Search

The objective of this study was to estimate the lifetime risk of breast cancer in women with a BRCA1 or BRCA2 mutation with and without at least 1 first-degree relative with breast cancer. A total of 2835 women with a BRCA1 or BRCA2 mutation were followed. Age- and gene-specific breast cancer rates were calculated. The relative risks of breast cancer for subjects with a family history of breast cancer, compared to no family history were calculated. The mean age at baseline was 41.1 years, and they were followed for a mean of 6.0 years. The estimated penetrance of breast cancer to age 80 years was 60.8% for BRCA1 and 63.1% for BRCA2. For all BRCA carriers, the penetrance of breast cancer to age 80 for those with no first-degree relative with breast cancer was 60.4% and 63.3% for those with at least 1 first-degree relative with breast cancer. The risk of breast cancer for BRCA carriers with no first-degree relative with breast cancer is substantial, and as a result, clinical management for these women should be the same as those for women with an affected relative., (© 2017 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd.)

Background: Non-epithelial gonadal tumours largely comprise sex cord-stromal tumours (SCSTs) and germ cell tumours (GCTs). Specific somatic mutations in DICER1, a microRNA maturation pathway gene, have been identified in these tumours. We conducted a study that aimed to confirm, refine and extend the previous observations., Methods: We used Sanger sequencing to sequence the RNase IIIa and IIIb domains of DICER1 in 154 gonadal tumours from 135 females and 19 males, as well as 43 extra-gonadal GCTs from 26 females and 17 males., Results: We identified heterozygous non-synonymous mutations in the RNase IIIb domain of DICER1 in 14/197 non-epithelial tumours (7.1%). Mutations were found in 9/28 SCSTs (32%), 5/118 gonadal GCTs (4.2%), 0/43 extra-gonadal GCTs and 0/8 miscellaneous tumours. The 14 mutations affected only five residues: E1705, D1709, E1788, D1810 and E1813. In all five patients where matched and constitutional DNA was available, the mutations were only somatic. There were no mutations found in the RNase IIIa domain., Conclusion: More than half (8/15) of Sertoli-Leydig cell tumours (SLCTs) harbour DICER1 mutations in the RNase IIIb domain, while mutations are rarely found in GCTs. Genetic alterations in SLCTs may aid in classification and provide new approaches to therapy.

Background: Recent data show that mutations in RAD51D have an aetiological role in ovarian carcinoma, yet mutations do not appear to be associated with an increased risk for breast cancer. We studied ovarian and breast cancer families having at least one woman affected by ovarian carcinoma, to assess the importance of RAD51D mutations in such families., Methods: The coding region of the RAD51D gene was analysed in 175 BRCA1/2-negative families with family histories of both ovarian and breast cancer ascertained from two Canadian and two Belgian institutions., Results: We identified one previously reported deleterious mutation, p.Arg186(*) (c.556C>T), and two novel variants; missense substitution p.Cys119Arg and an intronic variant c.83-26A>G. p.Arg186(*) segregated with the disease in the family and two ovarian carcinomas available for analysis showed loss of the wild-type allele, but the novel variants are likely neutral., Conclusion: RAD51D should be included in genetic screening of ovarian cancer families that do not have BRCA1/BRCA2 mutations. We show that mutations are more likely to be found in families with two or more ovarian cancers, or in probands with first-degree relatives with ovarian cancer, and we feel testing should be preferentially offered to affected women from such families.

Background: An increased lymphoma risk is well documented in systemic lupus (SLE). Less attention has been focused on women's cancers, even though SLE affects mostly females. Our objective was to estimate the risk of breast, ovarian, and endometrial cancers in SLE, relative to the general population., Methods: Data were included from five recent studies of large SLE cohorts. The number of cancers observed was determined for each cancer type. The expected number of malignancies was ascertained from general population data. The parameter of interest was the standardised incidence ratio (SIR), the ratio of observed to expected malignancies., Results: The five studies included 47,325 SLE patients (42,171 females) observed for 282,553 patient years. There were 376 breast cancers, 66 endometrial cancers, and 44 ovarian cancers. The total number of cancers observed was less than that expected, with SIRs of 0.76 (95% CI: 0.69, 0.85) for breast cancer, 0.71 (95% CI: 0.55, 0.91) for endometrial cancer, and 0.66 (95% CI: 0.49, 0.90) for ovarian cancer., Conclusions: Data strongly support a decreased risk of breast, ovarian, and endometrial cancers in SLE. This may be due to inherent differences in women in SLE (vs the general population) regarding endogenous oestrogen, other medications, and/or genetic make-up.

