Elsevier

Gynecologic Oncology

Volume 95, Issue 1, October 2004, Pages 120-126
Gynecologic Oncology

High-risk endometrial cancer subgroups: candidates for target-based adjuvant therapy

https://doi.org/10.1016/j.ygyno.2004.06.042Get rights and content

Abstract

Objective.

To identify patients with endometrial cancer at risk for hematogenous, lymphatic, or peritoneal recurrence (or combinations of them) who might potentially benefit from target-based therapies.

Methods.

During a 13-year period, 915 patients had endometrial cancer managed with hysterectomy and standard adjuvant therapy. On the basis of our previous regression analyses, depth of myometrial invasion predicted the risk for hematogenous recurrence; positive lymph nodes and cervical stromal invasion predicted lymphatic recurrence; stage IV disease or combination of nonendometrioid histology, cervical stromal invasion, positive lymph nodes, and positive peritoneal cytology was predictive of peritoneal recurrence. Median follow-up was 66 months.

Results.

Applying the above criteria to the population of 915 patients, 24% were considered at risk for hematogenous recurrence, 18% for lymphatic recurrence, and 16% for peritoneal recurrence. The respective relapse rates at 5 years were 28% for patients who were at risk for hematogenous recurrence, 31% for lymphatic recurrence, and 42% for peritoneal recurrence. This contrasted with less than a 5% recurrence rate in the corresponding subgroups not at risk for relapse (P < 0.001). Collectively, of the 915 patients, 324 (35%) were considered at risk for recurrence in one or more of the above three sites. Overall, 89% of all recurrences were identified in this at-risk group. Importantly, 46% of the patients considered at risk subsequently had recurrence in one or more of the three sites, compared with only 2% of patients not at risk for relapse (P < 0.001).

Conclusion

Patients at risk for relapse had a 46% probability of experiencing recurrence within 5 years despite management with standard therapy. New target-based algorithms for the 35% of endometrial cancer patients deemed at risk should be incorporated in the development of future prospective multimodality clinical trials predicated on site(s) of recurrence.

Introduction

In the United States, endometrial cancer is the most common malignancy of the female reproductive tract and is exceeded annually in overall frequency only by breast, colon, and lung cancers. It is estimated that during calendar year 2004, 40,320 new cases of endometrial cancer will be diagnosed and 7090 women will die of this disease [1]. This neoplasm generally becomes manifest early in its natural history, resulting in approximately 80% of patients presenting with stage I disease. Nevertheless, nearly one of every three women who die of endometrial cancer presents with presumed localized disease.

The majority of treatment failures and the accompanying compromised longevity are the result of the failure to recognize sites of occult extrauterine dissemination at the time of primary treatment. Furthermore, adjuvant therapy has generally been dictated by traditional preferences (modality-based) rather than target-based algorithms as determined by patterns of recurrence.

The natural history of epithelial corpus cancer includes four potential routes of metastasis: contiguous extension, hematogenous dissemination, lymphatic embolization, and exfoliation with intraperitoneal spread. The associated recurrences for each of these diverse routes of spread would presuppose different adjuvant treatment strategies. In addition, such target-based therapies are predicated on the cataloging of specific pathologic or molecular factors that identify patients at high risk for harboring occult disease disseminated via one or more of these routes.

In a preliminary analysis, we reported independent risk factors for recurrence based on hematogenous [2], [3], lymphatic [4], and intraperitoneal [5] routes of dissemination in endometrial cancer (Table 1).

The objective of the present study was to identify subgroups of patients with predictable regional or distant patterns of recurrence who might potentially benefit from target-based adjuvant therapies.

Section snippets

Materials and methods

From 1984 to 1996, 1109 patients with endometrial cancer were managed surgically at Mayo Clinic (Rochester, MN). Based on data from their medical records, 915 patients with epithelial endometrial cancer met the following inclusion criteria: (1) treatment included hysterectomy and removal of existing adnexal structures and (2) no other malignancy was diagnosed within 5 years before or after the diagnosis of endometrial cancer (except for carcinoma in situ or skin cancer other than melanoma).

Results

The mean age ± SD of the 915 patients was 64.5 ± 11.0 years (range, 22–95 years). The clinical and pathologic characteristics of the patients are summarized in Table 2.

Overall, lymph node dissection was performed in 514 patients (56%), specifically, pelvic lymphadenectomy in 497 (54%) and paraaortic lymphadenectomy in 152 (17%) (paraaortic lymphadenectomy only in 12 and concomitant pelvic and paraaortic lymphadenectomy in 140); the site from which the nodes were harvested was not identified in

Discussion

Approximately 70% to 80% of patients with endometrial cancer present with localized disease that potentially can be cured with surgery alone. However, approximately one of every three women dying of endometrial cancer was considered to have early-stage locoregional disease [1]. The main reason for treatment failure after traditional modality-based therapy is the presence of documented or occult extrauterine systemic disease and our inability to recognize and treat it successfully.

Traditional

Acknowledgments

Supported by the Mayo Cancer Center (P30CA15083) and the Rochester Research Committee, Mayo Foundation.

References (26)

Cited by (0)

Presented at the 13th International Meeting of the European Society of Gynaecological Oncology (ESGO), Brussels, Belgium, April 6–10, 2003.

No conflicts of interest exist for this manuscript.

View full text