use of cancer screening and prophylactic surgery as a part of a cancer prevention regimen. Endometrial cancer is common in women with HNPCC [4]. If these cases are more likely to have prophylactic total hysterectomy and bilateral salpingoophorectomy, then they might be more likely to have occult ovarian cancers discovered. Furthermore, screening aimed at detecting endometrial cancer might detect occult ovarian cancer. Some of these women may have adopted a regimen of ovarian screening, as well. However, 87% of the cases in this HNPCC series were reported to be symptomatic at the time of diagnosis, so this is unlikely to account for the stage distribution. Certainly the frequent synchronous endometrial pathology may have produced symptoms which led to the early discovery of these ovarian cancers. Finally, it has been observed that CRC in HNPCC patients has a better prognosis than other CRC [3, 21–23] and also that they are less likely to metastasize [3, 22]. This observation has not been explained, but possibly a related phenomenon affects ovarian carcinogenesis in HNPCC patients. All conclusions drawn from our study must be limited by the nature of the study data, that is, clinical information from medical records. Diagnoses were made over a broad span of years and in a variety of countries. Terminology and standards of judgment used by pathologists will have varied widely. Surgical practice will also have varied widely, impacting the quality of stage information. To decide whether the clinical features observed were different from the general population, we used norms based on the U.S. population rather than on the populations of the countries from which the HNPCC cases were drawn. Ideally, information on the clinical features of any subset of ovarian cancer should be based on cases staged and followed in a standard way and on centrally reviewed pathology specimens, and comparison (non-HNPCC) cases staged and reviewed in the same way, and selected from the same populations. However, given the rarity of HNPCC and the rarity of ovarian cancer in HNPCC, the ideal information is unlikely to be available soon. Clinicians treating women carrying HNPCC-associated mutations face questions about ovarian cancer prevention right now. We conclude that ovarian cancer in HNPCC differs from ovarian cancer in the general population in several clinically significant respects. It occurs at a markedly earlier age: onethird of the cases are before the age of 40 and one-half are between the ages of 40 and 50. It is very likely to be an epithelial cancer. It may be any histological type. If it is a frankly invasive epithelial cancer, it is unlikely to be poorly differentiated or to be advanced stage at diagnosis. HNPCC

227

patients with ovarian cancer are more likely to have a synchronous endometrial cancer than other ovarian cancer patients. These conclusions are tentative, being based on a retrospective review of medical records. Clinical, pathological, and molecular genetic studies of tumor materials are warranted to verify and extend these observations. ACKNOWLEDGMENTS This article was supported by revenue from Nebraska cigarette taxes awarded to Creighton University by the Nebraska Department of Health and Human Services. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the State of Nebraska or the Nebraska Department of Health and Human Services. In addition to the listed authors, the following contributed to this study and manuscript: Miguel Rodriguez-Bigas, M.D., Roswell Park Cancer Institute (Buffalo); Rolf Sijmons, University of Groningen; Judy Ho, M.D., University of Hong Kong; Jae-Gahb Park, M.D., Seoul National University Hospital; Timothy Bishop, Ph.D., ICRF Genetic Epidemiology Laboratory, Leeds; Annika Lindblom, M.D., Karolinska Hospital, Stockholm; Diana Eccles M.D., MCRP, Princess Anne Hospital, Southampton.

