|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 23, No 1 (January 1), 2005: pp. 127-132 © 2005 American Society of Clinical Oncology. DOI: 10.1200/JCO.2005.04.109 Risk-Reducing Salpingo-Oophorectomy in BRCA Mutation Carriers: Role of Serial Sectioning in the Detection of Occult MalignancyFrom the Gynecologic Oncology Program and Cancer Risk Program, University of California San Francisco Comprehensive Cancer Center, San Francisco, CA Address reprint requests to John L. Ziegler, MD, MSc, PO Box 0808, UCSF Comprehensive Cancer Center, 2340 Sutter Street, Room N424, San Francisco, CA 94143; e-mail: ziegler{at}itsa.ucsf.edu
PURPOSE: Women who carry deleterious mutations of BRCA1 or BRCA2 genes have up to a 54% lifetime risk of developing ovarian cancer. After childbearing, women at high risk increasingly choose bilateral risk-reducing salpingo-oophorectomy (RRSO). Two recent studies of BRCA mutation carriers reported occult malignancy in 2.5% of women undergoing RRSO. This study aimed to increase this detection rate using a protocol. METHODS: In 1996, the University of California San Francisco Gynecologic Oncology Program instituted a surgical-pathologic RRSO protocol that was composed of complete removal and serial sectioning of both ovaries and fallopian tubes, peritoneal and omental biopsies, and collection of peritoneal washings for cytology. We report the pathologic findings in 67 BRCA mutation carriers according to the degree of adherence to this protocol. RESULTS: Of the 67 procedures, the protocol was followed completely or partially in 41 (61%). Seven occult malignancies were discovered, four in the fallopian tube and three in the ovaries. Six of these were microscopic, and all seven (17%) were found in specimens from complete or partial protocol procedures as opposed to standard procedures (P = .026). Other variables such as age, parity, BRCA1 or BRCA2 mutation, or type of surgery did not alter the strong effect of protocol procedure on the cancer detection rate. CONCLUSION: A rigorous operative and pathologic protocol for RRSO increases the detection rate of occult ovarian malignancy in BRCA mutation carriers nearly seven-fold. If confirmed, this finding will alter postoperative management because additional staging, chemotherapy, and follow-up may be necessary in affected women.
Women who carry deleterious mutations of BRCA1 or BRCA2 genes have up to a 54% lifetime risk of developing ovarian cancer, accounting for approximately 10% of incident ovarian cancers.1-4 The risk of ovarian cancer in BRCA1-mutation carriers is about twice the risk in BRCA2 carriers, especially if the mutation occurs in the "ovarian cancer cluster" region of the gene. BRCA mutations are also prevalent in women with fallopian tube cancer and primary peritoneal carcinomatosis, particularly in those of Ashkenazi Jewish descent.3,4 Because screening for ovarian cancer has low predictive value, women who have completed childbearing often choose risk-reducing bilateral salpingo-oophorectomy (RRSO) to lower their chances of cancer.5 Both clinical and occult ovarian and tubal cancers have been found at the time of risk-reducing procedures. Several centers have reported their institutional experience of occult carcinomas in the ovaries removed at RRSO. These studies are retrospective and include periods before BRCA testing was offered. Therefore, the reports are heterogeneous with respect to BRCA testing, surgical procedures, and pathologic examination. Lu et al6 reviewed 50 women at high risk of ovarian cancer who underwent RRSO; 37 (74%) had ovaries and fallopian tubes fully sectioned. They found four carcinomas in a subset of 33 (12.1%) of 50 women whose risk of BRCA mutations was more than 25%; one was seen at surgery and three were discovered in the pathology specimen. All four women were BRCA mutation carriers, and three had normal transvaginal ultrasonography within 6 months of prophylactic surgery. The authors recommend "pathologists section and histologically examine ovaries and fallopian tubes in toto."6 Colgan et al7 found five occult carcinomas (8.3%) in a series of 60 consecutive prophylactic RRSO procedures that contained 20 fully-sectioned fallopian tubes; all