Abdominal Carcinomatosis in Women with a History of Breast Cancer

Abstract

This paper presents a retrospective review of patients with a history of primary breast cancer who underwent subsequent treatment for abdominal carcinomatosis at the authors' institution from 1988 through 2002. Inclusion criteria included diagnosis of recurrent breast cancer or breast cancer with a second primary cancer, clinical documentation of initial breast cancer, stage of disease, and histologic findings, and laparotomy or laparoscopy with pathologic documentation of abdominal carcinomatosis. The primary breast cancer was retrospectively staged according to the criteria established by the American Joint Committee on Cancer. During the study period, 1501 women were identified who were treated for recurrent breast cancer. Of these, 79 met the study criteria. Patients had an average age of 50 years, and slightly less than half (47%) were postmenopausal. Sixty-two percent of the patients had stage I breast cancer, 21.5% were stage II, 13.9% were stage III, and 2.5% were stage IV. Most women (81%) were diagnosed with infiltrating ductal carcinoma, and 15% had lobular carcinoma. Eighteen of the 79 patients had a previous, nonabdominal, breast cancer recurrence. The median time between diagnosis of breast cancer and surgery for abdominal carcinomatosis was 5.39 years (range, 0.10-33.55 years; mean, 8.37 years). Most women (81%) with abdominal carcinomatosis were postmenopausal. Of the 79 women with abdominal carcinomatosis, 54 (68.4%) were diagnosed with primary ovarian cancer, and 5 (6.3%) were diagnosed with primary peritoneal cancer. The FIGO stage distribution in these 59 patients was stage IIIB in 7 (11.9%), stage IIIC in 41 (69.5%), and stage IV in 11 (18.6%). One woman had a granulose cell tumor; the rest were epithelial tumors. Almost half (49%) of these women had a family history of breast cancer and 8 (13.6%) had a family history of ovarian cancer. Recurrent breast cancer was diagnosed in 19 (24%) of the 79 patients with abdominal carcinomatosis. Although 52.6% had a family history of breast cancer, none of the women with recurrent breast cancer had a family history of ovarian cancer. In a multivariate logistic analysis of variables associated with abdominal carcinomatosis resulting from primary ovarian or peritoneal cancer, stage I breast cancer, and the absence of a prior breast cancer recurrence were significantly more likely to be predictive of primary ovarian/peritoneal cancer than recurrent breast cancer (odds ratio [OR] = 10.73, 95% confidence interval [CI] = 2.6-43.7; and OR = 10.60, 95% CI = 2.5-45.2, respectively; P = 0.001 for both). A strong but nonsignificant association with primary ovarian/peritoneal cancer versus recurrent breast cancer was seen for longer interval between diagnosis of breast cancer and diagnosis of abdominal carcinomatosis and higher levels of serum CA-125. Most patients underwent exploratory laparotomy with bilateral salpingo-oophorectomy (n = 68), hysterectomy (n = 54), and various degrees of tumor cytoreduction, although 9 patients underwent only diagnostic laparoscopy with biopsies. Optimal cytoreduction, that is, the maximal diameter of the largest residual tumor mass was 1 cm or less, was performed in 33 (55.9%) patients. Over 90% of patients received a platinum-based chemotherapy alone or in combination with other agents. After a median follow up of 51.6 months, the median survival for these patients was 30 months. In 19 of the 33 patients with optimal cytoreduction, 18 had no visible residual disease. The median survival time for these patients was 82 months compared with 33 months for optimally reduced patients with some visible disease (hazard ratio [HR] = 10.07, 95% CI = 3.34-31.26, P < 0.0001), and 18 months for patients with suboptimal cytoreduction (HR = 26.37,95% CI = 8.30-83.74, P < 0.0001). At the last follow up, 10 patients, all of whom had no visible residual disease, were alive after more than 5 years. In the women with a final diagnosis of metastatic breast cance