The BRCAPRO, Couch, Myriad I and II, Ontario Family History Assessment Tool (FHAT), and Manchester models have been used to predict BRCA1 or BRCA2 mutation carrier status of women at high risk for developing the heritable form of breast and ovarian cancers. We have evaluated these models for their accuracy in classifying 224 French Canadian families with at least three cases of breast cancer (diagnosed before the age of 65 years), ovarian cancer, or male breast cancer where mutation status was known for an index affected case used to assess the model. This series includes 44 BRCA1 and 52 BRCA2 mutation-positive families. Using receiver operator characteristics analyses, the C-statistics were found to be 0.81, 0.80, 0.79, and 0.74 for the BRCAPRO, FHAT, Manchester, and Myriad II models, respectively, when incorporating both BRCA1 and BRCA2 mutation carrier predictions. For the BRCAPRO model, 75% scored greater than a 0.43 probability in the mutation-positive group and 75% scored less than 0.50 in the mutation-negative group. Only 38 of 128 (30%) mutation-negative group had a probability greater than 0.43 with the BRCAPRO model. While all models were highly predictive of carrier status, the BRCAPRO model was the most accurate where a cut-off of 10% would have eliminated 60 of 128 (47%) mutation-negative families for genetic testing and only miss 10 of 96 (10%) mutation-positive families. A review of the cancer phenotypes with high BRCAPRO probabilities showed that significantly more metachronous bilateral breast cancer cases occurred in BRCA1/2 mutation carrier families in comparison to mutation-negative families, a feature which is not discriminated in the BRCAPRO model.

Oral contraceptives have been shown to be protective against hereditary ovarian cancer. The variant progesterone receptor allele named PROGINS is characterized by an Alu insertion into intron G and two additional mutations in exons 4 and 5. The PROGINS allele codes for a progesterone receptor with increased stability and increased hormone-induced transcriptional activity. We studied the role of the PROGINS allele as a modifying gene in hereditary breast and ovarian cancer. The study included 195 BRCA1 and BRCA2 carriers with a prior diagnosis of ovarian cancer, 392 carriers with a diagnosis of breast cancer and 249 carriers with neither cancer. Fifty-eight women had both forms of cancer. Five hundred and ninety-five women had a BRCA1 mutation and 183 women had a BRCA2 mutation. Overall, there was no association between disease status and the presence of the PROGINS allele. Information on oral contraception use was available for 663 of the 778 carriers of BRCA1 or BRCA2 mutations. Among the 449 subjects with a history of oral contraceptive use (74 cases and 365 controls), no modifying effect of PROGINS was observed [odds ratio (OR) 0.8; 95% confidence interval (CI) 0.5-1.3]. Among the 214 carriers with no past exposure to oral contraceptives, the presence of one or more PROGINS alleles was associated with an OR of 2.4 for ovarian cancer, compared to women without ovarian cancer and with no PROGINS allele (P = 0.004; 95% CI 1.4-4.3). The association was present after adjustment for ethnic group and for year of birth.

Background: The authors investigated the acceptability to women of the preventive strategies available for dealing with hereditary breast/ovarian carcinoma in France, the United Kingdom, and Canada, countries selected because of their cultural differences. The authors aimed to discover the existence of specific factors that may affect acceptability of these preventative measures., Methods: A cross-sectional, multicenter survey was conducted in Marseilles, France (n = 141), in Manchester, England (n = 130), and in Montreal, Quebec (n = 84). All of the women attending cancer genetic clinics for the first time because of a family history of breast-ovarian carcinoma completed a self-administered questionnaire before their clinic consultation., Results: Variations in responses to different preventative options presented on the questionnaire were seen within the sample of patients considered as a whole. The highest levels of acceptability were obtained for mammographic screening (87%) and chemoprevention (58%). In contrast, prophylactic oophorectomy and prophylactic mastectomy were thought to be acceptable at an early age (before 35 years), an age associated with the highest prophylactic efficacy, by only 19% and 16% of the respondents, respectively. After multivariate adjustment, the statistical data showed that the British respondents were more in favor of oophorectomy (P < 0.0001) and more in favor of chemoprevention than the French (P < 0.001) and the Canadian respondents (P < 0.001). The British (overall adjusted response [OR(adj)] = 3.9; P < 0.001) and Canadian respondents (OR(adj) = 3; P < 0.01) were more in favor of prophylactic mastectomy than the French. The cumulated acceptability of mammography before the age of 40 years was found to be greater in the French (OR(adj) = 2.8; P < 0.01) and Canadian (OR(adj) = 3.1; P < 0.05) samples than in the British sample., Conclusion: These results demonstrated the existence of international variations in the acceptability of the preventive strategies available for women at risk for hereditary breast/ovarian carcinoma. Therefore, these results suggested that when establishing medical recommendations or planning public health interventions, physicians must integrate the population's perception of advantages and drawbacks with the patient's individual decision making., (Copyright 2001 American Cancer Society.)

