Langer Lends Insight on Transformed NSCLC Landscape

Corey J. Langer, MD

The discovery of actionable mutations and the increasing importance of immunotherapy has created a brighter outlook for patients with non—small cell lung cancer (NSCLC), but a better understanding of resistance mechanisms is necessary to continue advancement in the field, said Corey J. Langer, MD.

“In short order, really in the last 10 to 15 years, there are at least 4 actionable oncogenic drivers. EGFR was the first, then ALK, and more recently ROS1 and BRAF. Very few other cancers have that array of potential oncogenic drivers,” said Langer, director of Thoracic Oncology at the Abramson Cancer Center and a professor of medicine at Perelman School of Medicine and the University of Pennsylvania. “We are starting to see other drivers emerge like HER2, NTRK, and MET exon 14 skipping mutations, all of which are actionable.”

The emergence of effective targeted therapies transformed the treatment landscape for patients with oncogenic drivers, added Langer. For example, frontline osimertinib (Tagrisso) demonstrated unprecedented efficacy in patients with EGFR-mutant NSCLC, according to expansion cohort data from the phase I AURA trial (NCT01802632) that were presented at the 2019 European Lung Cancer Congress. Specifically, results showed an objective response rate (ORR) of 67% in patients who received osimertinib at 80 mg; ORR was 87% in those who were given an elevated dose of 160 mg. Median progression-free survival (PFS) in the overall population was 20.5 months, and approximately 25% of patients maintained a response at 36 months.

In an interview with OncLive, Langer highlighted how precision medicine has emerged in the treatment of patients with NSCLC via targeted therapies and checkpoint inhibitors.

OncLive: What are some of the actionable oncogenic drivers in NSCLC? How have targeted therapies impacted the treatment of these patients?

Langer: The targeted therapies have irrevocably transformed the therapeutic landscape, at least for patients with oncogenic drivers. In the past, patients with EGFR-mutated disease would have received chemotherapy alone or chemotherapy plus bevacizumab (Avastin). With that approach, response rates were about 30%; PFS, at best, was 6 months. Median overall survival (OS) was between 12 and 14 months. Now, with the advent of gefitinib (Iressa), erlotinib (Tarceva), and, most recently, osimertinib, response rates are 70%. PFS with osimertinib is nearly 19 months, and median OS in many of these studies has not been reached at 3 or 4 years; median OS in this population is clearly north of 4 years. Remember, we are seeing some long-term survivors. I don't fool myself for a moment that these patients are cured, but as long as they stay on therapy, they seem to be doing well—even at reduced doses.

We have seen even more astounding success [in patients with] with ALK translocations. This is a smaller population; unlike EGFR, which occurs in about 10% to 15% of patients with NSCLC, ALK occurs in about 3% to 8% of patients. The drugs work exceedingly well in this space. Alectinib (Alecensa) and brigatinib (Alunbrig) are now the frontline agents of choice with response rates north of 70%. PFS, at least in the ALEX study, is about 3 years. We have never seen that before. We were happy to see survivors at 3 years; now we are talking about a median PFS of 3 years. OS in [patients with] ALK [translocations] and probably ROS1[-positive disease] as well looks to be about 6, even 7 years.

ROS1-positive NSCLC is treatable with crizotinib (Xalkori), and we have seen some recent data with entrectinib. In one study, the median PFS was about 19 months. In patients with BRAF-mutated disease, the same combinations that work in melanoma also work in NSCLC. We are seeing response rates between 60% and 65% with these combinations, and PFS is 10 to 11 months. However, unlike those with EGFR, ALK, and ROS1 fusions, patients with BRAF mutations are also sensitive to immunotherapy. As such, that approach is a possible path for second- and third-line treatment.

Finally, RET translocations have been treated with older drugs, such as cabozantinib (Cabometyx) and vandetanib (Caprelsa), and we are seeing response rates of 30% to 40%. Newer drugs that don't have names yet and are not FDA approved are being tested in these patients. Growing up in this field and finishing my fellowship as a junior faculty, I could not have envisioned what we are seeing now. The landscape has been completely transformed, but oncogenic drivers are only seen in, at most, 33% of patients.

Could you speak to the importance of next-generation sequencing (NGS) in evaluating these oncogenic drivers?

NGS is critical, but the big problem with it is that the turnaround time is still about 2 to 3 weeks, sometimes 1 month. Therefore, if a patient is really ill, you have to get going with some sort of treatment. You might not be able to wait that long. Of course, our default option is chemotherapy plus immunotherapy. If they are not terribly ill and you think you can wait, you can probably palliate their symptoms with spot radiation to a bone lesion or brain lesion, for example.

