Joseph T. Chambers

Uterine Papillary Serous Carcinoma: A Study on 108 Cases with Emphasis on the Prognostic Significance of Associated Endometrioid Carcinoma, Absence of Invasion, and Concomitant Ovarian Carcinoma

One hundred eight cases of uterine papillary serous carcinoma (UPSC) were analyzed to characterize its histologic features and biologic behavior. Special situations that could conceivably modify the behavior and therapeutic approaches were considered: (1) the occurrence of areas of endometrioid carcinoma in otherwise typical UPSC; (2) the confinement of UPSC to an otherwise benign endometrial polyp or the endometrial mucosa or absence of residual tumor at the time of hysterectomy; and (3) the coexistence of a superficial UPSC and a serous ovarian carcinoma. There was coexistence of endometrioid and UPSC in 22 cases, and tumor was confined to an endometrial polyp or endometrium in 19 cases. There was simultaneous pathologic stage I UPSC and papillary serous ovarian carcinoma in 10 cases. In patients with pathologic stages I and II UPSC the presence of areas of endometrioid carcinoma intermixed with the UPSC did not improve survival. Patients with stage I disease and no residual tumor or tumor confined to an endometrial polyp/endometrial mucosa and without vascular invasion had a survival not statistically different from those with stage I disease but with myometrial and/or vascular invasion. Patients with stage I UPSC with concomitant ovarian serous surface papillary carcinoma had survival not statistically different from patients with stage IV UPSC.

Early pathologic stage clear cell carcinoma and uterine papillary serous carcinoma of the endometrium : comparison of clinicopathologic features and survival

Clear cell carcinoma (CCC) and uterine papillary serous carcinoma (UPSC) are aggressive variants of endometrial carcinoma that may coexist or share some clinicopathologic features suggesting a similar biologic spectrum and the need for a common therapeutic approach. Twenty-nine cases of pathologic FIGO stage I and II CCC and 47 cases of FIGO stage I and II UPSC seen and treated at Yale-New Haven Hospital were reviewed, and the survival rates with regard to various pathological parameters were compared. Both groups of patients had similar clinical profiles with respect to presentation, age, weight, and medical problems. The 5-year survival for pathologic stage I patients with CCC was 72% and for those with UPSC 44%. The 5-year survival for pathologic stage II patients with CCC was 59% and for those with UPSC 32%. Analysis of survival showed that the depth of myometrial invasion, the presence of vascular space invasion, and the admixture of endometrioid features did not influence survival in either group of patients. In CCC, survival was also not influenced by the predominant histologic pattern, i.e., papillary versus nonpapillary. The results of this study suggest that early stage CCC and UPSC have similar clinicopathologic profiles, suggesting the need for aggressive approaches including a staging laparotomy and possibly similar therapy. However, the stage I CCC patients had a significantly better survival than the stage I UPSC patients.

Borderline ovarian tumors

Abstract Ninety-four patients with borderline ovarian tumors were retrospectively analyzed for clinical features, treatments, and survival characteristics. There were 46 patients with FIGO stage IA cancer, 7 with stage IB, 20 with stage IC, 4 with stage IIB, 5 with stage IIC, 5 with stage IIIA, 3 with stage IIIB, and 4 with stage IIIC tumors. Seventy patients had at least a total abdominal hysterectomy and bilateral salpingo-oophorectomy, 20 patients had conservative surgery including unilateral salping-oophorectomy or ovarian cystectomy, and 4 patients had bilateral salpingo-oophorectomy. Fifteen patients with stage I disease received adjuvant melphalan therapy and 2 received external beam radiation for concomitant gynecologic cancers; 7 with stage II tumors received adjuvant melphalan therapy and 1 received external beam radiation; and 5 with stage III tumors received melphalan therapy and 6 patients received cisplatin-based combination chemotherapy. Follow-up ranged from 1 to 117 months, with a median of 33.5 months. Eighty-seven patients were alive. Seven patients died, two of disease. The overall 5-year survival rate was 83.0%; those treated with adjuvant therapy had a 79.5% survival, whereas the others had 84.6% survival. Second-look surgery was performed in 10 patients; six results were negative after melphalan therapy, one was negative after cisplatin combination therapy, and one was negative after no adjuvant treatment. Two patients had positive second-look surgery, one with stage IIIC disease treated with a cisplatin combination and the other with stage IC disease treated with melphalan. This review did not demonstrate that patients with borderline ovarian tumors benefited from adjuvant therapy.

Endometrial sampling: When? Where? Why? With what?

