Localized Non-Small Cell Lung Cancer (NSCLC) with an EGFR Mutation

DR HERBST: Today I’m going to talk about the management of localized EGFR mutation-positive non-small cell lung cancer.

EGFR TKIs have really redefined treatment in EGFR mutated advanced disease. I show a timeline on this slide, and you can see that over 20 years, EGFR TKIs have really changed the course of care. You can see all the way back to 1997, when the first EGFR inhibitors went into practice, all the way to present day. And now, osimertinib is a third-generation EGFR TKI that has demonstrated a statistically significant and clinically meaningful improvement in PFS and OS versus comparator EGFR TKIs, those being erlotinib and gefitinib, in EGFR mutated advanced non-small cell lung cancer, with efficacy also seen in the central nervous system.

So the efficacy and safety profile of osimertinib in EGFR mutated advanced disease suggested that it may be an effective treatment for patients with early stage disease. So you can see from the earliest disease, now to the most specific settings of early disease, this drug has moved forward.

Well, what do we know about resectable lung cancer? Well, first off, lung cancer is the leading cause of cancer death worldwide, accounting for more than 1.7 million deaths. And about 85% of lung cancers are non-small cell lung cancer, with an estimated 30% of these patients presenting with resectable disease. As you can see on this slide, localized/early disease, Stage IB, and more regional/locally advanced disease, Stage II and Stage III.

Now the outcomes in early stage disease clearly need to be improved. Surgery is the primary treatment for patients with early stage non-small cell lung cancer. And a survival benefit is seen with adjuvant cisplatin-based chemotherapy and it's recommended for Stages II to IIIA disease, and selected patients with Stage IB disease. And as many of you know, results from large, randomized trials and meta-analysis showed a 5-year overall survival benefit with adjuvant chemotherapy in early stage disease. Those benefits were real, but small. Overall survival hazard ratio 0.89, disease-free survival favoring chemotherapy, hazard ratio 0.84.

If you want to look at it by 5-year recurrence in Stage IB resected disease, the 5-year recurrence is still 45%. In regional/locally advanced disease, in Stage II disease, 5-year recurrence rate is 62%, and Stage III disease, it's as high as 76%. So, clearly there’s an unmet need for better therapy.

Based on a limited number of studies, the prevalence of EGFR mutations appears to be broadly similar across disease stages. Why do I show you this? Because we know that we can find the EGFR mutations just as easily in Stage I, II, and III disease, as in Stage IV disease, where these are used routinely as first-line therapy. So why not bring them earlier?

Now what’s the prevalence of these mutations? That's shown on this slide. In an Asian population, as many as 30 to 40% of patients have EGFR mutations. In a Caucasian population, 10 to 20%. If EGFR TKIs were available in the resectable setting, a similar proportion of patients could possibly benefit. This is a reason why we did the ADAURA trial.

Phase III ADAURA Trial — Study Design and Characteristics of Enrolled Patients

DR HERBST: I’m going to tell you today about data that are still less than a year old for osimertinib as adjuvant therapy in patients with Stage IB to IIIA, EGFR mutation-positive non-small cell lung cancer after complete tumor resection, the ADAURA trial. I presented this at the ASCO meeting last May, a plenary talk. It was also presented in October at ESMO and published last year in The New England Journal of Medicine.

Well here’s the trial. It's a Phase III trial. Patients to be eligible, on the left, had to have completely resected R0 resections, either Stages IB, II, or IIIA, non-small cell lung cancer, with or without adjuvant chemotherapy. The key inclusion criteria are shown. I won’t read through all of them, except to point out that the patients had to have non-squamous non-small cell lung cancer, exons 19 or 21 mutations in EGFR; they’re shown. Brain imaging was done if not completed preoperatively. They must have a resection, as I mentioned, with negative margins. And the time between surgery and the randomization was 10 weeks without adjuvant chemotherapy or 26 weeks with adjuvant chemotherapy. So if patients were considered candidates for chemotherapy, as I just showed you, a therapy that improves survival, they got it.

