ADAURA Trial: Breaking Through the Hype

H. Jack West, MD


June 05, 2020

Highly anticipated data from the phase 3 ADAURA trial were presented at the virtual American Society of Clinical Oncology (ASCO) meeting, held May 29-31. In this interview, H. Jack West, MD, talks with Lecia V. Sequist, MD, MPH, director of the Center for Innovation in Early Cancer Detection at Massachusetts General Hospital in Boston, about the impact of these practice-changing results for patients with EGFR mutation–positive non–small cell lung cancer (NSCLC).

H. Jack West, MD: The ADAURA trial was one of the plenary session highlights at ASCO and the highest-profile discussion within the lung cancer community. Even for those of us who are concerned by its shortcomings, we acknowledge that it is practice-changing and needs to be incorporated in some way.

To briefly recap, patients with stage IB-IIIA resected NSCLC harboring an activating EGFR mutation were randomized on a 1:1 basis to receive either osimertinib at 80 mg per day or placebo for up to 3 years. Patients were allowed, but not required, to have undergone adjuvant chemotherapy. The primary endpoint was disease-free survival (DFS), which was met to such a strong degree that in April 2020, upon recommendation by the Independent Data Monitoring Committee, the trial was unblinded early.

In patients with stage II to IIIA, the original focus group, we saw DFS with a hazard ratio of 0.17; when including patients with stage IB disease, we saw DFS with a hazard ratio of 0.21. This huge DFS benefit was seen across almost all clinical subgroups. Overall survival (OS) was obviously compromised and perhaps not even feasible to get a result on.

I think the main concerns are whether DFS in a setting where you're giving a targeted therapy to patients with a driver mutation is really a test of anything that we shouldn't expect to happen, whether DFS is a high enough bar, and whether we should be asking for an OS difference. I think that's the division between the proponents and the detractors, in addition to its exceptional cost.

Lecia, what are your thoughts? Will DFS translate to OS?

Lecia V. Sequist, MD, MPH: We are finding ourselves in a situation where we honestly don't have a lot of experience. You and I started practicing before adjuvant chemotherapy was standard. We saw a number of trials for adjuvant chemotherapy come through, most positive, some negative. We debated certain factors of those trials. Even when those trials were positive, though, the marginal benefit was much smaller.

Are we helping to cure more people or are we just prolonging the inevitable?

One thing I've been thinking about: In an adjuvant study, I'm not sure how unblinding at this premature point is going to affect overall survival. We need to understand from the study team what happened to those patients who, upon unblinding, found out that they had been on placebo. Roy Herbst, MD, PhD, showed us that, according to the DFS data, 55% of patients on placebo had already recurred at the time of the data cut. In a metastatic clinical trial, patients can cross over to the other arm after unblinding. Here, though, the majority of patients had already recurred, so "crossing over" is a term that doesn't really apply. For the 45% of patients who hadn't recurred at the data cut, they were at least a year out, and I'm not sure if starting adjuvant osimertinib more than a year after surgery is going to happen, or if we can consider it to even be the same thing as immediate postoperative therapy.

So, I'm struggling in my mind to piece together exactly how much the OS analysis is going to be compromised, as you said.

West: It would be instructive to see the outcomes of those patients who relapsed and how they fared based on what they received. It would also be important to clarify what proportion received EGFR tyrosine kinase inhibitor (TKI) therapy, particularly osimertinib. Perhaps the trial we need is one of upfront osimertinib as an adjuvant therapy compared to surveillance with immediate introduction of osimertinib at the time of relapse.

Another concern of mine is that these trials are typically done in places where management outside of them is variable and often suspect; it heavily biases the trial in favor of the investigational arm because patients are given what would be considered substandard care outside of the trial parameters.

In this trial, approximately 45% of patients did not receive adjuvant chemotherapy. It would be inappropriate to not have patients with at least stage II or IIIA disease receive adjuvant chemotherapy. That is an evidence-based standard of care with a modest survival benefit, but a survival benefit that is more than EGFR TKI therapy and definitely more than placebo. Dr Herbst showed curves that demonstrated that the benefit was comparable in patients whether they received adjuvant chemotherapy or not. But should it not be a concern to us about the overall quality of care when nearly half of patients are failing to get a treatment that we would consider to be a clear standard of care?

Sequist: We need to see more granularity, especially about why 45% of patients did not receive adjuvant chemotherapy. I agree that stage II and stage IIIA patients failing to receive adjuvant chemotherapy is not consistent with today's standard of care. I'd like to see a stage breakdown for those who did not get chemotherapy.

To your point, knowing what patients in either arm received as treatments at the time of recurrence will be key. The ADAURA trial began before osimertinib was shown to be superior to other TKIs in the FLAURA study and approved by the FDA as a first-line therapy. Depending on the timing, some patients may have received a first- or second-generation TKI upon recurrence, even in the United States. Furthermore, first-line osimertinib is not available in many countries, and that can influence the results.