Germline mutations in the LKB1 (STK11) gene (chromosome sub-band 19p13.3) cause characteristic hamartomas and pigmentation to develop in patients with Peutz-Jeghers syndrome. Peutz-Jeghers syndrome carries an overall risk of cancer that may be up to 20 times that of the general population and Peutz-Jeghers patients are at increased risk of benign and malignant ovarian tumours, particularly granulosa cell tumours. Loss of heterozygosity (allele loss, LOH) has been reported in about 50% of ovarian cancers on 19p13.3. LKB1 is therefore a candidate tumour suppressor gene for sporadic ovarian tumours. We found allele loss at the marker D19S886 (19p13.3) in 12 of 49 (24%) sporadic ovarian adenocarcinomas. Using SSCP analysis, we screened ten ovarian cancers with LOH, 35 other ovarian cancers and 12 granulosa cell tumours of the ovary for somatic mutations in LKB1. No variants were detected in any of the adenocarcinomas. Two mutations were detected in one of the granulosa cell tumours: a mis-sense mutation affecting the putative 'start' codon (ATG --> ACG, M1T); and a silent change in exon 7 (CTT --> CTA, leucine). Like BRCA1 and BRCA2, therefore, it appears that LKB1 mutations can cause ovarian tumours when present in the germline, but occur rarely in the soma. The allele loss on 19p13.3 in ovarian cancers almost certainly targets a different gene from LKB1.

We have identified four mutations in each of the breast cancer-susceptibility genes, BRCA1 and BRCA2, in French Canadian breast cancer and breast/ovarian cancer families from Quebec. To identify founder effects, we examined independently ascertained French Canadian cancer families for the distribution of these eight mutations. Mutations were found in 41 of 97 families. Six of eight mutations were observed at least twice. The BRCA1 C4446T mutation was the most common mutation found, followed by the BRCA2 8765delAG mutation. Together, these mutations were found in 28 of 41 families identified to have a mutation. The odds of detection of any of the four BRCA1 mutations was 18.7x greater if one or more cases of ovarian cancer were also present in the family. The odds of detection of any of the four BRCA2 mutations was 5.3x greater if there were at least five cases of breast cancer in the family. Interestingly, the presence of a breast cancer case <36 years of age was strongly predictive of the presence of any of the eight mutations screened. Carriers of the same mutation, from different families, shared similar haplotypes, indicating that the mutant alleles were likely to be identical by descent for a mutation in the founder population. The identification of common BRCA1 and BRCA2 mutations will facilitate carrier detection in French Canadian breast cancer and breast/ovarian cancer families.

Objective: The objective was to compare risk factors between familial and sporadic ovarian cancer by means of a case-control approach., Study Design: We conducted a case-control study among French Canadian women in Montreal during 1995-1996. One hundred seventy women 20 to 84 years old with histologically confirmed diagnoses of primary ovarian carcinomas or borderline tumors were interviewed concerning their reproductive, family, and medical histories. During the same period 170 randomly selected population control subjects, frequency-matched to the case patients according to age and ethnic group, were also interviewed. Unconditional logistic regression methods were used for data analysis., Results: The major factors influencing the risk of development of ovarian cancer were as follows: (1) family history of breast or ovarian cancer, (2) a late age at use of oral contraceptives (a protective effect), and (3) a late age at last childbirth (a protective effect for familial case patients only)., Conclusion: These factors had equally great impacts in familial and sporadic cases, implying that the underlying mechanisms of carcinogenesis in sporadic and familial ovarian cancer may be similar and that hereditary ovarian cancer may be preventable.