They were then randomized, based on 3 stratifications: stage, 1B versus II versus IIIA. I’ll show you in a moment we had equal numbers of each of those 3 stages. The EGFR mutation type, exon 19, deletion, or L85R in exon 21, and race: Asian versus non-Asian.

Then the treatment is shown, osimertinib at 80 mg once daily. The standard dose, the same dose that's used in metastatic disease, randomized 1:1 to a placebo, once daily, 682 patients randomized. People often ask, why placebo? Because other studies in this area with earlier drugs, erlotinib, for example, had not shown any benefit statistically in disease-free survival/overall survival. So it was felt that there was no known standard of care.

The planned treatment duration was 3 years. Treatment continued until disease recurrence, treatment was completed or discontinuation criteria were met. And the follow-up was as shown, until recurrence, then weeks 12 and 24, and then every 24 weeks till 5 years.

As I mentioned, the primary endpoint was disease-free survival in Stage II to IIIA patients and it was designed for superiority of DFS hazard ratio of 0.70, so a 30% improvement. Secondary endpoint DFS in the overall population, landmarks at 1, 2, 3, 4 and 5 years, safety and health related quality-of-life. I did not expect we’d have these data last year, but the Safety Committee, having a routine safety meeting, noticed superior efficacy in the trial and the trial was unblinded early. And what I report here is that analysis. At the time of the unblinding, all patients had been followed for at least 1 year.

Here again are the inclusion and exclusion criteria written out a little bit more clearly. I won’t go through this again, except to say that patients were taken from all over the world on this global study.

Here are the baseline characteristics of the patients. You can see the ratio of male to female were equal in both groups, about 30% male, more women on this EGFR mutation trial, that's to be expected. The median age similar between the 2 groups. There were smokers and non-smokers on the trial. As I mentioned, Asian and non-Asian, 64% of the patients were Asian. Performance status was either 0 or 1. All patients had brain imaging at randomization. People often ask by what modality? You can see it's about equal in both groups, about half MRI, half CT — this being a real-world trial. The staging at diagnosis, as I mentioned, equal, IB, II and IIIA in both groups. Histology, pretty much all adenocarcinoma, a few other. EGFR mutations, a few more exon 19s. And very important, adjuvant chemotherapy: 60% received adjuvant chemotherapy in both arms, equal across the arms.

Efficacy Data from the Phase III ADAURA Study Evaluating Adjuvant Osimertinib in Resected NSCLC with an EGFR Mutation

DR HERBST: Well, here are the data, and they, of course, are extraordinary. I expected this trial would be positive to some extent, an EGFR inhibitor versus a placebo in an EGFR mutated population, but at this magnitude — I mentioned that we were looking for a hazard ratio of .7, 30% benefit. If you look at the primary population, the II and IIIA patients, the hazard ratio is 0.17 or an 83% benefit. And you can see that on the curves shown here, they separate early and the continue apart all the way forward. If you add in the IBs, a group of patients who would have done extremely well even on their own, more than 50% are probably cured, you can see the hazard ratio goes up only subtly to 0.2 or an 80% benefit.

So, really very promising results and very exciting results at the time of this analysis.

Now what about stage? If you look at disease-free survival by stage, you can see that all 3 stages have a significant benefit. I’ve broken it down in this slide by stage. On the upper left, Stage IB. The lower left, Stage II and Stage III on the bottom right. The blue being treatment arm with osimertinib; the yellow being the placebo control. Hazard ratios of 0. 5, 0.17 and 0.12 for Stage IB, II and IIIA, respectively. And as you would expect, the patients at highest risk benefit the most, the IIIA patients.

Here’s a forest plot and it doesn’t get much better than this. With every one of the variables looked at, to the left of unity, to the left of 1 being positive; sex, age, smoking history, race, stage — I just showed you that — EGFR mutation type, or the presence or absence of adjuvant chemotherapy. You can see yes versus no. They’re statistically the same, but actually numerically, those who got adjuvant chemotherapy might have even got numerically just a slight bit better, but really no differences.

