SEARCH

SEARCH BY CITATION

The origin of ovarian cancer, and specifically high-grade serous carcinoma, has been the subject of considerable debate over the past 25 years, and the elucidation of this issue has been hampered because >90% of these tumors are discovered after considerable tumor growth has occurred, obscuring not only the ovaries but also the fallopian tubes and peritoneal surfaces. This propensity for discovering high-grade serous carcinomas at a late stage has not been altered appreciably by conventional approaches to early detection, and the beneficial role of serologic screening for cancer-related biomarkers remains to be proven. Thus, identifying the precursors to ovarian malignancies and their relevance to the detection or prevention of ovarian cancer has been a major objective in ovarian cancer research.

At least a partial answer to the question of tumor origin has come from the study of women who harbor an inherited mutation in either the BRCA1 or BRCA2 breast cancer gene (BRCA-positive). The inheritance of a heterozygous loss of function mutation of BRCA1 or BRCA2 gives rise to an average 10-fold increased risk of developing ovarian cancer. 1 Eight percent of BRCA-positive women who undergo risk-reducing surgery will have an early serous (and, less commonly, endometrioid) carcinoma identified. The accumulated studies in the past decade increasingly have implicated the distal fallopian tube rather than the ovary as the source of disease in these women, and approximately 80% of these early cancers are traced to an early (intraepithelial) carcinoma in the fimbrial mucosa. 2-5 This same intraepithelial carcinoma also can be identified in up to 47% of presumed sporadic ovarian and primary peritoneal serous cancers, suggesting that the salpingeal mucosa is prone to malignancy in all women, albeit more exquisitely in the BRCA-positive population, which is handicapped by the germline mutation in 1 functioning allele. 6, 7 It is noteworthy that 1 previous report suggested a 120-fold greater risk of fallopian tube cancer in BRCA1 mutation carriers. 8

Inactivating p53 mutations are identified in virtually all serous carcinomas and most often lead to the accumulation of altered protein product in the tumor cell nuclei. In a previous study of benign tubal mucosa from BRCA-positive women, investigators observed an increase in epithelial atypia and p53 positivity; and, recently, a candidate serous cancer precursor—the p53 signature—was described. 9, 10 Like its malignant counterpart, the p53 signature was observed most commonly in the fimbria, harbored evidence of DNA damage, increased p53 staining and p53 mutations, involved the tubal secretory cell, was observed in continuity with malignancy, and was present more commonly in women who had serous carcinoma. 9, 10 Significantly, the p53 signature was nearly equally common in BRCA-positive women and population controls; thus, it was proposed as a generic precursor that was independent of BRCA status. 10, 11 On the basis of the high frequency of BRCA1 or BRCA2 gene inactivation in familial and sporadic ovarian cancer, it was presumed that loss of heterozygosity (LOH) or other functional loss of the remaining BRCA1 or BRCA2 allele was an important rate-limiting event in the transition from p53 signature, to serous tubal intraepithelial carcinoma, and, finally, to serous carcinoma. 12

In this issue of Cancer, Norquist et al have brought the p53 signature (or, as they call it, p53 foci) into a clearer focus, albeit with some controversy. 13 These authors from the University of Washington School of Medicine make at least 3 important observations, including 1) that p53 signatures are more common in BRCA-positive women, 2) that the index of tubal epithelial proliferation is higher in the BRCA-positive group, and 3) that the transition from p53 signature to intraepithelial carcinoma is marked by a loss of function of the remaining BRCA allele. The latter observation is the first to make it clear that nonfunctioning BRCA and p53 genes not only are compelling dance partners in early serous carcinogenesis but will not be discovered together until p53 signatures have transitioned to malignancy. This finding also dispels the notion that the pathogenesis of the p53 signature requires previous inactivation of both alleles of the BRCA1 or BRCA2 gene, clearly indicating that p53 typically is inactivated before LOH of the wild type BRCA allele during serous carcinogenesis. This further supports the prior observations that p53 signatures are not highly dependent on BRCA status. 10, 11