It is not clear if hereditary site-specific ovarian cancer exists as a genetic entity distinct from the hereditary breast-ovarian cancer syndrome. We have identified a large Ashkenazi Jewish kindred with 8 cases of ovarian carcinoma and no cases of breast cancer. Initially, linkage analysis for this kindred generated a negative LOD score to BRCA1, but subsequent mutation and haplotype analysis of key individuals demonstrated a BRCA1 185delAG mutation segregating with all but 1 of the ovarian cancer cases. This observation has important implications for genetic counselling of families with site-specific ovarian cancer. Hereditary site-specific ovarian cancer is likely to be a variant of the hereditary breast-ovarian cancer syndrome, attributable to either BRCA1 or BRCA2. We consider women from these families to be at increased risk of breast cancer and counsel them accordingly.

In breast cancers and sarcomas, a variant polymorphism in the cell cycle inhibitor P21CIP1/WAF1 is under-represented in those individuals whose tumours contain mutated TP53. The aim of this study was to determine whether this variant polymorphism was also under-represented in those with ovarian carcinoma and TP53 mutations. We studied lymphocyte DNA from 104 women with ovarian carcinoma, 15 with borderline tumours and 16 with benign tumours, using a previously-reported PCR-RFLP technique. 96 of the ovarian carcinoma cases had been previously examined for mutations in TP53 and/or for overexpression of the TP53 protein. The variant allele was seen in 11 out of 104 women (10.6%) with ovarian carcinoma. There was no significant difference in the distribution of the variant allele in the women whose tumours had (7/47) or did not have (4/49) TP53 mutations (P = 0.523). It does not appear that the presence of this variant allele of P21CIP1/WAF1 has any aetiological role in ovarian carcinomas. Studies in other tumours support this finding.

The detection of loss of heterozygosity, indicative of the presence of a tumor suppressor gene, has been reported to occur frequently on chromosome 22q in human ovarian cancer. In this study, 110 sporadic ovarian tumors were analyzed using 8 polymorphic loci to define a minimum region of loss. Fifty-eight (53%) tumors showed loss of heterozygosity, and of these 6 exhibited partial loss, enabling the identification of two candidate tumor suppressor gene loci. One region, of less than 0.5 cM, is flanked by D22S284 and CYP2D, and a second region lies distal to D22S276. Analysis of loss of heterozygosity with respect to grade and stage suggests that chromosome 22q loss of heterozygosity is of more relevance in tumor progression rather than initiation.

Loss of heterozygosity on chromosome 11q23.3-qter is a frequent event in ovarian carcinoma, implying the existence of an important ovarian tumor suppressor gene(s) within the region. To refine a minimum region(s) of loss, 67 ovarian tumors were analyzed for loss of heterozygosity with eight microsatellite markers spanning 11q23.3-qter. Forty tumors (61%) demonstrated allelic losses. Twenty-seven of these had allelic losses on only part of 11q23.3, which enabled the identification of two distinct regions likely to harbor ovarian tumor suppressor genes. The proximal region, flanked by markers D11S925 and D11S1336, is less than two megabases while the second more distal region, flanked by markers D11S912 and D11S439, is approximately eight megabases. The refinement of these candidate tumor suppressor gene loci will facilitate future loss of heterozygosity studies and enable the isolation of candidate genes from these regions.