Now everyone always asks what about survival? Of course, survival is the ultimate endpoint to any trial. It was very early — this is just a snapshot. This is in the appendix of the New England Journal of Medicine paper. There’s only maturity here of 5%. But you can see even at this early point, the hazard ratio is 0.4. Again, way too early to make any conclusions. But at least now, at this time, the blue curve being osimertinib, the yellow being the control, they’ll be more on this in years to come.

Now, as I mentioned, the reason why this was felt to be a reasonable placebo controlled trial, is because other studies had not translated into approval or a change in practice using either erlotinib. There was a study, a single-arm study using osimertinib. Several studies using gefitinib. So, none of these studies, even though some of them hinted at some disease-free survival benefit, none of them achieved statistical significance and went forward to an approval.

So, what else are we looking at in the ADAURA trial? I’ll show you a little bit now about local versus disease recurrence as we await the survival results. Sites of disease recurrence; I think that's important. Lung cancer tends to metastasize to the brain, the liver and the bones.

What about subsequent therapies? What do patients get after they were on osimertinib? And of course, quality of life being extremely important in a placebo-controlled adjuvant trial. I’ll show you some of these data now.

Well, type of recurrence, as I mentioned, is a key consideration. Local/regional recurrence is associated with longer post-recurrent survival than distant recurrence. And that makes sense, if you have a locoregional recurrence you can usually do a locoregional therapy whether it be more surgery or radiation. And on the bottom, I show that CNS is a common site of disease recurrence among patients with EGFR mutated lung cancer. And in the RADIANT, 37% of recurrence in patients with non-small cell lung cancer treated with erlotinib occurred in the brain.

Osimertinib, as I mentioned, has been shown to be highly brain penetrant, with high exposure levels. So that's one of the reasons we felt going into this trial that there was a good chance that it might be more active. And it's also a more selective EGFR inhibitor as well, meaning less toxicity.

In the advanced non-small cell lung cancer setting, first-line osimertinib demonstrated superior overall survival and a 52% reduction in risk of CNS progression compared to erlotinib or gefitinib.

So here’s the types of disease recurrence. You can see 11% of the patients in the osimertinib arm had disease-free survival events, disease recurrence or death, versus 46% in the placebo arm. And you can see that, that's the purple here.

If you look in the osimertinib arm, the majority of patients with recurrence had locoregional recurrence, shown here in blue, and only 38% of patients had distant metastatic recurrence. If you look in the placebo arm, you can see only 39% of patients had locoregional recurrence and you can see here 61% had recurrence in a more metastatic way, the more difficult type of recurrence to treat.

Looking at it another way, you can see disease recurrence, only location, where the blue is or the osimertinib-treated patients, the yellow/orange are the placebo patients, and you can see the 11 versus 46, I just mentioned, on the top line. And then you can see the main sites of recurrence being lung, lymph nodes, brain. Look at the brain: 10% versus 1% for the patients on placebo versus osimertinib. A big improvement for being on osimertinib. And you can see visceral recurrence here and so forth. In fact, if you look at it, 45 patients had CNS/DFS events at this early analysis, 6 on osimertinib and 39 on placebo. And if you look here, I’ve highlighted with a box, in the overall population, 10% of the patients on a placebo arm recurred in the brain and only 1% on the osimertinib selected EGFR inhibitor arm.

And here’s a disease-free survival curve in the overall population, now for disease-free survival as the brain of the site of a recurrence. The top curve is the osimertinib arm, the bottom is the placebo. The hazard ratio here is 0.18, p-value less than 0.0001. So an 82% improvement in disease-free survival, recurrence-free survival in the brain taking this drug.

Tolerability and Toxicity of Osimertinib in the ADAURA Trial; Impact of Adjuvant Chemotherapy on Disease-Free Survival Outcomes

DR HERBST: Now what about safety? The safety is shown here, osimertinib in blue on the left. Placebo on the right. And you can see if you look AE leading to death, the third line down, 0 on the osimertinib arm, 1 patient in the placebo arm. Serious adverse events, 16% versus 13%, AE leading to discontinuation, 11% versus 4% in favor of the osimertinib, and any AE leading to dose reduction, 7% versus 1%. I will contest that these are very favorable numbers, considering the drug’s been compared to a placebo.

And this tornado plot helps look at things a little bit differently. I’ve talked to many patients since this trial has started and since we’ve unveiled the results, and they remind me that these can be significant for them: the skin, the dry mouth, the nails, the diarrhea, but, fortunately, they’re all mostly Grade 1 and 2, with very few, if any, Grade 3 which is good. The osimertinib is on the left. The placebo is on the right.

The most serious concerns were interstitial lung disease. Actually, we did not allow any adjuvant radiation in this trial because we were concerned about interstitial lung disease, but there were no Grade 3 events, and it was only reported in 10 patients in the osimertinib arm, 3%. And QTC prolongation, which have been seen with osimertinib in animal studies, only 7% of the patients had it in the osimertinib arm, all low grade.

Now, a few questions. Adjuvant chemotherapy use. As I mentioned, 60%, 410 out of the 682 patients, received adjuvant chemotherapy for a median duration of 4 cycles, consistent across both arms. The majority of patients received platinum-based chemotherapy, almost all, most with Stage II to IIIA disease, 76% and fewer with Stage IB disease, 26%. Exactly as you would expect in Stage IB, many patients would elect or their physicians might elect not to give them chemotherapy.

Adjuvant chemotherapy was more frequent in patients age less than 70 and in patients enrolled in Asia. It was not influenced by performance status. And you can see here Stage I, 26%; Stage II, 71% and Stage IIIA, 80% received adjuvant chemotherapy. So, it was used selectively.

Here are the curves. Very important, on the left, the ADAURA curves, now disease-free survival for the group broken down with chemotherapy on the left, without chemotherapy on the right. You can see with adjuvant chemotherapy, the hazard ratio is 0.16; without adjuvant chemotherapy, the hazard ratio is 0.23. In both cases, you see the landmark at 2 years, 89% versus 49% with adjuvant chemotherapy, 89% versus 58% without adjuvant chemotherapy.

Now here you can see just the forest plot, adjuvant chemotherapy, the overall is shown above, .2, .19. But if you look at Stages II to IIIA with adjuvant chemotherapy, the hazard .14; without adjuvant chemotherapy, hazard .15.

Stage IB, there are too few patients with adjuvant chemotherapy to look at that, but without adjuvant chemotherapy, 0.38. And Stage II disease with adjuvant chemotherapy, 0.15; without adjuvant chemotherapy, 0.2. And Stage III with adjuvant chemotherapy, 0.13; without adjuvant chemotherapy, a small number, 0.10. Which has begged some to say, should we do a trial in those patients and avoid the chemotherapy all together? And that's something that should be looked at.

Here’s disease-free survival on patients with and without adjuvant chemotherapy in the Stage IB patients. With adjuvant chemotherapy on the left and without adjuvant chemotherapy on the right. You can see with adjuvant chemotherapy, it's such a small number of patients — it's only 15 patients — but, still, you can see the osimertinib curve on the top, but there’s no significance in a trial like this with small numbers. But on the right, without adjuvant chemotherapy, hazard ratio and significant.

Here are the curves for Stage II with adjuvant chemotherapy on the left; without adjuvant chemotherapy on the right. As I mentioned, hazard ratio of .15 with adjuvant chemotherapy; .2 without adjuvant chemotherapy.

Here on the next slide are the patients with Stage IIIA disease with adjuvant chemotherapy, hazard ratio 0.13; without adjuvant chemotherapy, hazard ratio of 0.10.

Finally, patient reported outcomes are just so important. And Margarita Majem presented this on behalf of our team at the World Lung Conference in Singapore, at least virtually in Singapore, in January 2020. Patient reported outcomes from ADAURA: osimertinib as adjuvant therapy in patients with resected disease.

Here, health-related quality of life was collected using a metric, a platform known as SF-36. You can see that there both physical components and mental components that were part of this assay. And you can see that it was collected post-surgery in the patients, with or without chemotherapy, until the time of disease recurrence.