With adjuvant therapy, especially in a genotype-selected population where the cells are so biologically predisposed to treatment response, the question becomes whether we can change the natural history with adjuvant TKIs. From my experience with the SELECT trial, which assessed adjuvant erlotinib in patients with fully resected EGFR-positive NSCLC, patients who were re-treated upon recurrence with erlotinib did well, with some patients on first-line erlotinib for up to 5 years, which is not typical. This is anecdotal, of course, but it really stuck out to me how well these patients did on first-line treatment, especially with a first-generation drug. Did the adjuvant therapy change the natural history in some way? We've seen that type of phenomenon with immunotherapy, where even patients who do not respond appear to have a much more indolent course as if the natural history had shifted.

I think the core of this argument about OS versus DFS is: Are we helping to cure more people or are we just prolonging the inevitable? We still have a lot to learn.

West: We also have to think about whether it's socially responsible to embark on a treatment approach that costs over $200,000 per year per person, which on a 3-year course equates to at least $600,000 if they don't progress. This approach may be reasonable if it translates to a survival benefit, but we don't know that yet. Usually with such long-term treatment, we're essentially not eradicating but suppressing disease, and if you stop at 3 years, you've just restarted the clock you delayed.

Sequist: The cost is astronomical. It's hard to draw a line in the sand around this particular treatment for this particular group of patients when so many other treatments are equally expensive and more toxic, with potentially less gain. Globally, there is such a variety in the accessibility of care for cancer patients, largely because of the cost of cancer treatments and the type of health system they live under. It is hard to comprehend that patients in some countries can't access treatments for financial reasons, and it's equally hard to see patients who can't afford insurance in the United States or even go bankrupt for health-related reasons. Treatment costs are a constant struggle that oncologists face, and not just with adjuvant osimertinib.

West: When you're talking about immunotherapy or chemoimmunotherapy, you have something to follow in your patients who you know have the disease, whereas with adjuvant osimertinib, you know that some patients are cured, especially as you move into earlier stages. You're overtreating with an expensive medication with ongoing adverse effects.

Patients are bearing a portion of the costs, yet we haven't shown that it translates to better OS, which is the endpoint we should be focused on. We have become inappropriately complacent about only using DFS; it's convenient, faster, and easier than OS, but it shouldn't allow us to take our eyes off the ball.

Sequist: We obviously need to see OS data, but it will take years to mature. What feels uneasy to me, though, is waiting for that OS data before talking to patients about adjuvant osimertinib, especially now that we have the DFS data and it is so markedly positive. But I think your point is that the trial should have been designed in a way where we didn't have access to the data until we had the definitive results. We need to keep demanding the answers to the questions you're bringing up, but in the meantime, I don't think we should hold our patients hostage until we have that data. We do a lot of things in the metastatic setting that are a little bit ahead of the data or sometimes completely unfounded by the data. As you mentioned, though, the difference here is that some patients, unbeknownst to us, may be cured by surgery alone or with chemotherapy, and we are overtreating them by giving osimertinib when in reality they need nothing further.

There is a lot to struggle with, and I don't have a great answer. You bring up good points, but these data are compelling enough that it reaches a threshold in my mind where we can't ignore it. We can't let perfection be the enemy of the good.

West: Even though this has been a pro/con kind of discussion, we both recognize the complexity of the issue. This trial's result is remarkably good, even if the endpoint is suboptimal, in my opinion. But we need to have the opportunity to either demand increased rigor or go down a slippery slope that allows for lower standards moving forward. We can't pursue gamesmanship in how trials are run that force our hands. I would say I feel manipulated in that, yes, I'm going to offer osimertinib to my patients in this setting, but I don't feel good about having a conversation with them where I can't speak to whether there is a strong OS benefit. I would have hoped that we could do that, but we can't. Nonetheless, this is poised to change practice, arguably not only for EGFR mutation–positive lung cancer but also for the whole paradigm of molecular oncology and targeted therapies extending into earlier-stage disease.

Sequist: This is a turning point for EGFR-mutated lung cancer, one that has been more than 10 years in the making. Personally, it is very satisfying to see the first properly designed randomized trial of adjuvant TKI hit the goal. I'm also happy for patients who will have the opportunity to take advantage of this strategy.

H. Jack West, MD, associate clinical professor and executive director of employer services at City of Hope Comprehensive Cancer Center in Duarte, California, regularly comments on lung cancer for Medscape. Dr West serves as web editor for JAMA Oncology, edits and writes several sections on lung cancer for UpToDate, and leads a wide range of continuing education programs and other educational programs, including hosting the audio podcast West Wind.

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.