Today, evaluating women with abnormal uterine bleeding generally is initiated in the office with an endometrial biopsy. The indications and contraindications for endometrial sampling along with situations which do not, per se, demand sampling are listed in Figure 1. In women younger than 40 years of age, it may be appropriate in some clinical situations to initiate hormonal therapy after an endocrine evaluation before endometrial sampling; however, with the newer sampling devices that cause minimal discomfort, a histologic evaluation can be performed easily. Furthermore, the endometrial biopsy may help to distinguish anovulatory from ovulatory bleeding and exclude a hyperplastic condition or carcinoma. If the patient does not respond to medical therapy, then hysteroscopy may identify endometrial polyps or submucosal myomas. Bleeding in postmenopausal women requires endometrial sampling. If a diagnosis of cancer can be made in the office, this will expedite treatment. For those cases in which, for technical reasons, it is impossible to do an office biopsy or in which an examination under anesthesia is necessary for evaluation, then a D&C is indicated. The refined technology of transvaginal ultrasonography and hysteroscopy in the future may influence more directly the evaluation of women with abnormal uterine bleeding. As noted, transvaginal ultrasonography may determine which women would benefit from an endometrial biopsy, both for symptomatic and asymptomatic women. Likewise, the hysteroscope, under certain circumstances, may help identify pathologic findings missed by endometrial biopsy and/or reassure the patient or physician that a negative biopsy is the result of an atrophic mucosa. Because of the increase in the use of hormonal therapy, both in postmenopausal women for replacement and in women with breast cancer as adjuvant therapy, endometrial sampling must be performed for screening. Follow-up for women with premalignant changes of the endometrium treated with hormones also would require sampling to assess response. The overwhelming arguments in favor of the accuracy of an office-based endometrial biopsy, the convenience to the patient and physician, and the cost containment have been established firmly in the literature. Office screening procedures will continue to play important roles in the diagnostic skills of the gynecologist.

Parametrial involvement, regardless of nodal status: a poor prognostic factor for cervical cancer.

Objective To evaluate the effect of resection of central disease when the parametria are involved by tumor in high-risk stage I cervical cancer patients. Methods Thirty-two patients with high-risk stage I cervical cancer who underwent radical hysterectomy and had pathologic findings of positive lymph nodes (N = 13), positive parametria (N = 7), or both (N = 12) were identified retrospectively. The effects of various histopathologic findings on disease-free interval and survival were evaluated, including the effect of resection of central disease with and without positive nodal disease. Kaplan-Meier survival curves were compared with the log-rank test. Multivariate analyses nusing a stepwise regression model were performed. Results Compared with other histologies, adenocarcinoma was associated with a significantly shorter disease-free interval (P = .037). Among patients with parametrial involvement, lymph node status did not affect disease-free interval or survival. However, when patients with positive lymph nodes were examined, the additional finding of parameterial positivity significantly worsened both diseasefree interval (P = .039) and survival (P = .036). When the 19 patients with positive parametria, regardless of lymph node status, were compared with those with positive lumph nodes alone, the former group had a significantly shorter disease-free interval (P = .038). The tumor recurred in 12 of these 19 patients; all cases involved the pelvis, with a median time to recurrence of 15 months. Multivariate analysis showed that adenocarcinoma histology (P = .038) and parametrial involvement (P = .043) were independent, poor prognostic indicators for disease-free interval. Conclusion Involvement of the parametria, regardless of lumph node status, and adenocarcinoma histology confer a poor prognosis in high-risk patients undergoing radical hysterectomy. Caution should be used when contemplating resection of bulky tumors as part of primary therapy if the parametria appear to be involved by tumor.

HIGH LEVEL EXPRESSION OF FMS PROTO-ONCOGENE mRNA IS OBSERVED IN CLINICALLY AGGRESSIVE HUMAN ENDOMETRIAL ADENOCARCINOMAS

Six micron paraffin sections of paraformaldehyde-fixed endometrial currettings of 21 benign and neoplastic endometrial specimens were assayed for tumor cell-specific oncogene expression by in situ hybridization with probes for six oncogenes, beta-actin, and the E. coli plasmid pBR322. In the benign hyperplasias and invasive adenocarcinomas, multiple oncogenes, including erbB, fms, c-myc, and Ki-ras were expressed at significant levels. For the adenocarcinomas, statistical analysis demonstrated that high levels of expression of fms-complementary mRNA correlated strongly with clinicopathologic features (high FIGO histologic grade, high FIGO clinical stage, deep myometrial penetration) predictive of aggressive clinical behavior and poor outcome. The authors discuss the role which M-CSF receptor (the fms gene product) and locally-produced M-CSF may play in the development of the observed aggressively-malignant phenotypes. They also propose that pre-hysterectomy assay of fms gene expression in endometrial currettings in FIGO Stage I patients might be clinically useful to help identify preoperatively those patients with deep myometrial penetration or other locoregional spread.