five had BRCA1 mutations. Scheuer et al8 found two occult carcinomas in 90 women (2.2%) undergoing prophylactic surgery. Both were stage I, and neither were detected at transvaginal ultrasonography within the month before surgery. Among 30 women who had RRSO for BRCA mutations (73%) or a suggestive family history (27%), Leeper et al9 found four (13.3%) occult carcinomas (three fallopian tube) on histologic review. Kauff et al10 followed 98 BRCA mutation carriers, who had prophylactic RRSO, for an average of 2 years. Three (3%) had occult cancers in surgical specimens (two ovarian and one fallopian tube). Rebbeck et al11 followed 259 BRCA carriers for up to 8 years. They found six ovarian stage I cancers at the time of surgery (2.3%). Thus, in the largest two series of known BRCA carriers reported to date, nine occult cancers were found in 357 women (2.5%) who had risk-reducing RRSO,10,11 although the surgical and pathologic procedures were not standardized. In 1996, the Gynecologic Oncology Program at the University of California San Francisco (UCSF) Comprehensive Cancer Center instituted a surgical-pathologic protocol to increase the ability to detect occult cancer in BRCA mutation carriers who elect RRSO. The procedure includes serial sectioning of the ovaries and fallopian tubes, random peritoneal and omental biopsies, and cytology of peritoneal washings. We hypothesized that protocol adherence is the main predictor in the ability to detect occult malignancy in RRSO specimens. This report describes the pathologic findings in 67 BRCA carriers from the UCSF Cancer Risk Program database who underwent RRSO.
The UCSF Cancer Risk Program has provided genetic counseling and testing for hereditary breast and ovarian cancer since 1996. All patients signed an informed consent, approved by the UCSF Committee on Human Research, to undergo genetic counseling, testing and follow-up for 20 years. Women who test positive for a BRCA1 or BRCA2 gene mutation are counseled about RRSO using the National Institutes of Health consensus guidelines, which recommend prophylactic oophorectomy at 35 years of age or when child bearing is complete.12 At UCSF, our surgical protocol involved complete surgical resection of the ovaries and fallopian tubes up to their insertion into the cornua of the uterus. We take peritoneal washings as well as biopsies of the pelvic peritoneum, gutter peritoneum, and the omentum. The ovaries and fallopian tubes are then serially sectioned for pathology review. The recommendation for a more extensive operative procedure and pathologic examination is discussed with each high-risk patient before consent for surgery. From our cumulative database of 342 BRCA mutation carriers, we identified 118 women who had RRSO during the period of January 1996 to December 2003. We were able to obtain pathology records from 67 (57%) of these women. We next determined the extent to which the UCSF-recommended protocol was followed by the surgeon and pathologist in each case. To be considered as having "full" adherence to protocol, resection, washings, biopsies, and serial sectioning had to be performed. "Partial" adherence included cases where only some of the protocol was followed. Standard (nonprotocol) procedures varied somewhat by institution and generally consisted of laparoscopic removal of the ovaries and partial removal of the fallopian tubes with representative pathologic sectioning after inspection. We then tabulated the findings of cancers in pathology specimens according to protocol adherence (full, partial, none).
Using the Mantel-Haenszel
Table 1 outlines the protocol at UCSF for operative and pathologic procedures recommended for women at high risk of hereditary ovarian cancer who choose RRSO.
Of 342 BRCA carriers, 118 had RRSO between 1996 and 2003: 72 were probands and 46 were female relatives (Table 2). Of the 72 probands, complete operative and pathologic records were available from 59 (82%). Eight records (17%) were available from their relatives. These 67 patients form the basis for this study. The main reason for the low accrual from relatives was the need to obtain records through the intercession of the proband because of privacy concerns.