Enhanced matrix metalloproteinase-2 (MMP-2/72-kd type IV collagenase) action correlates with invasion in neoplasia. MMP-2 is inhibited in vivo by tissue inhibitors of metalloproteinases (TIMPs)-TIMP-1 and, especially, TIMP-2. A synthetic, biotinylated inhibitor specific for activated MMP-2 in solution phase, and immunohistochemistry were used to detect MMP-2 and TIMP-2 expression in cell lines and ovarian tumors and to analyze the surface-binding capacity of the inhibitors, which are potential therapeutic agents. Characterization of novel monoclonal antibodies to MMP-2 and TIMP-2 is described together with immunocytochemical staining of 83 paraffin-embedded ovarian tumors (67 malignant, 7 borderline, 9 benign) and 9 cell lines. Synthetic MMP-2 inhibitor binding under controlled conditions was visualized by immunofluorescence and avidin-biotin complex immunoperoxidase methods in cell lines and cryostat sections of ovarian tumors. MMP-2 and TIMP-2 showed heterogenous immunoreactivity, with enhanced staining on high-grade tumors, specifically at the invasive front and in vascular invasion. TIMP-2 immunoreactivity was maximal in malignant cell cytoplasm and less intense in desmoplastic fibroblasts. One monoclonal antibody to MMP-2 showed membrane immunoreactivity, apically polarized in benign and low-grade tumors but depolarized and strong in 37 of 44 cases of high-grade invasive tumors. Eleven of eighteen ovarian carcinomas and six of nine cell lines showed membrane localization of the synthetic inhibitor. Maximal binding occurred in the ovarian cell line OVCA 432 and the breast cell lines MCF 7 and MDA MB 435, all of which were immunoreactive for MMP-2. Cell lines propagated on type I collagen showed no enhancement in inhibitor binding. This study demonstrates cell surface binding of a synthetic MMP-2 inhibitor and provides new evidence of MMP-2 and TIMP-2 immunoreactivity in ovarian carcinomas and cell lines.

Inherited predisposition to cancer has become an increasingly important part of the practice of clinical genetics. Hereditary breast and ovarian cancer form a significant part of this new field. There are approximately 13,000 new cases of breast cancer diagnosed every year in Canada, and 5% of these can be expected to have a hereditary basis. Large studies implicated a dominant gene with a population frequency of 0.33% and high penetrance. Linkage analysis localised one such gene to chromosome 17q21 in 1990, and 1994 BRCA1 was cloned. The BRCA1 gene is large, and mutations are spread over the entire gene. There is evidence of common origins for some of the mutations. Early detection of hereditary breast and ovarian cancer is problematic. There is no evidence that early diagnosis by population-based mammography improves survival in sporadic premenopausal breast cancer. Multimodal screening for ovarian cancer has not been adequately assessed. Moreover, there is little evidence that such procedures are more likely to be effective in high-risk groups. Women who may gain initial benefit from the cloning of BRCA1 are from BRCA1-linked families who have living affected relatives. Over the next few years, it is likely that predictive testing for hereditary breast and ovarian cancer will become a routine clinical service in large centres.

The CDKN2 gene encodes a cell cycle regulatory protein and is located on chromosome 9p21, a region deleted in a wide variety of primary tumours. While mutations in the CDKN2 gene itself are frequently observed in tumour cell lines, they are less common in primary tumours. We have investigated the role of the CDKN2 gene in ovarian cancer by analysis for allelic loss of 9p21 and single-strand conformational polymorphism analysis of exons 1 and 2 of CDKN2 in 67 primary ovarian tumours. Loss of heterozygosity on 9p21 was frequently observed (24/50 informative tumours) and was common in early-stage tumours, suggesting that it is an early event in ovarian tumorigenesis. Homozygous deletion of the CDKN2 gene was detected in only 1 tumour. No somatic or germline mutations were observed in CDKN2, though a codon 140 polymorphism was detected in 2 cases. This suggests that CDKN2 is not involved in ovarian tumorigenesis and that another gene(s) may be the target of the frequent 9p allelic losses observed.

Somatic mutations in TP53 are seen in many human cancers. In addition, the protein product of the wild-type TP53 can be sequestered by the protein MDM2 (murine double minute 2). This protein is commonly overexpressed in human sarcomas and gliomas, usually as a result of gene amplification. In this study, 43 ovarian carcinomas (OCs) were analysed for aberrations in the TP53 gene by immunohistochemistry (IHC), loss of heterozygosity (LOH) or mutation analysis. The MDM2 gene and its product was studied by Southern blotting and IHC. Over 50% of the OCs studied showed mutations in TP53 by either direct sequencing (19/36, 53%), positive IHC (23,43, 53%) or both, whereas 0/32 had amplification of MDM2 and only 1/37 tumours had positive IHC using the anti-MDM2 antibody IF-2. The solitary example of positive IHC in this series was seen in a mixed müllerian tumour with sarcomatous differentiation and was not accompanied by MDM2 DNA amplification. These results support previous data showing that around 50% of OCs have mutations in TP53 and in addition, suggest that MDM2 is not amplified in OC, but the presence of sarcomatous features in mixed müllerian tumours may result in positive immunohistochemistry with IF-2.

