Surgery in Recurrent Ovarian Cancer Called Into Question

Alexander M. Castellino, PhD

November 13, 2019

A management strategy for women with recurrent ovarian cancer has been called into question after data from a prospective clinical trial showed no benefit and, indeed, suggested harm.

The strategy is cytoreductive surgery (secondary surgery) followed by chemotherapy.

The new results, from the first prospective trial to address this issue, show that for patients who underwent secondary cytoreductive surgery followed by chemotherapy, survival was not improved in comparison with patients who received chemotherapy alone.

Indeed, the patients who underwent secondary cytoreductive surgery were at a 29% increased risk for death.

The data come from the Gynecologic Oncology Group (GOG)-0213 (NCT00565851) trial and were published online November 13 in the New England Journal of Medicine.

"It is our obligation to test dogma," lead investigator Robert L. Coleman, MD, professor of gynecologic oncology and reproductive medicine at the University of Texas M.D. Anderson Cancer Center, Houston, told Medscape Medical News.

"This practice [of performing cytoreductive surgery followed by chemotherapy] is often held as an attractive and beneficial option for selected patients with recurrent disease," he added.

Indeed, the National Comprehensive Cancer Network (NCCN) lists it as an option for patients who experience disease recurrence following a treatment-free interval of 6 months or longer after complete remission from previous chemotherapy.

"Based on these new data, the NCCN should reexamine and reconsider its recommendation," Coleman said. He has already changed clinical practice at his own institution on the basis of these data.

Before these new results came to light, Coleman had been offering secondary cytoreductive surgery to select patients who had experienced recurrence 12 months after initial chemotherapy, had one or two lesions, and were healthy. "Women with ascites, nonoligometastatic disease, and who had a short interval between frontline chemotherapy and recurrent disease were not good candidates for this approach," he said. He acknowledged that different physicians have different criteria for offering this approach to patients with recurrent disease.

Reactions From Experts

"This well-conducted study quite clearly reveals the lack of survival benefit associated with an attempt at secondary cytoreduction in recurrent ovarian cancer," commented Maurie Markman, MD, president of medicine and science, the Cancer Treatment Centers of America, Philadelphia, Pennsylvania. "Further, the study demonstrates the value of evidence-based trials to document the validity of claims for the utility of surgical interventions in cancer management," he told Medscape Medical News.

Also approached for comment, gynecologic oncologist Eloise Chapman-Davis, MD, at Weill Cornell Medicine and New York–Presbyterian, New York City, offered a different perspective.

"I believe this study helps to confirm the already known benefit of chemotherapy for recurrent ovarian cancer," she said. "Providers should also make patients aware of how the addition of newer targeted therapies such as bevacizumab has changed the landscape for the need for surgery with ability to obtain the same or improved overall benefit," Chapman-Davis said.

"In the time of minimally invasive surgery, I don't think we can say all surgical interventions will cause significant morbidity and poor quality of life. We need more corroborating data," she added.

Chapman-Davis told Medscape Medical News that secondary surgical cytoreduction should not be abandoned solely on the basis of this study. "We need to carefully select appropriate patients with good baseline performance status and minimal disease who may have the best chance of complete gross resection without affecting quality of life," she said.

Most oncologists would consider this approach only if recurrence occurred 6 months after completion of platinum-based chemotherapy, if the sites of recurrence are limited, or if complete gross resection to R0 can be achieved, she continued. "Improvement in survival was associated with patients who underwent secondary cyotreduction to no gross residual disease, identifying this as a possible prognostic factor," she said.

At the same time, there needs to be an understanding that chemotherapy will always be needed, she added.

Patients should be told that if ovarian cancer recurs, they will need more chemotherapy regardless of whether or not they undergo secondary surgery, Chapman-Davis explained. "This study highlights the importance of chemotherapy and possible clinical benefit of addition of bevacizumab to traditional chemotherapy in this setting," she said.

GOG-0213 Trial Details

The surgical objective of the open-label phase 3 global study was to find out whether secondary surgery followed by chemotherapy provided significant survival benefit for patients with platinum-sensitive ovarian cancer. The study also had a chemotherapy objective, which has already been published (Coleman RL et al. Lancet Oncol. 2017;18:779-91).

For the surgical objective, women with ovarian cancer that had recurred after responding to initial platinum-based chemotherapy and who were treatment free for at least 6 months were randomly assigned to receive either surgery followed by chemotherapy (n = 240) or chemotherapy alone (n = 245).