Table 3 shows the characteristics of the 67 women who underwent RRSO. Overall, no gross cancers were detected at the time of surgery, but at subsequent pathologic examination, four occult fallopian tube carcinomas and three occult ovarian carcinomas were found. The median age of women with cancer was 53 years (range, 49 to 54 years). Among 43 BRCA1 carriers, five had occult cancer. Among 24 BRCA2 carriers, two cancers were found—a nonsignificant difference (P = .67). One additional patient had normal pathologic findings but her peritoneal washings contained atypical cells. Although malignancy could not be ruled out, her physicians felt that the finding was due to recent chemotherapy for invasive breast cancer. In addition, six women had benign pathologic findings. Two patients who had normal findings at RRSO subsequently developed primary peritoneal carcinoma 5 years after surgery.
Table 4 lists the age at surgery, details of pathologic findings in the operative specimens, and the associated BRCA mutations. One tumor was grossly visible as a 12-mm pedunculated tumor in the lumen of the fallopian tube (patient H); the remaining malignancies were microscopic. All seven malignancies were serous type and moderately to poorly differentiated. Six were microscopic and would not have been detected by gross examination of the organ.
Table 5 shows the relationship of pathologic findings according to the surgical/pathologic protocol recommended at UCSF. Overall, 55 women had laparoscopic surgery and 12 underwent laparotomy. In some cases, the operating surgeon at the referring community hospital adopted the protocol after discussions with consulting physicians at the Cancer Risk Program. Reasons for incomplete protocol adherence varied considerably. In some cases, only the ovaries, but not fallopian tubes, were serially sectioned. In others, peritoneal washings or biopsies were omitted. Because of the small numbers of these subsets, we elected to combine all non–fully adherent procedures into a "partial" category. All seven malignant tumors were discovered in women who underwent full or partial protocol procedures (P = .026). No cancers were found in random peritoneal or omental biopsies.
We analyzed other variables that might have influenced the main result: age at surgery (grouped in tertiles); parity; type of surgery; ethnicity; hospital where the procedure was performed; BRCA mutation (1 or 2); family history of ovarian cancer; and history of prior breast cancer. With the exception of older age at surgery, as expected, none of these variables independently influenced the finding of malignancy (data not shown). Further, none of the variables individually influenced the statistical significance of full or partial protocol adherence (ie, submission of all tissue) as a determinant of finding occult malignancy (Table 6). We had insufficient data on use of hormones or infertility drugs for this analysis.
This study examined the ability of a rigorous operative protocol for RRSO to detect occult tumors in carriers of BRCA mutations. The UCSF-recommended protocol was developed to thoroughly screen for ovarian and peritoneal cancer using a variety of techniques. A fraction of patients were expected to have an underlying malignancy at the time of RRSO. Peritoneal biopsies and omentectomy are considered part of surgical staging for early-stage ovarian cancer and were therefore included in the protocol. Overall, we detected seven malignancies (10.4%) in 67 patients, six of which were microscopic. All seven cancers (four in the fallopian tube and three in the ovary) were found in 41 women whose surgeons and pathologists followed all or part of the intensive protocol. This rate (seven of 41; 17%) is almost seven-fold higher than the rate of 2.5% reported in recent studies of BRCA mutation carriers whose prophylactic RRSO procedures are routine (ie, laparoscopic removal of the ovaries and fallopian tubes but without serial sectioning).10,11 No cancers were discovered in omental or peritoneal biopsies and the yield of these procedures must await further study. At least one in situ fallopian tube malignancy in our series was accompanied by the finding of malignant cells in the peritoneal washings (Table 4). Peritoneal washings are inexpensive and easy to perform; we believe the procedure may be a useful adjunct but will also require further study. In the past, the diagnosis of in situ dysplasia and adenocarcinoma of the fallopian tube has been controversial. More recently, gynecologic pathologists have accepted carcinoma in situ as a specific diagnostic entity that is