Remember, with NGS, you can do a lot more with a lot less. NGS enables us to evaluate multiple oncogenic drivers with limited specimens. If it can be done, it should be done. If you have a high index of suspicion, say you have a never-smoker, adenocarcinoma, and East Asian ethnicity, you probably should do spot testing to get a quick result. There is fairly high probability this type of patient would have an EGFR or ALK mutation.

Switching gears to immunotherapy, what level of benefit have you been seeing with this approach in the real-world setting?

I would say the benefit with immunotherapy is not quite as striking as what we see with targeted therapy. Remember, it's not binary; targeted therapy either has a driver or it doesn't. If a patient has a driver mutation, they are highly likely to respond to treatment. Immunotherapy is more of a continuum. The higher the level of PD-L1 expression, the higher the likelihood of response, but even then, the response rates are, at best, 45% to 50%. We are not seeing these 70% response rates. Still, single-agent pembrolizumab (Keytruda) in 2 separate studies, KEYNOTE-024 and KEYNOTE-042, has shown a clear advantage to histology-appropriate chemotherapy in the PD-L1—high group.

The big question is, “Can we do even better with chemotherapy plus pembrolizumab versus pembrolizumab alone?” There are now trials ongoing that will address that question. To this extent at least, immunotherapy is a personalized approach and PD-L1 status is critically important. Remember, PD-L1 status can be established on a simple immunohistochemistry test; you don't need NGS.

Now that immunotherapy has moved into stage III NSCLC, how has that impacted the landscape?

Here, too, we are seeing a transformation in the therapeutic landscape. We spent about 20 years running clinical trials in locally advanced disease comparing chemoradiation with chemoradiation, plus other agents like cetuximab (Erbitux). We tried increasing the radiation dose or giving it upfront versus out back. These were all negative trials. PACIFIC sort of washed over us like a tsunami, and it was a strikingly positive trial. There was striking PFS benefit, 3-fold improvement from about 6 months to 17 months. There were still some skeptics, like me—remember, this was in patients who had not been staged like we would in the United States; it was a global trial. PET imaging is not routine in other parts of the world.

Essentially, in anyone who is a candidate who doesn't have pneumonitis and has stable disease after chemoradiation, durvalumab (Imfinzi) for up to 1 year is the new standard of care. Needless to say, there is now great interest in moving durvalumab or other PD-1/PD-L1 inhibitors upfront with concurrent chemoradiation. A ton of research is ongoing in this area.

What are some unanswered questions regarding resistance mechanisms?

Unfortunately, the vast majority of these patients will still develop disease progression. At least in the advanced setting, the vast majority of patients will die from their disease. Fifteen or 20 years ago, most patients died within 2 years. Now, we've at least pushed that out to 3 or 4, and even 5 or 6 years. We are seeing a small minority of patients, as with melanoma, who are out 3 or 4 years either on or off treatment who seem to be enjoying long-term PFS. They may not be cured—their scans show a fair amount of disease—but [their situation is not] getting worse.

Nevertheless, we have not even begun to crack the code for resistance in checkpoint inhibition; it is a major focus of research. This is much harder to understand because it isn't driven by a specific genetic abnormality. Really, we are talking about the immune system. Targeting 1 part of the immune system might be like “Whack a Mole;” you hit 1 target and then you miss another 1 that suddenly appears. For those who develop resistance on TKIs, there appears to be several potential mechanisms. In some cases, adding a new drug to the original TKI has salvaged therapy when disease was progressing. There is a much more binary approach in the targeted therapy space.

What does a diagnosis of lung cancer mean today versus a decade ago?

It is still a bad diagnosis, but we are definitely more hopeful. When I started out—I finished my fellowship in 1987—it was bleak. Median OS was 6 to 8 months for those with this disease. All of a sudden, we are seeing trials of single-agent, nonchemotherapy approaches, where OS is 20 to 30 months. Up to 3 or 4 years ago, you could color me as an immune skeptic. Interleukin-2 was a highly toxic therapy without much benefit. Seemingly overnight, the checkpoint inhibitors have altered the landscape. We need to encourage patients to go on clinical trials. My only regret is that I'll probably retire before I see the full thrust of all these advances.

Yang J C-H, Ramalingam SS, Lee CK, et al. Osimertinib as first-line (1L) treatment for epidermal growth factor receptor (EGFR) mutation-positive advanced non-small cell lung cancer (NSCLC): final efficacy and safety results from two phase I expansion cohorts. Ann Oncol. 2019;30(suppl 2, abstr 122p). doi: 10.1093/annonc/adz063.020.