Nine randomized trials and one published data meta-analysis provide the evidence base for assessing the role of adjuvant hormone therapy in women with stage I endometrial cancer. There are several factors that limit the interpretability of the results, but the greatest limiting factor is that the trials, which span a thirty-year period, generally have inconsistent reporting throughout. This limits the quality assessment of internal validity related to patient and study characteristics, as well as outcomes. There were no quality of life data reported, little data on adverse events or treatment compliance, and limited data on recurrence and survival outcomes, especially concerning data on hazard ratios and time-to-event estimates. There were also differences in patient populations, unexpected findings that were not consistent with the results of similar randomized trials, and noted discrepancies between patients at baseline, despite the randomization process. These limitations affect the external validity of the trials; however, these trials provide the only randomized data that inform the role of adjuvant hormonal therapy in this patient population.
In spite of the limitations, the evidence was consistent in the direction of effect to indicate that adjuvant hormonal therapy does not confer survival advantages on patients with stage 1 endometrial cancer. Eight of the nine randomized trials failed to detect any differences in survival between treatment and control groups. Although the remaining trial did demonstrate a survival difference, the quality of the trial is subject to criticism because of important differences in baseline characteristics between patient groups, and it is not consistent with the results of the other randomized controlled trials identified. The magnitude of effect is also highly unexpected when one looks at the results of the other similar trials reported. In addition, in two meta-analyses, no survival advantages were detected with adjuvant hormonal therapy.
In seven of the nine randomized trials, recurrence rates were not significantly different between patients in the adjuvant hormonal groups as compared to patients in the control groups. Of the two trials that detected lower recurrence rates in patients in the progestin group, the trial by Urbanski et al. reported more favourable baseline characteristics for patients in the treatment group, and the trial by Quinn reported data on patients at high risk of recurrence. Although the current meta-analysis, as well as the previously published meta-analysis did show a marginal reduction in recurrence rate, this was not statistically significant at the 0.05 level.
Finally, although not consistently reported, the adverse events related to patients in the hormonal groups were generally higher than those in the control groups. Minor side effects were reported to be higher in the treatment groups, though tests of statistical significance were not performed. Non-cancer related deaths were shown to be higher with progestagen in one randomized trial mainly due to cardiovascular or thromboembolic events (p=0.04). In contrast, Urbanski et al. reported a 10% non-cancer related death rate in the control group and a 0% rate in the treatment group; an unexpected finding not seen in the other randomized trials. The published data meta-analysis by Martin-Hirsch et al. did not demonstrate a statistically significant difference in the number of non-cancer related deaths.
Given the lack of an overall survival benefit, a marginal decrease in recurrence rates seen mainly in patients at higher risk of recurrence, and the need for treatment regimens that can span years, with possible increases in adverse events, there is currently insufficient evidence to support the use of hormonal therapy as adjuvant treatment for patients with early stage endometrial cancer.