Ovarian germ cell malignancies: The Yale University experience

Abstract Eighty-one patients with ovarian germ cell malignancies (immature teratoma 29, dysgerminoma 26, endodermal sinus tumors 15, mixed germ cell tumor 8, other 3) seen or consulted on at Yale University over a 15 year period are presented. Initial therapy was successful in 70 of 81 (86.4%) patients and 75 (92.6%) are currently alive and disease free. Early stage dysgerminomas may be safely treated with surgery whereas advanced disease is exquisitely sensitive to vincristine, actinomycin D, and cyclophosphamide (VAC) therapy. Early stage immature teratoma is uniformly successfully treated with short-term VAC whereas advanced disease requires longer treatment. Early stage endodermal sinus tumor (EST) and mixed germ cell tumors may be effectively treated with VAC or platinum-based therapy but advanced disease should be treated with paltinum-based regimens. Serial alpha fetoprotein assays should determine duration of therapy in tumors containing EST elements. Conservative surgery to preserve reproductive function is appropriate for all patients with early stage ovarian germ cell malignancies and selected patients with advanced disease.

Colony-stimulating factor-1 in primary ascites of ovarian cancer is a significant predictor of survival

OBJECTIVE Our purpose was to determine whether the concentration of colony-stimulating factor in ascites of ovarian carcinoma is a prognostic factor for survival. STUDY DESIGN Forty-four ascites samples from patients undergoing primary surgery for ovarian carcinoma were measured for colony-stimulating factor-1 by radioimmunoassay. Retrospective analysis of clinical data allowed comparison of accepted prognostic factors to ascites colony-stimulating factor-1 concentration for impact on survival by means of life-table analysis (Kaplan-Meier) by the Wilcoxon test and the Cox regression methods. RESULTS In patients with advanced disease (International Federation of Gynecology and Obstetrics stages III and IV, n = 37) ascites colony-stimulating factor-1 concentration levels below a critical cutoff of 8.59 ng/ml were associated with longer overall survival (p < 0.05) and were a better predictor of survival than any other prognostic factor except zero residual disease after cytoreduction. International Federation of Gynecology and Obstetrics stage, tumor histologic type, malignant cells in fluid, grade of tumor, age, and performance status at presentation were not predictive of outcome. CONCLUSION Colony-stimulating factor-1 in ascites may be an independent indicator of prognosis in patients with epithelial ovarian cancer.

Estrogen and progestin receptor levels as prognosticators for survival in endometrial cancer.

The survival of 213 postmenopausal patients with primary endometrial cancer was analyzed as a function of clinicopathologic features and cytosol steroid receptor levels. Estrogen receptor (ER) levels (P = 0.008) and progestin receptor (PR) levels (P = 0.0001) were negatively correlated with grade. ER and PR levels were positively correlated with each other (P = 0.0001), but neither was correlated with age. In 187 patients with stages I and II, ER positivity (greater than or equal to 20 fmole/mg cytosol protein (cp] was statistically associated with grade (P = 0.007); and PR (greater than or equal to 7 fmole/mg cp) was statistically associated with grade (P = 0.001). Univariant analysis revealed that survival for the early endometrial cancer patients was significantly dependent upon ER status (P = 0.0003), PR status (P = 0.0016), and grade (P = 0.0002). Multivariant analysis of ER status, PR status, age, and grade showed that the ER status was a significant prognostic factor for survival (P = 0.0168), even if the positivity of the PR status was defined at greater than or equal to 50 fmole/mg cp. If ER status was divided at 0-19, 20-100, and greater than 100 fmole/mg cp, survival was significantly different between the low range group and the other two groups. If PR status was divided at 0-6, 7-50, and greater than 50 fmole/mg cp survival was significantly different between the first two groups and the high range group. Thus, survival in these endometrial cancer patients was better predicted by ER status than grade.

Sequelae of lateral ovarian transposition in unirradiated cervical cancer patients

The sequelae of lateral ovarian transposition (LOT) in cervical cancer patients has been examined only in the light of the effect of pelvic radiation therapy on ovarian preservation. Preservation of ovarian function has not been examined in the absence of radiation therapy, and symptomatic ovarian cyst formation in transposed ovaries with the need for subsequent surgery has not been addressed in either radiated or unirradiated cervical cancer patients. We studied 84 premenopausal FIGO stage IA or IB cervical cancer patients treated by primary radical hysterectomy between the years 1978 and 1988. None of these patients received adjuvant radiation therapy. Fifty-nine of eight-four patients had radical hysterectomy (RH) without LOT. These patients were compared to 25 of 84 patients who had LOT in addition to RH. The incidence of symptomatic ovarian cysts, the majority requiring operative intervention, was 24% in the ovarian transposition patients as compared to 7.4% in those who had RH alone. This threefold increase in symptomatic benign ovarian cyst formation in the translocated ovary was significant (P = .048). On the other hand, LOT in these RH patients does not appear to increase the incidence of early menopause (P greater than 0.05). On follow-up of those patients who did not incur additional surgery or radiation, 4.3% became menopausal, as compared to 4.1% of those patients undergoing RH alone, with the mean ages of the two groups being comparable.