Stephen Liu, MD, discusses ongoing developments in the treatment of patients with ALK-positive NSCLC.
Stephen Liu, MD
The first-line treatment landscape of ALK-positive non—small cell lung cancer (NSCLC) has been flooded with multiple targeted agents with more on the way, according to Stephen V. Liu, MD.
Alectinib (Alecensa) has become the current standard of care for most of these patients, but positive results with frontline brigatinib (Alunbrig) has the potential to shake up the landscape once more.
“The first-line space continues to rapidly evolve. We are talking about being on your first treatment for 3 years with metastatic NSCLC,” said Liu. “It really is a new era, and we have become accustomed to excellent outcomes with these drugs.”
In an interview during the 2018 OncLive® State of the Science Summit™ on Advanced Non—Small Cell Lung Cancer, Liu, an associate professor of Medicine at Medstar Georgetown University Medical Center, recapped ongoing developments in the treatment of patients with ALK-positive NSCLC.Liu: If we think of our frontline treatment for patients with ALK-positive cancer, it is almost an embarrassment of riches. It has evolved a lot in a pretty short period of time. Going back to PROFILE 1014, we saw that crizotinib (Xalkori) was superior to chemotherapy, but now after more than 4 years of follow-up, we can see that those patients have a significant improvement in survival if they receive the right treatment. What is the most interesting is that patients who start with crizotinib instead of chemotherapy and continue with ALK-targeted therapy afterward have the best possible outcomes. This suggests that targeted therapy is the best first treatment, and also the best subsequent treatment. For most patients, we want to keep them on targeted therapy for as long as we can.
The first-line space has gone onto evolve. We saw ceritinib (Zykadia) in the ASCEND-4 study was superior to chemotherapy, but the biggest change was with the ALEX trial when alectinib was head-to-head compared with crizotinib. Alectinib was far superior, and chemotherapy is no longer an appropriate standard of care. We saw an updated progression free survival (PFS) of almost 3 years compared with 11 months with crizotinib at the 2018 ASCO Annual Meeting.
At the 2018 ESMO Congress, we saw that the ALESIA trial, which was an Asian study, but had the same design of ALEX. We saw that again, [alectinib] was far superior. In the investigator-initiated PFS comparing alectinib with crizotinib, the hazard ratio was 0.22, which is “ridiculous.” This clearly cemented it as our preferred first-line option. Alectinib is an excellent, well-tolerated, highly effective drug.
This space will continue to evolve with brigatinib. Brigatinib had a very compelling PFS of longer than 16 months in the second-line space—much greater than even alectinib in that space—allowing for cross-trial comparison. What will the activity of brigatinib be in the frontline setting? We saw D. Ross Camidge, MD, PhD, of the University of Colorado, present some relatively immature data on the ALTA-1L trial after an 11-month follow-up. [ALTA-1L] trial compared brigatinib with crizotinib, which is another very effective, well-tolerated, and highly central nervous system (CNS)-penetrant drug. We saw a very impressive investigator-assessed PFS, but the median PFS was not reached.The role of crizotinib is very small. For specific mutations that it can restore sensitivity in, then crizotinib may play a role. It is going to be few and far between. Really, I view crizotinib as sub-standard care. Could you sequence therapy, starting with crizotinib and save other drugs for later? I really think that is an inferior strategy. In part because the act of progression itself is such a violent one at times, and if you spare patients that progression—especially if it is in the brain—it’s a much better outcome.
Even beyond that, the fewer drugs a patient is on, the less heterogeneity you are introducing. We know the resistance profile for patients who start with crizotinib is much different than those who start with alectinib. The more lines that you introduce over a shorter period of time, the more complex that biology gets, and the more difficult it will be to overcome things. We want to start with our best drugs upfront, not just for practical reasons but for biological reasons. Right now, we have 3 FDA-approved drugs in the first-line setting of ALK-positive NSCLC: crizotinib, ceritinib, and alectinib. We also have brigatinib, which is a category 1 National Comprehensive Cancer Network guideline-recommended drug that is not yet FDA approved. For the approved drugs, the advice is to start with alectinib. This is a highly CNS-penetrant drug, is probably the best tolerated out of all those drugs, and is clearly the most effective.
Assuming that brigatinib will receive an FDA approval, choosing between brigatinib and alectinib will be very complex. There, we might want to consider practice issues such as insurance coverage and cost. They have slightly different toxicity profiles, and we may want to tailor that to individual patients. The better decision will take a little more time. As the data mature, we will be able to compare the long-term outcomes for these 2 trials. The difference in resistance to brigatinib and alectinib, and how we can overcome that, is important. Brigatinib and alectinib are both great options, and we don't have sufficient information to decide which is the appropriate one upfront. We know a little bit about alectinib resistance. Most of what we know is based on pretty small studies. In a small study out of Japan, we saw that patients who started on alectinib and then move onto ceritinib has a response rate of 25%, although it did include 1 complete response. A small retrospective group looked at brigatinib following alectinib and saw a response rate of about 17%. There is a prospective study being led by Thomas Stinchcombe, MD, of Duke Cancer Institute, that is ongoing now.
We have some understanding in vitro about what resistance mutations these drugs can overcome. Translating that into the patient is something that we need to do prospectively. The short answer is that we don't really know yet, but we can envision a future where we will use those resistance mutations in a very similar way that we do with EGFR and plan our next lines of therapy. We just aren't quite there yet.
Lorlatinib (Lorbrena) showed a lot of activity in pretreated patients of 2 or more TKIs. We saw a response rate of 42%—fairly durable responses. Empiric use with agents such as lorlatinib with pretty broad activity will probably be the norm until we get a handle on resistance.What we keep coming back to with ALK-positive cancer, and really with all targeted therapy, is how can we fold in immunotherapy? Immunotherapy receives all the headlines and gets all the attention. Rightfully so—these are the drugs that can provide very meaningful benefit that can be measured in years. In some patients, it may never stop working. How can we offer that to our targeted therapy patients? Targeted therapy is very effective and much more reliable, but it has a transient benefit—resistance is something that we expect.
We have been trying to marry these for a long time, but it is really a forced marriage. We have been disappointed at some of the results. While cohort G of KEYNOTE-021, KEYNOTE-189, and KEYNOTE-024 introduced pembrolizumab to frontline therapy in NSCLC, ALK-positive patients were excluded. While they were included in IMpower150, those are patients who already received ALK TKIs.
We know that combining TKIs with immunotherapy is something that should be avoided outside the context of a clinical trial. If you look at the experience with crizotinib and nivolumab (Opdivo) in only 13 patients, almost 40% had severe hepatic toxicity, and 2 deaths were possibly related. When we have 2 very well-tolerated drugs and we combine them, sometimes the results are surprising and humbling. Outside of a trial, I cannot endorse using TKIs and immunotherapy at the same time. There is certainly some biologic rationale for using TKIs to affect the microenvironment, but we need to find a safe and effective way to do that. We are working toward that, but we are pretty far.