Mutation Testing to Guide ALK TKI Selection in NSCLC

Transcript:Alice Shaw, MD, PhD: ALK and ROS1, while they’re both sensitive to crizotinib, they actually are distinct types of lung cancer, and their resistance mechanisms to crizotinib are also—while there are some similarities—distinct. And this does translate into different therapies for patients with ALK and ROS1 once they failed on crizotinib. For ALK-positive patients, the second-generation class of inhibitors—alectinib, ceritinib, and, soon, brigatinib—are all very active in ALK-positive patients who have failed crizotinib. The FDA approved alectinib and ceritinib for crizotinib-pretreated patients without regards to the mutation status of the patient, meaning they did not require a biopsy post crizotinib in order for patients to be eligible for alectinib or ceritinib. So, I would say that biopsies are not absolutely mandatory for an ALK-positive patient who fails on crizotinib because their chance of responding to a second-generation inhibitor is quite high, regardless of whether or not they have an ALK-resistant mutation or not.

At our institution, we do routinely biopsy ALK-positive patients when they fail on crizotinib because there is a small proportion of patients, probably less than 10%, who will have developed a very resistant ALK mutation, such as G12O2R. G12O2R does not respond to the second-generation inhibitors, like alectinib and ceritinib, and so for those patients, we would actually prefer to treat them with a third-generation inhibitor, like lorlatinib. So, there is a potential role for doing biopsies after crizotinib for those rare cases where a patient may have an unusual resistance mutation. However, overall, I would say it’s not absolutely mandatory since most patients will respond to a second-generation inhibitor after crizotinib.

ROS1 is very different than ALK in this regard. For ROS1, when patients relapse on crizotinib, about one-half of them have a very resistant mutation called ROS1 G2032R. This is analogous to the ALK mutation, G12O2R. They’re both very refractory mutations to many of the targeted therapies. It’s very important to point out that while crizotinib is a dual ALK and ROS1 inhibitor, not all of the second-generation ALK inhibitors work on ROS1. And in particular, alectinib—which we use all the time for ALK-positive patients who have relapsed on crizotinib and even for crizotinib-naïve patients—is not a ROS1 inhibitor, so it should definitely not be considered an option for a ROS1 patient who has failed on crizotinib. Even drugs like ceritinib and brigatinib, while they have some ROS1 activity, they cannot overcome the most common ROS1 G2032R mutation, and we’ve seen very little activity with the second-generation ALK inhibitors for ROS1. So, ROS1 is a bit unique in this way, and there are studies now exploring new next-generation ROS1 inhibitors for patients who have failed on crizotinib. These include drugs like lorlatinib, which is a dual ALK/ROS inhibitor. Old drugs, like cabozantinib, are being repurposed potentially for ROS1 and a few other new experimental therapies for ROS1. So, ROS1 has to be considered very much a different disease than ALK when patients relapse on crizotinib.

Liquid biopsies are a very exciting new tool that we’re beginning to use, not just for research, but also for clinical monitoring and testing of patients. Liquid biopsies, of course, are referring to analyzing the circulating tumor DNA that can be detected in the blood and looking for a specific mutation in the DNA. There’s a lot of literature supporting the use of liquid biopsies in identifying some resistance mutations, like the EGFR T790M gatekeeper mutation and also EGFR-activating mutations. However, in the ALK setting and the ROS1 setting, there are less data on how well these liquid biopsies work in detecting the mutations. We do know that the liquid biopsies are not that sensitive for detecting the actual rearrangements themselves. However, mutation detection is much easier than rearrangement detection. Going forward, I imagine that we’ll be using more liquid biopsies to track both ALK and ROS1 patients as they’re on treatment and to detect resistance mutations, like the ALK G12O2R or the ROS1 G2032R in patients, so that we can then direct patients to the correct therapies.

Tony Mok, MD: Although ceritinib and alectinib are very effective second-generation TKIs, there is still the time that it becomes resistant. And then, the 2 well-known gatekeeper mutations that are resistant to these 2 agents are namely ALK F1174C and also G1202R; it’s just the location. And also there have been extensive IC50 studies to show that the dosage of the drug actually has to be very high before we can inhibit this type of mutation. We have to find a new solution, and so it comes to the third-generation, namely neratinib, which had been studied both preclinically and also the occasional case report, to be able to inhibit the patient with a G1202R gatekeeper mutation.

Alice Shaw, MD, PhD: We recently published a study on a large series of patients who had undergone repeat biopsies after failing on crizotinib or a next-generation inhibitor like alectinib, ceritinib, or brigatinib. And it was very interesting because depending on the ALK inhibitor, different resistance mutations and resistance mechanisms would emerge. So, for example, with crizotinib, it’s only a minority of patients who develop on-target ALK-resistant mutations; it’s about 20% of patients. However, in contrast with alectinib, ceritinib, and brigatinib, we see that about one-half of patients—50% or more of patients—now have on-target ALK mutations. And this likely reflects the greater potency of these second-generation inhibitors in shutting down ALK.

So, for alectinib, in particular, what we’ve identified is that there are 2 very common mutations that emerge in patients. One is the G12O2R mutation, which is really refractory to second-generation inhibitors, but fortunately responsive to third-generation inhibitors like lorlatinib. Another very common mutation that we see emerging after alectinib treatment is I1171 mutations, and these have been reported by a number of groups looking at patients who have failed on alectinib. These mutations are actually sensitive to ceritinib. This would be an example where you may actually move a patient from one second-generation inhibitor, alectinib, to another second-generation inhibitor based on their mutation profile of their resistant biopsy.

There are a number of other mutations that we’ve been seeing that emerge post alectinib, post ceritinib, or post brigatinib, and our hope is that over time, as we gather more of this information, this will allow us to better tailor the selection of ALK inhibitors based on the underlying mutation.

Transcript Edited for Clarity