When most of us hear the word “cancer,” our first instinct is “take it out!” However, in the case of endometrial cancer, it’s not so simple. A big question remains- what is the best way to take it out? Is it better to do a big surgery and open the whole abdomen to take out the uterus or to use minimally invasive techniques? Should the surgeon take the ovaries out when removing the uterus or leave them (ovarian conservation)? Does taking out lymph nodes (lymphadenectomy) improve survival?
The January 2016 Green Journal features three excellent epidemiologic studies answering these pressing questions about surgery for endometrial cancer.
The first study by Dr. Fader and her colleagues at Johns Hopkins answers the question: Are hospitals using minimally invasive surgery for non-metastatic endometrial cancer (early cancers that have not spread outside the uterus)? Using the approximately 32,500 cases in the US Nationwide Inpatient Sample database from 207 to 2011, the authors found that surgeons are performing more minimally invasive hysterectomies than ever. 22% of hysterectomies for non-metastatic endometrial cancer were minimally invasive in 2007. Just 4 years later in 2011, a full 51% of hysterectomies were minimally invasive. Patients with private insurance, white patients, and patients receiving surgery at a high volume hospital were more likely to receive minimally invasive surgery. Open surgery had a higher risk of infection and blood clots and a higher cost.
Dr. Wright of Columbia University led the next two studies. Dr. Wright and his team of researchers used The National Cancer Database to allow for population-based analyses. Their first article investigates ovarian conservation for premenopausal women with early stage endometrial cancer. It has been well established that surgeons should practice ovarian conservation when performing hysterectomies for benign indications (like endometriosis or fibroids). In other words, when removing the uterus, they should leave the ovaries. If the surgeon just takes the uterus, fertility is lost. But a woman only goes into surgically-induced menopause if the surgeon removes the ovaries. Early menopause can have numerous detrimental health effects. In fact, the Nurses Health Study showed that women under 50 whose ovaries were surgically removed and who never had estrogen therapy have an increased risk of dying. However, many oncologists are hesitant to leave ovaries when performing hysterectomies for endometrial cancer because they are worried the ovaries might harbor hidden cancer cells or provide estrogen that would stimulate hidden cancer cells.
This leaves us with two questions: One, are surgeons leaving the ovaries? And, two, should surgeons leave the ovaries? The authors isolated 15,648 cases of women under 50 years of age with Stage 1 (earliest stage) endometrioid adenocarcinoma of the endometrium who had a hysterectomy from 1998 to 2012. The rate of ovarian conservation hasn’t changed since 1998, staying stable at about 7%. Several patient characteristics were associated with ovarian conservation: younger women, black women, and women with lower stage and grade tumors. Interestingly, patient characteristics were not the only significant influencers. Both region and hospital type significantly changed a woman’s odds of having ovarian conservation. Community hospitals and hospitals in New England and the Mountain west performed more ovarian conservations.
Now on to the second question: Should surgeons leave the ovaries? The preponderance of evidence from this study supports performing ovarian conservation. Ovarian conservation was not associated with survival on any statistical analysis. As the authors conclude, “Despite the oncologic safety of ovarian conservation, the majority of young women with endometrial cancer still undergo oophorectomy at the time of surgery.”
In the Columbia group’s second article, the authors culled the National Cancer Data Base for the 151,000 cases of women with endometrioid adenocarcinoma who underwent hysterectomy from 1998 to 2011. 66% underwent lymphadenectomy, which is sampling of lymph nodes in specific regions at high risk of intercepting cancer cells attempting to travel to other parts of the body. 34% did not receive lymphadenectomy. The rate of lymphadenectomy has risen with time, from about 50% in 1998 to a height of 70% in 2007. For this study the statisticians did three separate types of statistical analysis to try to isolate the effects of just lymphadenectomy. The first statistical approach was the simplest: a “multivariable regression” adjusted for adjuvant therapy. The “multivariable regression” showed that lymphadenectomy was associated with a 16% reduction in mortality. Next, the statisticians performed a “propensity score analysis.” This type of analysis estimates the probability of treatment and controls for measured confounders. The “propensity score analysis” gave about the same results- lymphadenectomy was associated with a survival benefit. Finally, the statisticians performed an “instrumental variable analysis.” An “instrumental variable analysis” leverages variation in treatment to control for both measured and unmeasured confounders. The “instrumental variable analysis” did not find a statistically significant association between lymphadenectomy and survival. So, what does all this mean? The authors conclude that lymphadenectomy confers a modest, if any survival benefit.
But, let’s take a quick look at one of the sub-analyses. If there’s anyone who doesn’t benefit from lymphadenectomy, it would be women with the lowest stage disease. However, at 1, 3, and 5 years, for all stages of endometrial cancer, the multivariate regression model and propensity score continue to demonstrate a significant survival benefit. Of course, the benefit is greater with higher stage.
While this is a lot of information on endometrial cancer to absorb in just one blog post, this new body of research presents significant contributions to thinking about how we create a standard of care for a disease and how important it is to continually push forward to understand and re-asses if our treatments are actually the best we can do for patients.