Whether inactivating mutations of bothBRCA1 or BRCA2 and p53 are more common in the distal tube is not clear at this point; however, it is reasonable to assume that a cell that has lost 1 BRCA allele is at greater risk for BRCA inactivation—which requires only 1 more genotoxic event—than a cell that has not lost an allele. A graphic example of this gradient of genotoxic risk can be observed with fallopian tubes from women with Li Fraumeni syndrome, who are haploinsufficient for p53. The fimbria in these women exhibits a multiplicity of p53 signatures, likely reflecting the lower threshold for p53 inactivation in the face of a germline mutation. 14 However, these individuals have no increased risk of serous carcinoma, presumably because the BRCA1 and BRCA2 genes are intact. The link between the subsequent loss of the wild type allele of BRCA1 or BRCA2 and malignancy is consistent with previous observations that the inactivation of BRCA genes in normal cells leads to growth arrest as a consequence of accumulated DNA damage. However, loss of another tumor suppressor gene in this context, such as p53, may permit these cells to bypass cell cycle arrest, accumulate more genomic instability, and progress to malignancy. 15 Taken together, the results reported by Norquist et al provide convincing evidence for this sequence of these genetic events during the development of serous cancer.

The observations by Norquist et al that the frequencies of p53 signatures and DNA proliferation are greater in BRCA-positive women 13 are more difficult to interpret. Others have not reported a significant difference in the frequency of p53 signatures between BRCA-positive women and controls. 10, 11 Given the meticulous nature with which the analysis was done in this study, including BRCA genetic analyses in all patients, a greater frequency in BRCA-positive women cannot be dismissed, although additional studies will be required to resolve this issue. The findings by the authors that proliferation was increased globally in the tubes of these women, that p53 signatures were more prevalent in the distal tube, and that p53 signatures did not exhibit increased proliferation make it more problematic to establish a direct connection between proliferation and p53 mutations.

In support of the Norquist et al, Piek et al appreciated the increased proliferative index in the fallopian tubes of BRCA-positive women and the increased risk of gene-altering events in cells that undergo cell division. 9 It is possible that proliferation, which Norquist et al demonstrate is increased globally in the fallopian tubes of BRCA-positive women (and is increased even more in women with malignancy), and along with the inactivation of p53 and BRCA1 or BRCA2, is an independent event. This supports the hypothesis that the development of malignancy requires the convergence of multiple events, some of which might serve as markers of cancer risk. For example, in a recent study, “epigenetic signatures” (homeobox A methylation) were uncovered in benign uterine mucosa and were associated with an increased ovarian cancer risk. 16 Another study recently identified discrete secretory cell outgrowths in the fallopian tube that are not related directly to p53 mutations but may be increased in the presence of tubal malignancy. 17 Thus, the possibility that the fallopian tubes in certain populations are at greater risk of early genetic events is gaining traction, reinforcing the role of this organ both as the site of pelvic serous carcinogenesis and, possibly, as a potential source of biomarkers that could prove valuable in screening or ascertaining cancer risk. Advancing this concept will require teasing apart the influence of inherited (BRCA-positive) genetic predisposition, acquired mutations (p53), and more globally distributed gene dysregulation (histone modification and DNA methylation) on serous carcinogenesis in both BRCA-positive carcinomas and sporadic serous carcinomas. Norquist et al and their predecessors have made it increasingly clear that all of these perturbations may be complementary in the distal fallopian tube.

The fallopian tube is a challenging anatomic region and harbors microscopic precursors (some as few as 12 cells) that can test the limits of any technology and can spawn cancers that are much more conspicuous after they have left the tube. Nevertheless, it must be explored. Norquist et al have demonstrated that this is not an easy area of research, but those who make the effort may find it extremely fertile.

CONFLICT OF INTEREST DISCLOSURES

  1. Top of page
  2. CONFLICT OF INTEREST DISCLOSURES
  3. REFERENCES

Supported by National Institutes of Health grant 3R21CA124688-02S1.

REFERENCES

  1. Top of page
  2. CONFLICT OF INTEREST DISCLOSURES
  3. REFERENCES