The trial was initiated in 2007 when bevacizumab was still in clinical development. The protocol was amended in 2011 to allow physicians the choice of either paclitaxel-carboplatin or gemcitabine-carboplatin with or without bevacizumab for six cycles. Patients were allowed an additional two cycles if they experienced a partial response, and all patients who received bevacizumab also received maintenance therapy with bevacizumab every 3 weeks until progression or death.

GOG-0213 Results

The median time to initiation of chemotherapy was 40 days in the surgery group and 7 days in the no-surgery group. Coleman commented that this difference is within targeted timelines for postoperative chemotherapy initiation and does not skew the results.

Patients who undergo surgery need to recover, he pointed out. He noted that in other trials in the frontline setting, there has been no difference in outcomes as long as chemotherapy was initiated within 6 weeks.

The median follow-up was 48.1 months.

Median overall survival (OS), the primary endpoint of the study, was lower for patients in the surgery group (50.6 months vs 64.7 months for the no-surgery group; P = .08).

Three-year OS was 67% for patients in the surgery group and 74% for patients in the no-surgery group.

Median progression-free survival (PFS) was longer for patients who underwent surgery, but this difference was not statistically significantly (18.9 months vs 16.2 months for the no-surgery group; hazard ratio [HR]: 0.82; 95% confidence interval [CI]: 0.66 – 1.01). Three-year PFS was 29% for the surgery group and 20% for the no-surgery group.

Complete surgical resection was performed in two thirds (67%) of patients who underwent secondary surgery. "This highlights the importance of patient selection, which remains important bias in studies involving surgical interventions," Chapman-Davis said.

Although median OS was significantly higher for these patients compared with those who underwent incomplete surgical resection (median OS: 56 months vs 37.8 months; HR: 0.61; 95% CI: 0.40 – 0.93), the OS was not significant when compared with patients in the no-surgery group (56.0 months vs 64.7 months; HR: 1.03).

Coleman told Medscape Medical News that since 2007, when the study began, life expectancy for patients with ovarian cancer has increased substantially. Initially, the anticipated survival was set at a median OS of 22 months, but the actual median OS was 64.7 months, he pointed out.

"We underestimated the OS by approximately threefold," he said.

How Beliefs Start

Surgical studies are difficult to conduct, Coleman told Medscape Medical News. "There is an underlying bias that surgery is important in nearly every cancer, and in a curative setting it is difficult to argue that surgery in the frontline setting of ovarian cancer may not work," he said. However, he pointed out that the majority of patients (>75%) with ovarian cancer are diagnosed after the disease has already spread; in these patients, surgery alone does not result in cure, Coleman pointed out.

Surgical cytoreduction is a key component of the frontline treatment of ovarian cancer, but even in this setting, it is not supported by prospective trial data, Coleman noted.

Data from cohort and retrospective studies show that for patients who undergo complete surgical resection, outcomes are, on average, threefold better than for patients who undergo incomplete resection (ie, who have residual disease). This has been a surrogate for justifying surgery to patients, Coleman noted.

In the setting of recurrent disease, a few retrospective analyses have shown that secondary cytoreductive surgery may not be of benefit in this patient population, Coleman said. However, he emphasized that "this is the first randomized trial in this setting that shows that secondary cytoreductive surgery does not benefit these patients, and it challenges the current standard of care."

More Results Coming Soon

Coleman indicated that the results of three ongoing trials that are evaluating similar endpoints will be available soon. Of these, DESKTOP-III (NCT01166737) is closest to GOG-0213 in design. The other two studies are the SOCcer study from the Netherlands and the SOC 1 study (NCT01611766) from the Shanghai Gynecologic Oncology Group.

Chapman-Davis pointed out that DESKTOP-III revealed significant improvement in PFS among those who underwent secondary surgery followed by chemotherapy compared with those who received chemotherapy alone. However, DESKTOP-III differs from GOG-0213 with respect to patient-selection criteria and adjuvant therapy, and the overall survival data are still not mature, she said.

The trial was funded by the National Cancer Institute with supplemental funding support to GOG from Genentech, which also provided bevacizumab. Genentech played no role in the collection or analysis of the data or its interpretation. Coleman reports relationships with AstraZeneca, Merck, Tesaro, Medivation, Clovis, Gamamab, Genmab, Roche/Genentech, Janssen, Agenus, Regeneron, OncoQuest, V-Foundation, Gateway Foundation. Information regarding conflicts of interest for study coauthors is available in the original article. Chapman-Davis reports no conflicts of interest.

New Engl J Med. Published online November 13, 2019. Abstract

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: