Considerations for TKI Selection and Sequencing in ALK-Positive NSCLC

Rajat Thawani, MBBS, Assistant Professor of Thoracic and Phase 1 Oncology at the Knight Cancer Institute at Oregon Health & Sciences University

Edward Kim, MD, MBA, Physician-in-chief at City of Hope National Medical Center, professor in the Department of Medical Oncology & Therapeutics Research Construction Industries Alliance City of Hope Orange Count Physician-in-Chief Chair

Edward B. Garon, MD, MS, Professor in the Department of Medicine, Hematologic Oncology, and director of the Thoracic Oncology Program, and co-director of Signal Transduction and Therapeutics at the Jonsson Comprehensive Cancer Center at the University of California Los Angeles Health

As part of a Meeting of the Minds program, CancerNetwork® hosted a panel discussion focused on considerations for tyrosine kinase inhibitor (TKI) selection and sequencing in the first and second lines of treatment for patients with stage IIIB to IV ALK-positive non–small cell lung cancer (NSCLC). Highlighting data from the phase 3 CROWN trial (NCT03052608) presented at the 2024 American Society of Clinical Oncology Annual Meeting (ASCO), which assessed lorlatinib (Lorbrena) as first-line therapy in this patient population, the discussion initially covered the trial results, highlighting strong progression-free survival (PFS) outcomes, strong central nervous system (CNS) responses, and a manageable adverse effect (AE) profile. Additionally, the panel discussed sequencing options following progression after first-line lorlatinib, considerations for lorlatinib as a second-line therapy, and the feasibility of combining inhibitors targeting different biomarkers.

The panel was led by Rajat Thawani, MBBS, assistant professor of Thoracic and Phase 1 Oncology at the Knight Cancer Institute at Oregon Health & Sciences University. Panelists included Edward Kim, MD, MBA, physician-in-chief at City of Hope National Medical Center, professor in the Department of Medical Oncology & Therapeutics Research Construction Industries Alliance City of Hope Orange Count Physician-in-Chief Chair; and Edward B. Garon, MD, MS, professor in the Department of Medicine, Hematologic Oncology, and director of the Thoracic Oncology Program, and co-director of Signal Transduction and Therapeutics at the Jonsson Comprehensive Cancer Center at the University of California Los Angeles Health.

Phase 3 CROWN Trial Clinical Overview

Thawani: Are you able to provide an overview of clinical evidence supporting lorlatinib as the first-line therapy for these patients, specifically the phase 3 CROWN trial?

Kim: The CROWN trial [assessed] lorlatinib vs crizotinib [Xalkori]. These were patients who [were] ALK-positive, and it was a head-to-head study. Crizotinib was our initial ALK-approved drug. There were almost 300 patients in this study, and [it] showed a similar theme to where we have been going today [with] profound differences between the 2 drugs.

As far as efficacy, PFS was off the charts. I do not know if I have seen the final PFS number for the lorlatinib arm, whereas crizotinib was 9.1 months or so and the curves are [quite] disparate. There was a clear winner with the newer generation ALK inhibitor. With many of these newer-generation drugs, intracranial penetration works well, and it delays any onset of brain metastases. This is especially important in the ALK population, where [the brain] tends to be [the] first site of recurrence, or a major site of recurrence.

Part of that was crizotinib was not, and never was, developed to be a pure ALK inhibitor. It was almost an accident, how we got crizotinib. It targets MET and ROS1 more, but it also targets ALK––these types of molecules can have off-target effects. ALK was one of the major off-target effects, as the way I would put it. There is no doubt that lorlatinib is a drug that is approved in the first line [setting] and is one of our major drugs we use. There are some [adverse] effects associated with it.

Where there is a gap right now is that there was no direct head-to-head with alectinib [Alcensa]. If you do not like to do cross-trial comparisons, the numbers were better than alectinib when you compare them as they both beat up on crizotinib, but that is the lingering question out there.

Garon: I agree with that entirely. The lorlatinib data looked good. It beats crizotinib, which is a significant Achilles heel for managing these patients, [as] it has nearly no CNS activity. When you look at it, it looks like lorlatinib beat up on it a little bit more than alectinib. That being said, what is unique, and has led me to rethink this question about lorlatinib and alectinib is the 5-year data, in which I was at a meeting where people were discussing this. Prior to that, most [clinicians] were favoring alectinib. The fact that most patients remain progression-free 5 years out is meaningful. The low drop-off between years 4 and 5 also is meaningful, leading one to wonder whether or not you are seeing almost a towing of the curve there.

This is the debate––brigatinib [Alunbrig] is also approved in this space. I have not used it much in my own practice. I would say that when people have this debate, it is more frequently alectinib vs lorlatinib. It boils down to the idea that the tolerability, in general, has been a bit better with alectinib, but at least when one starts looking at the 5-year data––the PFS––there is a suggestion that you may do better with lorlatinib. That is [what] the field is grappling with when making these decisions.

Considering the Adverse Effect Profile of Lorlatinib in ALK-Positive NSCLC

Thawani: What are the usual challenges of incorporating lorlatinib in patient care?

Kim: When we first got crizotinib, there were a few neurocognitive [effects] that we would see, but not major. We did not see so much in ceritinib [Zykadia], but we tried to avoid using ceritinib at all costs. It was tough to give and toxic. We see these hyperlipidemias that occur, and I would not say that there was anything major that we saw with lorlatinib. We see some neurotoxicity that exists as well, [and] we see cognitive changes.

I have talked to colleagues who are sometimes profoundly describing these cognitive changes, [including] mood lability, to actual neurological aspects. It is something that is talked about and seems to be managed adequately with dose reductions, etc. [Many] of these CNS AEs do need to be [communicated] well to the patient up front. It seems like the mood swings can be [variable], and patients can [appear] up at first, but then can get down as well. A patient understanding this on top of the fact that they are stressed out with trying to fight lung cancer is important to know that the drug may be causing some of this. Mostly with drug reductions [and] dose holds [can you mitigate these].

There was some data that showed that even if you had to dose reduce lorlatinib up front, the efficacy was still there. Some have questioned whether that same level of dose reduction has the same intracranial efficacy, and that is a question that exists, but those are the major ones that we like to discuss.

Thawani: Is there any patient for who you would prefer using alectinib or brigatinib over lorlatinib considering these [adverse] effects or complications in patients?

Garon: The place where I tended to use lorlatinib is more in younger patients. I am impressed by that 5-year data. One of the things that’s been comforting is having patients who have a partner who is someone who is going to be able to notice changes that are inappropriate. I have never been in that problem where you have somebody who is having severe psychiatric issues, and there is nobody there to follow them or to check on them. That would be of some concern.

One other thing, and this may be a little unique to my practice, which is on the west side of Los Angeles has a bit of a unique patient population. I worry a bit if somebody weighs themselves every day. The weight gain issues are odd and can be significant. It is not that it is a problem every day, but it is giving you an idea of the extent to which somebody takes their weight into account in their life. I am not saying I would not use it in that group of patients, but if there is a group that you start piling on the potential for psychological difficulties, and the weight issues, I would want to have a particularly extensive discussion with [them].

Thawani: That is a good point. Broadly the metabolic and neuropsychiatric [adverse] effects, both of them [can be] challenging in lorlatinib, but it seems like with holds and temporary discontinuations, you can get it under control.

Sequencing Therapy Following TKI Progression

Thawani: As patients [experience progression] on lorlatinib, what do you do next? Not that it seems like [many] patients progress on frontline [therapy], but if they did, how do you select the next therapy?

Kim: The PFS curve is impressive. Luckily, it is not something that has required [much] urgency. [Considering] the field we are in, [therapy selection] is similar. We want to repeat next-generation sequencing [NGS] testing, preferably through tissue. Blood does not capture everything that we need. Right now, we do not have [many] TKI options. Post-TKI, we have seen some of this in the EGFR space that we have talked about, we have not seen the same thing in the ALK space. Now, there are TKIs being studied in clinical trials that can, in fact, help with these resistance mutation aspects that people are trying to assemble right now, but that is certainly not widely available. Many of them are in early phase studies, so the consideration goes back to chemotherapy.

We still have a relative indication to avoid immunotherapy in this population. It would be any number of choices of chemotherapy [for these patients]. If and when the chemotherapy stops working, [we reconduct NGS testing] to see if something else pops up. That is the strategy [we employ] right now.

Thawani: How do you decide on patients on other TKIs up front? Do you usually second line blindly give lorlatinib? Are you testing them before? What is your strategy?

Garon: We have tried to test it in this setting. The way that I tend to look at it is that my plan is to give lorlatinib unless something weird comes up, which happens. In fact, I got a little shaken in clinic a few weeks ago where, where we got from liquid biopsy, a clear resistance mutation in another gene where it was clear that it was not going to be helpful to give lorlatinib. When you have a bypass pathway in this patient, it was via the RAF gene, it is becoming clear that that is not going to be an effective strategy. I would say in the majority of the patients, it has been going to lorlatinib, but I get testing [done] to make sure there is not a red flag that says that hitting ALK a bit harder in this is not going to get you where you want to go.

Discussing the Feasibility of Combining Inhibition for Multiple Biomarkers

Thawani: Would you combine, for example, for a RAF mutation, a MEK inhibitor with alectinib or lorlatinib? How do you approach that?

Garon: This one is a bit easier, because it was a RAF fusion, and the approved therapies are directed against BRAF V600E. This is a situation where we did not have an approved agent, although what I would say is there are case reports, for instance, of treating with RAF-directed therapy and out-directed therapy in that case, but I would caution people about this. If you are doing this in your own clinic, at least make sure the patient is informed you do not have a significant database of efficacy [or] tolerability data.

What you will see if you do some of these things off trial, which many of us have had to do, is that you find ... sometimes in drug development, where you combine 2 things that you think will go fine together, and they do not. The big difference is that, in the context of a clinical trial, you have given somebody an informed consent form––they have evaluated the risks and benefits. In this case, you are saying, “This is a great idea,” but they have not gotten the results. We have to be cautious about what we describe to people. The questions [should include]: “Have I done it?” Yes, we have done things like that; “Have the results sometimes been good?” Yes; “Have the results sometimes been bad?” Yes.

Reference

Solomon BJ, Liu G, Felip E, et al. Lorlatinib versus crizotinib in patients with advanced ALK-Positive non–small cell lung cancer: 5-year outcomes from the phase III CROWN study. J Clin Oncol. 2024; 42(29):3400-3409. doi:10.1200/JCO.24.00581