EGFR-Targeted Approach Continues to Advance in NSCLC

Joshua Bauml, MD

Pivotal trials have laid the groundwork for available EGFR-targeted agents—afatinib (Gilotrif), erlotinib (Tarceva), and gefitinib (Iressa)—for patients with non—small cell lung cancer (NSCLC).

The EURTAC trial, for example, demonstrated the efficacy of erlotinib versus gemcitabine and docetaxel in 173 EGFR-mutant patients. Those who were treated with erlotinib had an overall response rate of 58% and median progression-free survival (PFS) of 9.7 months, compared with 15% and 5.2 months for gemcitabine and docetaxel. These findings led to the FDA approval of erlotinib in the first-line setting in 2013.

Today, ongoing studies are exploring how to target the most common resistance mutation that EGFR-mutant patients develop: T790M. Osimertinib (Tagrisso) was granted an accelerated approval by the FDA in November 2015 for patients with advanced EGFR T790M—mutant NSCLC following progression on a prior EGFR tyrosine kinase inhibitor (TKI).

Osimertinib is also being explored in a phase III study for use following tumor resection, with or without adjuvant chemotherapy, in patients with EGFR-positive NSCLC (NCT02511106). A second ongoing phase III study is comparing frontline osimertinib with gefitinib and erlotinib for patients with EGFR-mutant disease (NCT02296125).

Joshua Bauml, MD, shared the latest updates in EGFR-targeted therapy for NSCLC during the 2016 OncLive State of the Science Summit on Advanced Non—Small Cell Lung Cancer, which took place on September 17 in Philadelphia.

“Really the goal—the ‘holy grail’ of targeted therapy, so to speak—is to actually look at this disease like a chronic disease where we can give [patients] a pill and identify what is leading to development of resistance,” said Bauml, an assistant professor of Medicine at the University of Pennsylvania. “How we can prolong the response to these therapies is going to be critical. Once we identify resistance, we need to give a targeted therapy to treat that.”

OncLive: What were the main takeaways from your lecture on EGFR-targeted therapies?

In an interview with OncLive during the meeting, Bauml discussed the evolution of EGFR-targeted therapies, resistance mutations in patients with NSCLC, the evolution of next-generation sequencing (NGS), and ongoing trials that could have an impact in this space. Bauml: EGFR mutations are present in 10% to 15% of Caucasians. There’s a higher rate in East Asians, but this is the most common targetable mutation for NSCLC, and we have a lot of data at this point showing that these patients can respond very well to a targeted therapy. This is giving them a pill instead of giving them chemotherapy.

One of the studies I mentioned was the EURTAC study that compared erlotinib to platinum doublet chemotherapy in patients with EGFR-mutant adenocarcinoma. We found that there were markedly improved PFS and response rates. This is echoed in multiple trials that have compared TKIs to chemotherapy in EGFR-mutant lung cancers. You don’t necessarily see an overall survival advantage, but that’s probably due to the fact that the rate of crossover in these studies is approaching 100%.

The reason for that is quite simple. If you have a trial that you’re running, which is randomized between a pill that’s highly active in the cancer you’re treating and chemotherapy with its known side effects, how do you not put a single patient on that study? Unless any patient is randomized to chemotherapy, I know they will get the pill at crossover. That is, in fact, what we saw happen.

What about developing resistance to these EGFR-targeted therapies?

There are some questions regarding whether afatinib can prolong survival for exon 19 deletion patients. That analysis, which was done by Dr James Chih-Hsin Yang, et al in Lancet Oncology, has significant limitations. At this point, erlotinib, gefitinib, or afatinib are reasonable approaches for the first-line treatment of EGFR-mutant lung cancer.This is where things get a little bit complicated. At the time of diagnosis for a patient with EGFR-mutant lung cancer, their tumors are relatively homogonous. Any place you biopsy is going to have these same exon 19 or L858R mutation. At the time of resistance, however, that develops in a heterogeneous fashion. One part might be growing and have the T790M mutation—which is the most common resistance mutation to a first-generation TKI—but another area that’s stable may not have it or may have a different resistance mutation.

It is essential that we do a repeat biopsy at progression, and the purpose of that is to identify T790M when it’s present. The other reason that you have to do it is that sometimes you can have other histologic changes. There’s small cell transformation, which has been reported up to 14% of times of resistance to erlotinib. Our experience is a little bit lower than 14%, but you imagine that you would treat a small cell transformed lung cancer much differently than you would treat an EGFR-mutant adenocarcinoma.

Moving onto NGS, could you compare and contrast cell-free circulating tumor DNA NGS and tissue-based NGS?

For those patients who do harbor a T790M mutation, osimertinib has been approved with a very nice response rate and a disease control rate in excess of 80%, which is similar to what we see in the first-line treatment of EGFR-mutant lung cancer. That is a really great opportunity that we’re seeing—that we can give people 1 pill, and then hopefully another pill. As we advance the science, maybe we can give them a third pill, and that’s really what we are shooting for.NGS is sort of changing the paradigm of how we check for genetic abnormalities. The way we used to do it was with a pathological complete response-based assay. We would say, “I want to test for EGFR. Is that there? No? Okay. I’m going to check for KRAS. Is that there? No?”

Each time you do it, you have to have tissue, you have to wait for the results, and then you have to pick the next test. What NGS does is that it takes a small amount of tissue and at the same time, tests for potentially thousands of genes all off of that small piece of tissue. That means that we’re going from time and tissue being limiting factors, to actually now bioinformatics being the limiting factor. We will find out that a patient has an androgen receptor mutation. What am I going to do with that information? That means that the science now has to catch up with this technology.

The other thing that we are doing now is with cell-free DNA. This uses plasma, a blood test, or a urine test. We are using DNA that’s shed from the tumor. The reason why is that tumors are the things that are growing and dying the fastest in your body. If you pick up DNA in the blood, chances are it’s from the tumor.

Are there any ongoing studies in this space that you are particularly excited about?

What these assays allow us to do is to do that sort of NGS without putting a patient through a biopsy. There have been some early studies comparing tissue-based NGS with plasma- and urine-based testing, but when the vast majority of patients will test positive for T790M, they will be positive on all 3 tests. If a patient has disparate results, it seems that what drives their response is tissue. If the tissue is positive, they’re going to respond, and if their plasma is positive but their tissue is negative, their response rate is lower.There are a couple of studies. Osimertinib has the AURA3 study comparing osimertinib to docetaxel after progression on a first-line TKI for T790M-positive tumors, and that has been released in a press release. I’m excited to actually see the data to confirm our practice; that is, if a patient has T790M-positive disease, we’re giving them osimertinib.

Where do you envision the field of targeted therapies in this space heading in the next 5 to 10 years?

The other study that is being evaluated is moving osimertinib to first-line with the thought that, because it has suppression of T790M, you might be able to have a prolonged response more than the 9 to 12 months we expect with erlotinib, gefitinib, and afatinib. I’m interested to see, in a larger study, what they showed. The first study data, which was presented by Dr Suresh Ramalingam this past year, showed that there was markedly prolonged PFS with osimertinib in the first-line setting.Looking to the way that ALK is playing out, for instance, there’s not only 1 mutation— there are multiple mutations. We need to identify which drug will fit which mutation and build on what they do in things such as infectious disease, where we can identify, “This is the mutation and this is the drug that’s best suited for that.” That is the way they treat HIV, and that’s sort of how we should be viewing lung cancer, and hopefully we will soon.