Pleural mesothelioma is generally treated by extended pleurectomy decortication, and there has been little improvement in systemic treatment of early-stage, resectable mesothelioma, which has led to the recommendations of maximum cytoreduction. U.S. and European guidelines, as well as an international consensus statement, support this approach, but it has never been tested in a randomized, controlled trial.
Now it has, and the result is surprising: The conclusion was uncomfortable for Eric Lim, MD, who presented the results of the MARS2 trial at the annual World Conference on Lung Cancer. “Ladies and gentlemen, as a surgeon standing here, you have no idea how much it pains me to conclude that extended pleurectomy decortication, an operation that we have been offering for over 70 years, has been associated with a higher risk of death, more serious complications, poorer quality of life, and higher costs, compared to mesothelioma patients who were randomized to chemotherapy alone,” said Dr. Lim of the Royal Brompton Hospital, London, during his presentation.
Although the line drew laughter and applause from the audience, Paula Ugalde Figueroa, MD, who served as a discussant, raised some concerns about the study. Disease presence in one hemithorax was assessed only by chest CT scan, which is notorious for underestimating the volume of disease during surgery. There was also no information on pleural effusion or how many patients received it prior to intervention. Existing guidelines suggest staging of mesothelioma should also use PET scans, and invasive mediastinal staging should be assessed with endobronchial ultrasound. “None of these were performed during the trial,” said Dr. Figueroa, who is an associate thoracic surgeon at Brigham and Women’s Hospital, Boston. “At this point, my question is, are the arms of this study well balanced in regard to tumor volume? We don’t know,” she added.
Dr. Figueroa noted that the 90-day mortality seemed higher than that seen in other studies. “So, does the surgeon’s experience and center volume affect the outcome of this study?” she asked. Dr. Figueroa personally made phone calls to the participating centers and found that 45% of the patients in the trial were treated at low-volume centers, defined by her as two to three patients per year. “Should we assume that their surgical outcomes are similar between those centers? In this trial, with approximately half of patients from low-volume centers, extended pleurectomy decortication for mesothelioma had no significant difference when compared to those patients that underwent chemotherapy alone. Would the outcome be different in exclusively high-volume centers?” she concluded.
The study randomized 335 patients to receive surgery and chemotherapy, or chemotherapy alone. They received two cycles of platinum-based chemotherapy and pemetrexed prior to surgery and up to four cycles after surgery. The average age was 69 years; 86.9% were male, and 85.7% of tumors were epithelioid only. Among those in the surgery group, 88.5% underwent extended pleurectomy/decortication, 8.3% underwent pleurectomy decortication, 1.9% underwent partial pleurectomy, 0.6% exploration with no pleurodesis, and 0.6% were classified as “other” surgery. Completeness of resection was R0 in 3.2% of surgeries, R1 in 80.9%, and R2 in 15.9%. In-hospital mortality occurred in 3.8% of patients, and postsurgical 90-day mortality was 8.9%.
Over the first 42 months of follow-up, the hazard ratio for overall survival was 1.28 in the no-surgery group (P = .03). “The survival was so good in this early-stage cohort that we had to extend the trial by 6 months to get the prerequisite number of deaths, underscoring the phenomenal importance of having a randomized comparative cohort for all future studies on surgery for mesothelioma,” said Dr. Lim.
After 42 months, there was no survival difference between the two groups (hazard ratio, 0.48; P = .15). Dr. Lim attributed the change at 42 months to the fact that only 15 patients remained in each arm at that stage. There was no significant difference between the two groups with respect to progression-free survival.
The survival benefit of the no-surgery group was sustained after additional analyses, including adjustment of the number of first-line chemotherapy cycles and immunotherapy received after completion of the trial protocol.
Adverse events were more common in the surgery group (incidence rate ratio, 3.6; P < .001), including any cardiac disorder (IRR, 2.73; 95% confidence interval, 1.11-6.67); any infection or infestation (IRR, 1.99; 95% CI, 1.33-2.99); any respiratory, thoracic, or mediastinal disorder (IRR, 2.40; 95% CI, 1.52-3.80); and any surgical or medical procedure (IRR, 2.23; 95% CI, 1.04-4.78). The EORTC quality of life score favored the nonsurgery group at 6 weeks, but there was no significant difference at other time points.
Dr. Lim and Dr. Figueroa have no relevant financial disclosures.