<p>TKIs have transformed the treatment landscape for patients with ROS1-positive non–small cell lung cancer (ROS1+ NSCLC) by providing them with clinically meaningful benefits compared with those provided by chemotherapy. However, many patients with metastatic NSCLC may acquire resistance alterations while on or after receiving TKI therapy. Our featured experts discuss approaches for overcoming acquired resistance mechanisms in ROS1+ metastatic NSCLC.</p>

Resistance can pop up in a couple of different ways when we are treating with a TKI. It can develop within the gene itself or externally. We do not yet know as much about ROS1 resistance as we do about EGFR or ALK, but we are rapidly gaining knowledge. We know that MET is a common resistance pathway, and new resistance mutations within the gene itself can be seen. With MET amplification, we can think about using crizotinib, especially if there are no brain metastases, or potentially adding a MET-targeted TKI, and, sometimes, I will switch back to chemotherapy (preferably pemetrexed based). I have found that chemotherapy treatments can be quite effective, especially if it has been a few years since the patient has had any. There is also the broader TKI, cabozantinib; I have seen efficacy with that as well in heavily pretreated patients.

<br>

Ideally, you need to know what else is going on with the patient. What other treatments have they had? What are the other co-mutations? If we do not fully understand the resistance pathway, I will treat the patient with something that is mechanistically different from what was used first line. We do not really know enough about some of the newer drugs and how they work after other ROS1 agents, but I do think that the more similar they are, the less likely they are to affect the new resistance pathway.

<br>

We are now starting patients who have tumors with ROS1 alterations on more potent next-generation ROS1 inhibitors. So, it does not necessarily make sense that you can guess the resistance mechanism, particularly when using a potent drug. You really do not know what the resistance mechanisms are unless you test.

<br>

Even in the wake of the rapid development of newer drugs, it is still important to follow the basic principles of identifying the resistance mechanism and determining what makes sense from there. I think that an advantage of newer drugs is often improved tolerability. However, just because it is new does not necessarily mean that it is the right drug for a given patient.

We can think about acquired resistance mutations in the context of EGFR, where patients initially treated with EGFR TKIs were frequently developing a T790M mutation. But now, with osimertinib, we do not see this mutation anymore because it is such an effective drug. The lesson from this is to cover those on-target resistance mutations that are most likely to emerge. However, with ROS1, we do not have that type of situation, as at least half of the ROS1+ population is developing a single resistance mutation as a result of first-line treatment. Many of these acquired mutations are quite rare, so there may not be much information on what to do.

<br>

If you are more effective at inhibiting the kinase with a TKI in patients with an ROS1 mutation, it is likely that there is less opportunity for a secondary acquired mutation to arise, which means that you might see more varied off-target alterations, such as MET amplification. We also see off-target alterations in patients with an EGFR mutation in whom we have used a combination of targeted therapies. The literature is replete with trials adding a MET inhibitor to osimertinib with impressive responses. Overall, I think that the same paradigms will be true in the ROS1 space. ROS1 is much less common, so we just do not have the depth of experience that we have with EGFR.

<br>

So, when a patient progresses on a TKI, the question then becomes: Do you add chemotherapy and continue the TKI? Most of the data on this also come from the EGFR space. For example, recent data from the COMPEL trial demonstrated that continuing osimertinib in patients with an EGFR mutation did have an impact on progression-free survival. Most of the TKIs seem to “play” well with chemotherapy. Given the available data across oncogenic driver experiences, my sense would be to probably continue the TKI if I were to transition to chemotherapy.

At disease progression, clinicians have 2 options. The first is to choose treatment based on what acquired mutations are present, which was learned in advance. The second option is to just move to the next-available ROS1 inhibitor. However, if you do the latter, you do not know if the resistance mechanism is on- or off-target, so therapy may be ineffective. You really need to try to find the mechanism of resistance.

<br>

If someone has central nervous system progression, lorlatinib has great brain penetrance and works on some of the on-target ROS1 mutations, except the most common one (ie, G2032R). Repotrectinib and taletrectinib work on G2032R and most of those other on-target resistance mutations, except for L2086F. Cabozantinib is currently the only drug available that works for L2086F. It is a type 2 TKI as opposed to a type 1, like most of the others.

<br>

As Dr Wakelee indicated, the full journey of the patient must be taken into consideration. This can get complicated when you have a patient who has gone through multiple lines of therapy. They might have an identified resistance mutation that does not respond to a treatment that should work. This may be because it is not only the identified mutation that is present; another driver is also at play that you have not yet identified. For example, there is a new investigational drug, zidesamtinib, that everyone is currently excited about. However, we know that some patients may not have success with this agent because they have a secondary driver (eg, a patient with MET amplification). In the future, it will be critical to define the population in which a drug not only will work but also absolutely will not work.