Thread by @Garth_Nicholas1

November 8: Consorting with MARS

This year for Lung Cancer Awareness Month #LCAM I’m going to summarize 30 important lung cancer trials over 30 days. These posts are directed at non-medical professionals, with descriptions of the results and of what makes a good trial.#lcsm 1/15

Picking up on 6 November’s discussion of primary end points, today’s trial has a somewhat unusual endpoint. While most trials have a patient outcome, such as Time To Recurrence, this trial was undertaken to study the feasibility of doing a larger study. 2/15

Pleural mesothelioma is a tumour of the lining of the lung which is almost invariably associated with asbestos exposure. It is almost never discovered at a curable stage, and standard palliative treatments included chemotherapy and radiation. 3/15

Non-randomized case series of patients showed encouraging survival times when patients were also treated with an extrapleural pneumonectomy (EPP). EPP is a huge surgery, entailing removal of the entire lung and pleura in one piece (en bloc). 4/15

We saw in our discussion on 3 November that fitter patients have better outcomes. So it was controversial whether EPP actually improved survival, or whether patients selected for EPP were fitter to begin with, and therefore destined to have better survival. 5/15

This is a great question for a randomized trial, and the Mesothelioma and Radical Surgery (MARS) trial was born.
But the investigators had a problem: patients suitable for EPP are rare, and a properly-done trial would require almost 700 participants. Was this possible? 6/15

So this trial was designed to study feasibility. The investigators developed the protocol, brought 12 hospitals on board, and decided that they would declare a larger trial feasible if they were able to randomize 50 patients within one year. 7/15

To be eligible, people had to complete chemo and then have their candidacy for EPP assessed by a multidisciplinary team. If eligible they were then randomized to EPP or no EPP. Postoperatively, patients were meant to get radiation treatment. 8/15

The attached diagram is called a Consort Diagram. Its purpose is to account for the flow of all trial patients, assuring the reader that everyone is accounted for. In this case it shows a grueling process of attrition where 301 patients are whittled down to only 50 randomized.

Of 24 randomized to EPP, 16 actually had the surgery, of whom only 8 had the intended post-op radiation. All of this took place over 3 years of enrollment. Based on this, the primary endpoint was not met, and the authors concluded that a 700-patient study wasn't feasible. 10/15

92% of randomized patients were men, which probably reflects occupational exposure to asbestos in industries where most workers were traditionally male. This doesn’t really factor into our discussion, but gives me an excuse to post this picture. 11/15

Although the main purpose of the study was not to describe outcomes, data about survival was collected. The EPP arm had numerically worse survival than the no-EPP arm, although the difference was not statistically significant (p=0.082). 12/15

Single arm studies of intense therapies, like EPP, often show encouraging outcomes versus standard treatments. But it isn’t possible to tell whether those outcomes are related to the new therapy or to the selection of very fit patients who would have done well with any treatment.

Although not the definitive randomized trial of EPP, this is the best randomized data that we have on this question. Enthusiasm for EPP never recovered after this, and improving palliative systemic therapies over time have made it less appealing still. 14/15

Later this month we'll discuss immunotherapy and mesothelioma.
Tomorrow we’ll look at an example of how a negative trial can still give useful information, and will also discuss interim analyses.
See you then! 15/15