Pattern of Lymph Node Metastases in Stage I Invasive Epithelial Ovarian Carcinomas

Abstract & Commentary

Synopsis: Fifteen percent of patients with apparent stage I unilateral ovarian carcinoma have microscopic lymph node metastases, with contralateral nodal involvement seen in half the cases.

Source: Cass I, et al. Gynecol Oncol. 2001;80:56-61.

Cass and colleagues identified 96 patients with stage I ovarian cancer visibly confined to 1 ovary. Pathology reports were reviewed to identify metastatic lymph node involvement, number of involved nodes, and their locations. Patients with gross disease in the pelvis or abdomen or those who had grossly positive lymph nodes removed for debulking were excluded from this review. Fourteen of 96 patients (15%) had microscopically positive lymph nodes on pathologic review. All of these patients had grade 3 tumors. Grade 3 tumors were more commonly seen in lymph node-positive vs. lymph node-negative patients (P < .001). Pelvic nodes were positive in 7 patients (50%), paraaortic nodes in 5 patients (36%), and both in 2 patients (14%). Forty-two patients had lymph node sampling only on the side ipsilateral to the neoplastic ovary, 4 of whom (10%) had lymph node metastases. Fifty-four patients had bilateral sampling performed, 10 of whom (19%) had lymph node metastases. Of these 10 patients, isolated ipsilateral lymph node metastases were seen in 5 (50%) cases. Isolated contralateral lymph node metastases were seen in 3 (30%) cases, and bilateral metastases were seen in 2 (20%). Cass et al concluded that in this cohort of patients with clinical stage I ovarian cancer with disease limited to one ovary, bilateral lymph node sampling increased the identification of nodal metastases. Ipsilateral sampling may result in the understaging of patients. Bilateral pelvic and paraaortic lymph node sampling is recommended to accurately stage ovarian cancer.

COMMENT BY DAVID M. GERSHENSON, MD

Lymphatic drainage of the ovary is an understudied area. Available information, based on studies of the past 2 decades, suggests that lymph node involvement in ovarian cancer is common—as high as 75% in apparent stage III disease based on peritoneal assessment. The incidence of occult lymph node metastases in apparent stage I ovarian cancer in unknown, but existing studies estimate the rate to be between 5-25%. Thus, the results of this study are consistent with previous studies. A number of important points can be made about this study. First, studies of apparent early-stage ovarian cancer are difficult because stage I disease is relatively uncommon. This study is retrospective and comes from 2 large institutions. Second, the question addressed in this study is extremely difficult to study because of the low incidence of lymph node involvement in apparent stage I disease. Only 14 of 96 patients had positive findings; when one begins to subdivide patients, the numbers become small. Third, the prevailing thought is that apparent stage I ovarian cancer confined to one ovary only metastasizes to ipsilateral lymph nodes; this study demonstrates that this is a myth. Even at these two academic institutions, 42 of the 96 patients had only unilateral nodal sampling. And finally, as in endometrial cancer, there is no standard regarding the extent of lymph node sampling. The findings of this study would suggest that complete bilateral paraaortic and pelvic lymphadenectomy should be the standard in apparent stage I ovarian cancer. Ideally, the results of this study should be confirmed by a large, prospective, clinical trial.