Frequent loss of heterozygosity (LOH) has been reported on 22q in ovarian carcinoma, implying the presence of a tumour-suppressor gene. The neurofibromatosis type 2 gene (NF2) at 22q12 is a plausible candidate. Analysis of 9 of the 17 exons of NF2 by single-strand conformational polymorphism (SSCP) in 67 ovarian carcinomas did not detect any somatic mutations, suggesting that NF2 is not involved in the pathogenesis of ovarian carcinoma. LOH data support this conclusion and that the putative tumour-suppressor gene lies distal to NF2, beyond D22S283.

Frequent loss of heterozygosity in ovarian carcinoma (OC) has been reported on several different chromosomes. We have studied 27 OCs and corresponding normal tissue for loss of heterozygosity (LOH) using 10 markers detecting polymorphisms on chromosome 5 (two on 5p and eight on 5q). Three tumours showed extra copies, rather than loss, of one homologue. Twelve of 24 remaining tumours showed LOH on 5q (50%), and 8 of 21 on 5p (38%). Of the 12 showing LOH on 5q, 7 showed reduction to homozygosity at all informative markers over the chromosome. The remaining 5 showed LOH over all of 5q. These data are consistent with the localisation of a tumour suppressor gene on 5q involved in OC. A good candidate is the APC gene, which is mutated in a number of adenocarcinoma derived from several tissues and is located at 5q21-22. The APC gene was studied in 40 ovarian tumours, including all the OCs showing LOH, by single-strand conformation polymorphism (SSCP). Analysis of all the exons containing published mutations (approximately 4.7 kb of the cDNA) did not reveal any band shifts that could be attributed to mutations. However, a new polymorphism was detected, as well as 7 known polymorphisms. Together, these data indicate that (1) LOH is common on chromosome 5 in OC, (2) APC is not mutated in OC, and (3) another gene (or genes) on chromosome 5q is responsible for the LOH seen.

Frequent loss of heterozygosity has been described on several chromosomes in ovarian carcinoma (OC), but few tumour suppressor genes (TSGs) have been analysed. Mutations in the GTPase-related domain (GRD) of the TSG NF1 have been described in tumours not usually associated with neurofibromatosis type 1 (NF1). We analysed 36 OCs for mutations in this domain using single-strand conformation polymorphism. The NF1-GRD can downregulate the active form of p21RAS and, therefore, we analysed the same tumours for mutations in RASK. No cases of mutations in NF1-GRD were seen, and only two cases of RASK mutations were found. Thus, activation of the RAS signalling pathway by RASK or NF1 mutations does not appear to be common in OC.

Frequent loss of heterozygosity (LOH) on both arms of chromosome 17 has been described in ovarian carcinoma (OC) by a number of groups, and the recent fine mapping of an inherited breast-ovarian cancer gene (brcAI) to a small region at 17q12-21 has focused interest on this area. We studied 28 sporadic OCs with 21 markers at 18 loci on chromosome 17 (5 on 17p and 13 on 17q). LOH on 17p was 78%, and always involved p53. In 86% of cases showing LOH, all informative markers on chromosome 17 showed reduction to homozygosity. Using 6 markers flanking the brcAI locus on 17q, LOH was 75%, but only one tumour showed LOH with telomeric retention. The data therefore suggest that small deletions on chromosome 17 are very uncommon in sporadic OC. No genomic rearrangements by Southern blotting were seen in the brcAI candidate gene estradiol 17 beta dehydrogenase 2 (17hsd2), or in erbB2, prohibition (phb) and nmeI (previously nm23-HI). LOH on chromosome 17 was more common in high-grade, late-stage carcinomas, and no LOH was seen in any benign or borderline tumour. This study has clearly demonstrated that LOH at any one site on chromosome 17 is most commonly explained by LOH over the whole of the chromosome. We consider possible reasons for the absence of small deletions on chromosome 17 in OC.