REFERENCES 1. Lynch HT, Smyrk T. Hereditary nonpolyposis colorectal cancer (Lynch syndrome): an updated review. Cancer 1996;78:1149 – 67. 2. Lynch HT, Smyrk T, Lynch J. An update of HNPCC (Lynch syndrome). Cancer Genet Cytogenet 1997;93:84 –99. 3. Watson P, Lin K, Rodriguez-Bigas MA, Smyrk T, Lemon S, Shashidharan M, Franklin B, Karr B, Thorson A, Lynch HT. Colorectal carcinoma survival among hereditary nonpolyposis colorectal cancer family members. Cancer 1998;83:259 – 66. 4. Watson P, Vasen HFA, Mecklin J-P, Jarvinen H, Lynch HT. The risk of endometrial cancer in hereditary nonpolyposis colorectal cancer. Am J Med 1994;96:516 –20. 5. Watson P, Lynch HT. Extracolonic cancer in hereditary nonpolyposis colorectal cancer. Cancer 1993;71:677– 85. 6. Lynch HT, Fusaro RM, Roberts L, Voorhees GJ, Lynch JF. Muir–Torre syndrome in several members of a family with a variant of the cancer family syndrome. Br J Dermatol 1985;113:295–301. 7. Lynch HT, Lynch PM, Pester JA, Fusaro RM. Sebaceous neoplasia and visceral cancer (Torre’s syndrome) and its relationship to the cancer family syndrome. In Lynch HT, Fusaro RM, editors. Cancer-associated genodermatoses. New York: Van Nostrand Reinhold, 1982;366 –93. 8. Lynch HT, Fusaro RM. Muir–Torre syndrome: heterogeneity, natural history, diagnosis, and management. Prob Gen Surg 1993;10:1–14. 9. Vasen HFA, Sanders EACM, Taal BG, Nagengast FM, Griffioen G, Menko FH, Kleibeuker JH, Houwing-Duistermaat JS, Meera Khan P. The risk of brain tumours in hereditary non-polyposis colorectal cancer (HNPCC). Int J Cancer 1997;74:551–5. 10. Fishel R, Lescoe MK, Rao MRS, Copeland NG, Jenkins NA, Garber J, Kane M, Kolodner R. The human mutator gene homolog MSH2 and its association with hereditary nonpolyposis colon cancer. Cell 1993;75: 1027–38. 11. Leach FS, Nicolaides NC, Papadopoulos N, Liu B, Jen J, Parsons R, Peltomaki P, Sistonen P, Aaltonen LA, Nystrom-Lahti M, Guan X-Y, Zhang J, Metzler PS, Yu J-W, Kao F-T, Chen DJ, Cerosaletti KM, Fournier REK, Todd S, Lewis T, Leach RJ, Naylor SL, Weissenbach J, Mecklin J-P, Jarvinen H, Petersen GM, Hamilton SR, Green J, Jass J, Watson P, Lynch HT, Trent JM, de la Chapelle A, Kinzler KW, Vo-

228

WATSON ET AL.

gelstein B. Mutations of a mutS homolog in hereditary nonpolyposis colorectal cancer. Cell 1993;75:1215–25. 12. Bronner CE, Baker SM, Morrison PT, Warren G, Smith LG, Lescoe MK, Kane M, Earabino C, Lipford J, Lindblom A, Tannergard P, Bollag RJ, Godwin AR, Ward DC, Nordenskjold M, Fishel R, Kolodner R, Liskay RM. Mutation in the DNA mismatch repair gene homologue hMLH1 is associated with hereditary nonpolyposis colon cancer. Nature 1994;368: 258 – 61.

23.

24.

13. Papadopoulos N, Nicolaides NC, Wei Y-F, Ruben SM, Carter KC, Rosen CA, Haseltine WA, Fleishmann RD, Fraser CM, Adams MD, Venter JC, Hamilton SR, Petersen M, Watson P, Lynch HT, Peltomaki P, Mecklin JP, de la Chapelle A, Kinzler KW, Vogelstein B. Mutation of a mutL homolog in hereditary colon cancer. Science 1994;263:1625–9.

25.

14. Jass JR, Stewart SM, Schroeder D, Lane MR. Screening for hereditary nonpolyposis colorectal cancer in New Zealand. Eur J Gastroenterol Hepatol 1992;4:523–7.

26.

15. Palombo F, Gillinari P, Iaccarino I, Lettieri T, Hughes M, D’Arrigo A, Truong O, Hsuan JJ, Jiricny J. GTBP, a 160-kilodalton protein essential for mismatch-binding activity in human cells. Science 1995;268:1912– 4.

27.

16. Nicolaides NC, Papadopoulos N, Liu B, Wei Y-F, Carter KC, Ruben SM, Rosen CA, Haseltine WA, Fleischmann RD, Fraser CM, Adams MD, Venter JC, Dunlop MG, Hamilton SR, Petersen GM, de la Chapelle A, Vogelstein B, Kinzler KW. Mutations of two PMS homologues in hereditary nonpolyposis colon cancer. Nature 1994;371:75– 80. 17. Burke W, Petersen G, Lynch P, Botkin J, Daly M, Garber J, Ellis Kahn MJ, McTiernan A, Offit K, Thomson E, Varricchio C, for the Cancer Genetics Studies Consortium. Recommendations for follow-up care of individuals with an inherited predisposition to cancer: I. Hereditary nonpolyposis colon cancer. JAMA 1997;277:915–9. 18. Peltomaki P, Lothe RA, Aaltonen LA, Pylkkanen L, Nystrom-Lahti M, Seruca R, David L, Holm R, Ryberg D, Haugen A, Brogger A, Borresen A-L, de la Chapelle A. Microsatellite instability is associated with tumors that characterize the hereditary nonpolyposis colorectal carcinoma syndrome. Cancer Res 1993;53:5853–5. 19. Kinzler KW, Vogelstein B. Lessons from hereditary colon cancer. Cell 1996;87:159 –70. 20. Forster S, Sattler H-P, Hack M, Romanakis K, Rohde V, Seitz G, Wullich B. Microsatellite instability in sporadic carcinomas of the proximal colon: assocation with diploid DNA content, negative protein expression of p53, and distinct histomorphologic features. Surgery 1998;123:13– 8. 21. Lynch HT, Albano W, Recabaren J, Lynch PM, Lynch JF, Elston R. Prolonged survival as a component of hereditary breast and nonpolyposis colon cancer. Med Hypoth 1981;7:1201–9. 22. Myrhoj T, Bisgaard ML, Bernstein I, Svendsen LB, Sondergaard JO, Bu¨low S. Hereditary non-polyposis colorectal cancer: clinical features and

