Pelvic lymph node count is an important prognostic variable for FIGO stage I and II endometrial carcinoma with high-risk histology

https://doi.org/10.1016/j.ygyno.2005.11.032Get rights and content

Abstract

Objective.

To determine whether pelvic lymph node count is associated with patterns of recurrence or survival in patients with FIGO stage I and II endometrial cancer.

Methods.

Single institution retrospective study of 467 patients with FIGO stage I and II endometrial cancer treated with primary surgery including lymph node dissection. Analysis included pelvic lymph node count, histology, stage, age, race, BMI, year of surgery, depth of myometrial invasion, and adjuvant radiation. Kaplan–Meier life-tables were used to calculate survival; the Cox proportional hazards model was used to identify prognostic factors independently associated with survival.

Results.

Mean pelvic lymph node count was 12.6 (SD ± 8). Distant recurrence was associated with decreased pelvic lymph node count, high-risk histology, and postoperative pelvic radiation. Pelvic lymph node count was not associated with survival by univariate analysis, however, overall (OS) and progression-free (PFS) survival were significantly better with pelvic lymph node counts ≥12 among women with high-risk histology (P < 0.001), but not among women with low-risk histology. Multivariable Cox proportional hazards regression identified increasing age, non-Caucasian race, and high-risk histology as independent negative prognostic factors for both OS and PFI. Among patients with high-risk histology, pelvic lymph node count remained an independent prognostic factor for both overall (OS) and progression-free survival (PFS) in the model, with hazard ratios of 0.28 and 0.29, respectively, when ≥12 pelvic lymph nodes were identified. Pelvic lymph node count had no association with OS or PFS in women with low-risk histology.

Conclusion.

Pelvic lymph node count ≥12 is an important prognostic variable in patients with FIGO stage I and II endometrial cancer who have high-risk histology. Most likely, the association of survival and lymph node count in this group is the result of improved staging among patients with higher pelvic lymph node counts.

Introduction

In 1988, FIGO adopted the surgical and pathologic staging system currently in use for the staging of endometrial adenocarcinoma [1]. Surgical staging was specified to include hysterectomy with bilateral salpingo-ophorectomy, analysis of peritoneal cytology, biopsy of any suspicious intraperitoneal or retroperitoneal lesions, and retroperitoneal lymph node sampling [1]. However, the type and extent of lymph node dissection have not been specified by FIGO. Many patients are at low risk for lymphatic metastases based on tumor features including: superficial depth of invasion (<50% myometrial thickness) [2], [3], [4], small size of tumor (<2 cm) [3], and low tumor grade (grades 1 and 2) [2], [3], [4]. Some Gynecologic Oncologists omit pathologic assessment of lymph nodes for patients with very low-risk pathologic features because the minimal risk of lymph node metastases [3], [4]. These features can be assessed preoperatively by histology and by gross and frozen section assessment at the time of primary operation by the surgeon and pathologist. However, discrepancies between the preoperative histology, intraoperative assessment, and final pathology report may occur [5].

In patients who have less favorable tumor features, a retroperitoneal node dissection is often performed. The extent of pelvic lymph node sampling remains an area of controversy, with no consensus regarding the type of lymph node evaluation that should be performed. The original Gynecologic Oncology Group (GOG) studies and subsequent randomized trial of adjuvant pelvic radiation therapy in intermediate risk stage I and stage II patients required only sampling of lymph nodes from external iliac, obturator, and hypogastric areas [2], [6]. Chuang et al. have proposed that 9 different lymph node sites be assessed in order to accurately stage a patient with endometrial cancer [7]. Kilgore et al. proposed biopsy of multiple nodes from multiple sites [8]. Fanning reported systematic lymphadenectomies from his series, however, he did not report pelvic lymph node counts [9]. We recently reported improved outcome among patients with normal appearing lymph nodes whose pelvic and paraaortic lymph node count exceeded 11 nodes [10]. That patient cohort included patients with both histologically negative nodes and clinically occult lymph node metastases. The available literature strongly suggests that the number of lymph node metastases identified increases as the extent of lymph node dissection increases [8], [10]. Clearly, there is prognostic significance to identifying lymph node metastases. Reported rates of serious morbidity arising from lymph node dissection in patients with endometrial cancer have ranged from 6% to 19% [9], [10]. Whether there is a therapeutic benefit to this dissection in endometrial cancer patients without histological evidence of lymph node metastases remains controversial. The therapeutic merits of regional lymphadenectomy are debated in other malignancies [11], [12].

Our aims in this retrospective study were to confirm whether pelvic lymph node count has prognostic significance in endometrial cancer confined to the uterus and cervix and to investigate whether pelvic lymph node count has any influence on the patterns of recurrence in patients with FIGO stage I and II endometrial cancer.

Section snippets

Methods

We performed a single-institution retrospective study approved by the Institutional Review Board at Duke University Medical Center. Charts and medical records were abstracted for patient demographic information, clinical and pathological information from primary surgery and patient follow-up status. Patients eligible for this study included those with primary surgery at Duke University Medical Center between 1973 and 2002. Surgery consisted of hysterectomy and adnexectomy with at least pelvic

Results

Of the 1656 patients treated for endometrial cancer at Duke University Medical Center between 1973 and 2002, 707 patients underwent hysterectomy, bilateral salpingo-ophorectomy, and pelvic lymph node sampling for the primary treatment of endometrial cancer. After excluding patients found to have metastatic disease at surgical staging, positive peritoneal cytology, preexisting cancer at another site, those who received preoperative radiation, and those with incomplete data available for

Discussion

Our data strongly suggest that women with FIGO stage I and II endometrial cancer and high-risk histology or high–intermediate risk status have significantly improved survival when a larger number of pelvic lymph nodes are histologically evaluated. At our institution, the pelvic lymph node dissection has evolved to include removal of all nodal tissue from the common iliac bifurcation caudal to the deep circumflex iliac vein, all lymphatic tissue between the obturator nerve and external iliac

Cited by (162)

View all citing articles on Scopus
View full text