The compliance on this was excellent. This amazes me and it really speaks to all the investigators on this trial. You could imagine how many different languages this had to be in, given to patients in multiple sites. But you can see across the range of the trial, well over 90% compliance, trailing off a little bit at the end.

And here you can see baseline scores for physical and mental components in the different domains. And you can see the baseline scores were pretty much the same for the osimertinib arm or the placebo arms. So we’re starting out about the same.

Here you can see the data adjuvant mean change in SF-36 physical and mental component summary scores. And what you can see in disease-free patients receiving osimertinib, SF-36 physical and mental scores were maintained from baseline to week 96 with no clinically meaningful differences observed compared to the placebo arm. There were some very slight differences, but this scale has been looked at, and based on historical data the changes seen were minimal and that felt to have a measurable effect on patient quality of life.

And here you can see the adjusted mean scores. In a number of different domains, the osimertinib versus the placebo, and again, it's a little bit worse in the osimertinib arm but not to the level where it would felt to be a significant change in quality of life. Which is pretty good considering the patients have gotten the drug for as much as 3 years versus a placebo.

Here you can see the time to deterioration of physical and mental components and to summarize during the disease-free period, the majority of patients, 80% of patients across both arms, did not experience a clinically meaningful deterioration in physical or mental quality of life measures. And you can see that below. Here are the curves and you can see the lines are pretty much indistinguishable from each other.

FDA Approval of Osimertinib as Adjuvant Therapy in Early-Stage NSCLC with an EGFR Mutation; Future Direction of Clinical Research

DR HERBST: So to conclude, adjuvant osimertinib is the first targeted therapy in a global trial to show a statistically significant and clinically meaningful improvement in disease-free survival, the patients with Stage IB, II, or IIIA, EGFR mutated non-small cell lung cancer.

Overall, there was an 80% reduction in the risk of disease recurrence or death with osimertinib. Hazard ratio of 0.20 across all 3 groups.

Osimertinib versus placebo, disease-free survival rate at 2 years was 89% versus 52%, respectively.

Adjuvant osimertinib demonstrated a clinically meaningful improvement in CNS PFS compared with placebo.

The safety profile is consistent with the established safety profile of osimertinib, with mild EGFR TKI class effects reported. Median duration of exposure to osimertinib was 22 months.

Furthermore, a clinically meaningful DFS benefit with osimertinib was observed in patients with or without adjuvant chemotherapy. DFS hazard ratio of 0.16 and 0.23 respectively, regardless of disease state. Overall, health-related quality of life was maintained with adjuvant osimertinib treatment with no clinically meaningful differences versus placebo, despite prolonged treatment.

So to conclude, adjuvant osimertinib was proven to provide a highly effective, practice-changing treatment for patients with Stage IB, II, or IIIA, EGFR mutant non-small cell lung cancer after complete tumor resection. And on December 18, 2020, the FDA approved osimertinib as adjuvant after tumor resection in patients with non-small cell lung cancer who’s tumors have EGFR exon 19 deletions or exon 21 L858R mutations, as detected by an FDA-approved test. This was based on this trial.

Finally, in summary, this really talks about how biology has spoken. I actually remember my late mentor, Josh Fidler shown on this slide, who always taught us that the biology of a tumor will help with its treatment and that's true more than ever today.

We’ve taken, as you can see in the middle, patients with lung cancer and combined surgery and adjuvant chemotherapy, and now added-in the targeted therapy, in those patients who have EGFR mutations. By bringing in the targeted therapy early on, we now have brought our best therapies to the early disease, preventing metastases to the brain, liver and bone, with maintained quality of life.

Next steps will be the NeoADAURA trial, which is neoadjuvant therapy with EGFR inhibitors pre-surgery. The LAURA trial, osimertinib being used post-chemoradiation in Stage III.

Combination studies to look for even better activity. And of course other agents could be used, using the same paradigm. You could imagine that we’re going to start testing our patients with next-generation sequencing post-resection, and I can’t imagine why we can’t start to look at other inhibitors like ALK and RET, and who knows, maybe even RAS inhibitors someday.