The gene for tumor necrosis factor, TNF, was expressed in 45 out of 63 biopsies of human epithelial ovarian cancer. In serous tumors, there was a positive correlation between level of TNF expression and tumor grade. TNF mRNA was found in epithelial tumor cells and infiltrating macrophages, whereas TNF protein localized primarily to a subpopulation of macrophages within and in close proximity to tumor areas. mRNA and protein for the p55 TNF receptor gene localized to the tumor epithelium and tumor, but not to stromal macrophages. The p75 TNF receptor was confined to infiltrating cells. Cells expressing TNF mRNA were also found in ovarian cancer ascites and TNF protein was detected in some ascitic fluids. In 2 out of 12 biopsies of normal ovary, TNF mRNA was detected in a minority of cells in the thecal layer of the corpus luteum. Serum levels of TNF and its soluble receptor did not correlate with extent of TNF expression in matched biopsies. Northern and Southern analysis revealed no gross abnormality of the TNF gene. The coexpression of TNF and its receptor in ovarian cancer biopsies suggests the capacity for autocrine/paracrine action. TNF antagonists may have therapeutic potential in this malignancy.

Investigation of genetic changes in tumours by loss of heterozygosity (LOH) is a powerful technique for identifying chromosomal regions that may contain tumour suppressor genes. LOH has been described on chromosome 6 in ovarian carcinoma using restriction fragment length polymorphism analysis with a small number of probes. We studied 29 ovarian carcinomas with 19 probes mapping to chromosome 6. Sixteen of the 29 tumours showed LOH on 6q (55%). Of these 16, 63% showed loss of all informative markers on that arm. One tumour showed loss of 6q24-qter, localising the putative tumour suppressor gene to that region. Loss on 6p was 28% overall. However, using three dinucleotide repeat primer pairs from 6p to study LOH in seven selected tumours, LOH was demonstrated at both 6p22.3-pter and at 6p12-6p22. These results confirm that 6q harbours a tumour suppressor gene of relevance to ovarian carcinoma and suggest that there may also be a similar gene(s) on 6p. By Southern analysis, there was no evidence of genomic rearrangements of the oestrogen receptor gene, located at 6q25.1. LOH on 6q was more common in high than low grade tumours. The relevance of our findings to previous work in ovarian cancer and other solid tumours is discussed.

The 11q13 chromosomal region encodes oncogenes relevant to a variety of human cancers as well as a tumour suppressor gene implicated in multiple endocrine neoplasia type 1. In addition, high affinity folate receptor (FOLR1), which maps to 11q13.3-13.5, is expressed at an elevated level on the surface of over 80% of nonmucinous epithelial ovarian cancers. Further telomeric, 11q breakpoints are found in many cancers. We studied the involvement of 11q markers in ovarian cancer by looking for tumour-specific loss of heterozygosity (LOH), as well as amplification or rearrangements that might explain the overexpression of FOLR1. Twenty eight epithelial ovarian cancers, along with lymphocyte DNA from the same individual were used for Southern blotting with polymorphic probes from 11q. PCR primers from 11q23.3 were also used. The 11q13 band was amplified in four out of 28 cancers. The amplicon included the probe D11S146 as well as FGF3 (formerly INT2) and FOLR1 in one out of these four cases, thus crossing the bcl1 translocation breakpoint. LOH was seen in three out of 16 cases with FGF3 (11q13). A much higher frequency of LOH (8/12) was found at 11q23.3-qter, implying the presence of a tumour suppressor gene in this region.

We describe the isolation of a complementary DNA (cDNA) sequence encoding the ovarian cancer-associated antigen recognized by monoclonal antibody MOv18 and its identification as a high-affinity folate-binding protein (FBP). Functional cDNA clones were isolated using mRNA from the ovarian carcinoma cell line SKOV3 and colon carcinoma cell line HT29, by transient expression in WOP cells and selection of expressing cells by adhesion to antibody-coated magnetic beads. The cDNAs differed in the lengths of 5'- and 3'-noncoding regions, but they encoded identical peptides. A database search clearly showed them to be adult high-affinity FBPs with amino acid sequences identical with those isolated from normal placenta and several carcinoma cell lines. Reactivity of cell lines with MOv18 was quantitatively consistent with the expression of FBP mRNA. Southern hybridizations show evidence of a family of related genes and/or pseudogenes and were mapped to chromosome 11q13.3-14.1 by fluorescent in situ hybridization using cosmid clones containing part of this region. Also identified were two PstI polymorphisms of four and three alleles, respectively, and a two-allele MspI polymorphism. The folate-binding protein locus was not amplified in any of the 16 carcinoma cell lines tested and in only 1 of 10 serous adenocarcinomas, indicating that overexpression of FBP in ovarian cancer cannot, in general, be due to gene amplification.