28.

29. 30.

31.

32.

33.

34.

35.

survival: results from the Danish HNPCC Register. Scand J Gastroenterol 1997;32:572– 6. Sankila R, Aaltonen LA, Jarvinen HJ, Mecklin J-P. Better survival rates in patients with MLH1-associated hereditary colorectal cancer. Gastroenterology 1996;110:682–7. Vasen HFA, Wijnen JT, Menko FH, Kleibeuker JH, Taal BG, Griffioen G, Nagengast FM, Meijers-Heijboer EH, Bertario L, Varesco L, Bisgaard M-L, Mohr J, Fodde R, Khan PM. Cancer risk in families with hereditary nonpolyposis colorectal cancer diagnosed by mutation analysis. Gastroenterology 1996;110:1020 –7. Aarnio M, Sankila R, Pukkala E, Salovaara R, Aaltonen LA, de la Chapelle A, Peltoma¨ki P, Mecklin J-P, Ja¨rvinen HJ. Cancer risk in mutation carriers of DNA-mismatch-repair genes. Int J Cancer 1999;81: 214 – 8. Vasen HFA, Mecklin J-P, Meera Khan P, Lynch HT. The International Collaborative Group on Hereditary Nonpolyposis Colorectal Cancer (ICGHNPCC). Dis Colon Rectum 1991;34:424 –5. Kosary CL, Ries LAG, Miller BA, Hankey BF, Harras A, Edwards BK, editors. SEER cancer statistics review, 1973–1992: tables and graphs. Bethesda, MD, National Cancer Institute. 1995: NIH Publ. No. 96-2789. Kosary CL. FIGO stage, histology, histologic grade, age and race as prognostic factors in determining survival for cancers of the female gynecological system: an analysis of 1973– 87 SEER cases of cancers of the endometrium, cervix, ovary, vulva, and vagina. Semin Surg Oncol 1994;10:31– 46. DiSaia PJ, Creasman WT, editors. Clinical gynecologic oncology, 5th ed. St. Louis, Mosby–Year Book, 1997. Fujita M, Enomoto T, Yoshino K, Nomura T, Buzard GS, Inoue M, Okudaira Y. Microsatellite instability and alterations in the hMSH2 gene in human ovarian cancer. Int J Cancer (Pred Oncol) 1995;64:361– 6. King BL, Carcangiu M-L, Carter D, Kiechle M, Pfisterer J, Pfleiderer A, Kacinski BM. Microsatellite instability in ovarian neoplasms. Br J Cancer 1995;72:376 – 82. Catasus L, Machin P, Matias-Guiu X, Prat J. Microsatellite instability in endometrial carcinomas: clinicopathologic correlations in a series of 42 cases. Hum Pathol 1998;29:1160 – 4. Lin WM, Forgacs E, Warshal DP, Yeh IT, Martin JS, Ashfaq R, Muller CY. Loss of heterozygosity and mutational analysis of the PTEN/MMAC1 gene in synchronous endometrial and ovarian carcinomas. Clin Cancer Res 1998;4:2577– 83. Sheu B-C, Lin H-H, Chen C-K, Chao K-H, Shun C-T, Huang S-C. Synchronous primary carcinomas of the endometrium and ovary. Int J Gynecol Obstet 1995;51:141– 6. Young RC, Decker DG, Wharton JT, Piver MS, Sindelar WF, Edwards BK, Smith JP. Staging laparotomy in early ovarian cancer. JAMA 1983; 250:3072– 6.