Myometrial Invasion and the Tumor-Free Distance from the Uterine Serosa in Endometrial Cancer?

Abstract & Commentary

Synopsis: TFD as a single measurement carries significant prognostic importance in women with comprehensively staged endometrial cancer.

Source: Lindauer J, et al. Gynecol Oncol. 2003;91(3): 547-551.

Lindauer and colleagues set out to evaluate whether the tumor-free distance from the uterine serosa to the deepest invasive lesion was a better predictor of patient outcome from endometrial cancer than the traditional measure of myometrial invasion. To do this, they retrospectively evaluated all surgically staged endometrial adenocarcinoma patients between 1997 and 2000. Depth of myometrial invasion was defined as the distance in millimeters between the endometrial-myometrial junction and the deepest invasive lesion. Tumor-free distance was defined as the distance in millimeters between the uterine serosa and the deepest area of myometrial invasion. Depth of invasion and tumor-free distance were expressed as continuous variables in this report. To determine their predictive and prognostic significance, these 2 variables were compared with traditional surgicopathologic factors and against outcomes of recurrence and survival. A total of 153 patients met study criteria. The most common stage was IB, and 23 patients had positive nodes. The median depth of invasion was 0.5 cm and the median tumor-free distance was 1.4 cm. At a median follow-up of 29 months, 10 patients recurred. By univariate analysis, both invasion parameters were significant predictors of traditional surgicopathologic variables.

However, only tumor-free distance was predictive of recurrence. In addition, while both tumor-free distance and depth of invasion were significant predictors of survival, only tumor-free distance was correlative with surgicopathologic variables and predictive of recurrence and survival in the multivariate model. Depth of invasion became predictive of recurrence when myometrial thickness was included in the model. A tumor-free distance of 1 cm maximized the balance of sensitivity and specificity in predicting recurrence. Lindauer et al concluded that tumor-free distance as a single continuous measurement carries significant prognostic importance in women with comprehensively staged endometrial cancer.

Comment by Robert L. Coleman, MD

Epithelial malignancy of the uterine corpus remains the most common gynecological malignancy diagnosed in the United States. Characterized by frequent early stage at diagnosis, endometrial cancer is often considered a "curable" lesion. However, more than 20% of clinical stage I cases will have evidence of extrauterine disease, and up to 15% will recur at 5 years.1,2 Exhaustive evaluation of important clinicopathological risk factors relating to these clinical outcomes, along with survival, led FIGO to modify the staging schema in 1988, incorporating findings at surgery generated from formal exploration. Currently, surgical staging is recommended for all patients with the diagnosis.3 Nonetheless, it is a rare event for noninvasive lesions to be associated with metastatic disease, prompting some to use a measure of grade and gross myometrial invasion to triage those who should undergo the more extensive surgical staging procedure.4,5 However, unreliable and inconsistent correlation of intraoperative findings to final postoperative pathology hallmarks the challenge of making an accurate assessment of myometrial invasion and thus, leading potentially to improper staging—raising yet another challenge in determining accurate postoperative therapy.

Lindauer et al study another methodology for quantifying myometrial invasion by looking at, essentially, the complement of the standardized practice. This measure of tumor-free distance from the serosa to the deepest element of invasion is easier to measure and, through multivariate analysis, was better associated with parameters determining survival. A tumor-free distance of 1 cm was associated with the best probability testing characteristics for recurrence. Although not an a priori goal, Lindauer et al did have 2 pathologists evaluate a limited number of specimens for this parameter. The reproducibility was quite high, being identical in 4 of 5 cases tested.

It is of note that traditionally measured depth of myometrial invasion became an important independent variable when myometrial thickness was also considered. It is from this "denominator" that percent of myometrial invasion is calculated and reported. In the current study, depth of myometrial invasion was considered as a continuous variable without relation to the uterine wall. In this respect, it is not difficult to see why the variable fell out on multivariate testing. For instance, a 0.75 cm invasive lesion in a 3.0 cm uterine wall (25% invasion) would be expected to behave differently (and staged differently) than the same 0.75 cm lesion in a wall 1.0 cm thick (75% invasion). Calculating a percent invasion "normalizes" to some extent the invasion characteristic and as such brings the measure more in line with what tumor-free distance is essentially measuring. My suspicion is that the 2 variables would explain similar amounts of variance on regression testing if compared. Parenthetically, similar strategies for assessing biological behavior have been reported in carcinoma of the cervix, with short tumor-free distances frequently associated with nodal metastases. Whichever methodology is ultimately used, the growing trend in the surgical evaluation of endometrial cancer is to perform complete surgical staging wherein an accurate picture of the disease state can be constructed to effectively counsel and treat patients with this malignancy. 

Dr. Coleman is an oncologist in the Dept. of Gynecologic Oncology, University of Texas Southwestern Medical Center, Dallas, Tex.


  1. Creutzberg CL, et al. Lancet. 2000;355:1404-1411.
  2. Creasman WT, et al. Cancer. 1987;60:2035-2041.
  3. Straughn JM, et al. Gynecol Oncol. 2002;84: 194-200.
  4. Altintas A, et al. Eur J Gynaecol Oncol. 1999;20: 329-331.
  5. Goff BA, Rice LW. Gynecol Oncol. 1990;38:46-48.