SAN FRANCISCO – Patients undergoing had less morbidity and pain and similarly good oncologic outcomes, when the surgery was performed by robot-assisted laparoscopy instead of by the open technique, a phase 3 clinical trial has found.
Investigators of the ROBOT (Robot-assisted Thoracolaparoscopic Esophagectomy vs. Open Transthoracic Esophagectomy) trial, led by Pieter C. van der Sluis, MD, a surgeon at the University Medical Center Utrecht, the Netherlands, randomized 112 patients with resectable esophageal cancer to open transthoracic esophagectomy – considered to be the gold standard – or robot-assisted minimally invasive thoracolaparoscopic esophagectomy.
Dr. Pieter C. van der Sluis
“Robot-assisted minimally invasive thoracolaparoscopic esophagectomy versus open transthoracic esophagectomy improves postoperative outcome. There were no differences in oncologic outcomes, and our oncologic outcomes were in concordance with the highest standards nowadays,” Dr. van der Sluis summarized. “This trial provides evidence for the minimally invasive approach over the open approach, and especially the robot-assisted minimally invasive esophagectomy.”
The investigators will report a full cost comparison separately. “We see that costs are lower, though not significantly lower, with the robot,” he said, giving a preview. “We are going to show that the real costs of the operation are in the complications. When you have complications that involve the ICU and reoperations, some patients are in the hospital for months after the surgery. So by investing a little extra money in the surgical procedure, you might actually get it back by reducing the complications.”
When asked by an attendee why the trial did not compare robotic esophagectomy with thoracoscopic esophagectomy, Dr. van der Sluis noted that such comparison is complicated by many factors; for example, the challenge of finding surgeons skilled in both techniques, and the likelihood of small differences in outcomes, potentially requiring enrollment of thousands of patients to have adequate study power. “We concluded that such a trial might not be feasible,” he said.
Parsing the findings
“The complication rates [in this trial] are very high in the robotic and open groups, much higher than reported in some well-controlled prospective and retrospective studies,” commented session attendee Kenneth Meredith, MD, FACS, professor at Florida State University, Sarasota, and director of gastrointestinal oncology, Sarasota Memorial Institute for Cancer Care.
He wondered how extensive the investigators’ experience with robotics was and how many cases they had done on their learning curve. Data from his group suggest that surgeons must perform 29 cases of robotic esophagectomy before the complication rate drops (Dis Esophagus. 2017;30:1-7).
“That’s more then half of the patients in the robotic arm of their study,” he noted in an interview. “I find this needs to be explained. If the authors are past their learning curve, why were the complication rates so high?” Additionally, the 80% rate in the open group “is among the highest I’ve seen in many years.”
The lack of significant differences in complete resection rate and in lymph node harvest was also surprising, as he and other robotics users have found that this technique can improve these outcomes, Dr. Meredith added. This could likewise be a learning curve phenomenon.
Although ROBOT’s comparison of robotic with open esophagectomy is relevant, “it would have been more relevant to compare robotic to minimally invasive esophagectomy [MIE],” he maintained, as MIE has been shown to improve outcomes relative to open surgery (Lancet. 2012;379:1887-92).
“There are many high-volume centers in MIE but not necessarily robotics. The two are often mutually exclusive, and a multicenter trial in which each center performs high volumes of their respective technique, rather then mandating each center perform an operation they may not be facile in,” would be practical, Dr. Meredith concluded.
Study details
“The main objective in our trial was to reduce surgical trauma and reduce the percentage of complications,” Dr. van der Sluis told attendees of the symposium, sponsored by the American Gastroenterological Association, the American Society for Clinical Oncology, the American Society for Radiation Oncology, and the Society of Surgical Oncology.
Results showed that compared with peers in the open surgery group, patients in the robotic-assisted surgery group specifically had a lower rate of pulmonary complications (32% vs. 58%, P = .005), largely due to a reduction in rate of pneumonia (28% vs. 55%, P = .005), and a lower rate of cardiac complications (22% vs. 47%, P = .006), almost entirely due to a reduction in rate of atrial fibrillation (22% vs. 46%, P = .01).
There was a trend toward fewer wound infections with robotics (4% vs. 14%, P = .09), with a large difference in thoracic wound infections (0% vs. 9%, P = .06).
The two groups were statistically indistinguishable on rates of anastomotic leakage (24% and 20%) and recurrent laryngeal nerve injury (9% and 11%). The fairly high rate of anastomotic leakage was likely due to the center’s use of cervical anastomosis at the time of the trial, according to Dr. van der Sluis; they have since started using thoracic anastomosis, and will report results with that technique soon.
There was also no significant difference between groups in the rate of in-hospital mortality (4% with robotic surgery and 2% with open surgery), median hospital length of stay (14 and 16 days), and ICU length of stay (1 day in each group).
Patients in the robotics group more commonly had functional recovery within 2 weeks (70% vs. 51%, P = .04). And on the Quality of Life Questionnaire Core 30, they had better scores for health-related quality of life at discharge (57.9 vs 44.6, P = .02) and at 6 weeks (68.7 vs. 57.6, P = .03), and for physical functioning at discharge (54.5 vs. 41.0, P = .03) and 6 weeks (69.3 vs. 58.6, P = .049).
The two groups were similar on rates of R0 resection (93% and 96%) and median number of lymph nodes retrieved (27 and 25), reported Dr. van der Sluis. Pain during the first 14 days after surgery was lower for the robotics group (P = .003).
With a median follow-up of 40 months, the robotics and open groups did not differ significantly on disease-free survival (median, 26 and 28 months) and overall survival (not reached in either group).
Dr. van der Sluis disclosed no relevant conflicts of interest.
SOURCE: van der Sluis PC et al. 2018 GI Cancer Symposium, Abstract 156148