listed in the recent WHO histologic classification of tumors of the fallopian tube, and is recognized by the finding of nuclear hyperchromasia and atypia, mitotic figures, and nuclear stratification.12 Immunohistochemical stains for p53 protein are typically positive in adenocarcinoma in situ but negative in adjacent non-neoplastic epithelium, and staining for the proliferation marker MIB-1 (Ki-67) is increased.13 We acknowledge that the diagnosis of dysplasia or atypical hyperplasia with cytologic features that fall short of adenocarcinoma in situ remains controversial (eg, patient J, Table 4). The significance of dysplasia/atypical hyperplasia in patients with BRCA mutations is unclear, and more patients with this finding need to be identified and studied to determine whether it has any relationship to the development of tubal adenocarcinoma.3,7,14,15 If confirmed, the high rate of occult ovarian and fallopian tube malignancies discovered on rigorous tissue examination has clinical implications in postoperative management. First, it is evident that occult malignancies are more common in carriers of BRCA mutations than previously thought. Because the procedure followed was not fully compliant with the protocol in about half of the patients, this high rate may be an underestimate. The multidisciplinary nature of genetic counseling for hereditary cancer risk (primary provider, genetic counselor, surgeon, pathologist) is critical in communicating the patient’s cancer risk and surgical-pathologic procedures to the care team. Second, clinical management of women with high-grade occult malignancies should include additional staging procedures and consideration of chemotherapy, as occurred in five of our patients. Third, it would appear that RRSO is indeed risk-reducing, as nearly one in five women who underwent full or partial protocol procedures had occult malignancies. Finally, if we assume that screening for ovarian cancer continues to have a very low predictive value, an RRSO with complete removal of the fallopian tubes and serial sectioning of the ovaries and fallopian tubes may be the most effective and reassuring risk-reducing measure for BRCA mutation–positive women.5,10,11,16 An added benefit is the lowering of breast cancer risk in premenopausal BRCA carriers by 50%.11 Although follow-up time is short (median, 3 years), two (3%) of 67 patients developed primary peritoneal cancer following RRSO 5 years following surgery. Primary peritoneal cancer occurs at rates of approximately 1% in other series of RRSO.10,11 It is of interest that neither of these two patients was in the protocol group. In one case, the ovaries were removed with a 2-cm length of fallopian tube, and only representative sections of the ovaries, but not fallopian tubes, were examined. In the second case, the patient had a total abdominal hysterectomy and RRSO, but standard examination of the surgical specimens failed to reveal any abnormality. Our results are subject to the usual perils of a multicenter, retrospective analysis. Heterogeneity of data collection from several institutions, insufficient data from potential confounders (eg, hormone use), and small numbers weaken the robustness of the main findings. We were missing pathologic data from BRCA-positive relatives of UCSF patients who underwent RRSO because these records could only be obtained by proxy. This may have biased the result in favor of procedures done at UCSF where protocol adherence was most likely practiced and where pathologic records were most easily obtained. Misclassification is an unlikely source of error. We examined confounding variables that may have an a priori effect on the main outcome, and found none. We could not ascertain oral contraceptive use in sufficient numbers to judge its effect, but there is no reason to suspect that this practice would be differentially distributed by protocol adherence. Our numbers are small, and these results are preliminary. Nevertheless, we believe that this multidisciplinary procedure should be more systematically studied, as others have recommended,5,7,9,16-18 because of the important and potentially practice-changing implications.
The authors indicated no potential conflicts of interest.
We thank genetic counselors Nicola Stewart, Robin Lee, Lisa Ku, and Julie Mak for the excellent clinical services provided to our patients.
Preliminary results of this study were presented at the Western Association of Gynecologic Oncologists, Steamboat Springs, CO, June 28-29, 2003. Authors’ disclosures of potential conflicts of interest are found at the end of this article.
1. Wooster R, Weber BL: Breast and ovarian cancer. N Engl J Med 348:2339-2347, 2003 2. Risch HA, McLaughlin JR, Cole DE, et al: Prevalence and penetrance of germline BRCA1 and BRCA2 mutations in a population series of 649 women with ovarian cancer. Am J Hum Gen 68:700-710, 2001[CrossRef][Medline]
3. Levine DA, Argenta PA, Yee CJ, et al: Fallopian tube and primary peritoneal carcinomas associated with BRCA mutations. J Clin Oncol 21:4222-4227, 2003
4. King M-C, Marks JH, Mandell JB, et al: Breast and ovarian cancer risks due to inherited mutations in BRCA1 and BRCA2. Science 302:643-646, 2003
5. DeMichelle A, Weber BL: Risk management in BRCA1 and BRCA2 mutation carriers: Lessons learned, challenges posed. J Clin Oncol 20:1164-1166, 2002
6. Lu KH, Garber JE, Cramer DW, et al: Occult ovarian tumors in women with BRCA1 or BRCA2 mutations undergoing bilateral prophylactic oophorectomy. J Clin Oncol 18:2728-2732, 2000 7. Colgan TJ, Murphy J, Cole DE, et al: Occult carcinoma in prophylactic oophorectomy specimens: Prevalence and association with BRCA germline mutations. Am J Surg Pathol 25:1283-1289, 2001[CrossRef][Medline]
8. Scheuer L, Kauff N, Robson M, et al: Outcome of preventive surgery and screening for breast and ovarian cancer in BRCA mutation carriers. J Clin Oncol 20:1260-1268, 2002 9. Leeper K, Garcia R, Swisher E, et al: Pathologic findings in prophylactic oophorectomy specimens in high-risk women. Gynecol Oncol 87:52-56, 2002[CrossRef][Medline]
10. Kauff ND, Satagopan JM, Robson ME, et al: Risk-reducing salpingo-oophorectomy in women with a BRCA1 or BRCA2 mutation. N Engl J Med 346:1609-1615, 2002
11. Rebbeck TR, Lynch HT, Neuhausen SL, et al: Prophylactic oophorectomy in carriers of BRCA1 or BRCA2 mutations. N Engl J Med 346:1616-1622, 2002 12. Pathology and genetics of tumours of the breast and female genital organs. Lyon, France, IARC Press, 2003 13. Demopoulos RI, Aronov R, Mesia A: Clues to the pathogenesis of fallopian tube carcinoma: A morphological and immunohistochemical case control study. Int J Gynecol Pathol 20:128-132, 2001[Medline] 14. Piek JMJ, Van Diest PJ, Zweemer RP, et al: Dysplastic changes in prophylactically removed fallopian tubes of women predisposed to developing ovarian cancer. J Pathol 195:451-456, 2001[CrossRef][Medline] 15. Agoff SN, Mendelin JE, Grieco VS, et al: Unexpected neoplasms in patients with proven or suspected BRCA-1 or -2 mutations: Implications for gross examination, cytology, and clinical follow-up. Am J Surg Pathol 26:257-259, 2002[CrossRef][Medline] 16. NIH consensus conference. Ovarian cancer. Screening, treatment, and follow-up. NIH Consensus Development Panel on Ovarian Cancer. JAMA 273:491-497, 1995[Abstract]
17. Haber D: Prophylactic oophorectomy to reduce the risk of ovarian and breast cancer in carriers of BRCA mutations. N Engl J Med 346:1660-1662, 2002
18. Schwartz MD, Kaufman E, Peshkin BN, et al: Bilateral prophylactic oophorectomy and ovarian cancer screening following BRCA1/BRCA2 mutation testing. J Clin Oncol 21:4034-4041, 2003 Submitted April 21, 2004; accepted October 5, 2004. This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2005 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
2 test we examined the individual effect of other a priori variables in detecting occult carcinomas and their potential confounding effect on adherence to the protocol as a determinant of finding occult malignancy. Continuous variables were grouped into tertiles or quartiles, and all tests of statistical significance were two-sided. 
