Alexander Drilon, MD, discussed updated findings with larotrectinib in <em>NTRK</em> fusion–positive tumors and entrectinib in <em>ROS1</em> fusion–positive NSCLC.<br />
Alexander Drilon, MD
In subset findings from the LOXO-TRK-14001 (NCT02122913) and NAVIGATE (NCT02576431) studies presented during the 2019 European Lung Cancer Congress (ELCC), larotrectinib (Vitrakvi) demonstrated an overall response rate (ORR) of 71% in patients with nonsmall cell lung cancer (NSCLC) harboringNTRKgene fusions.1
The TRK inhibitor was approved by the FDA in November 2018 based on data from these 2 trials. The agent is indicated for the treatment of adult and pediatric patients with solid tumors that have anNTRKgene fusion without a known acquired resistance mutation, are metastatic or where surgical resection is likely to result in severe morbidity, and have no satisfactory alternative therapies or that have progressed following treatment.
Results from the studies showed, of 7 patients with metastatic lung adenocarcinoma who were evaluable for larotrectinib response, there was 1 complete response (CR), 4 partial responses (PRs), and 2 patients with stable disease. Additionally, the median time to response was 1.8 months, and the median duration of response (DOR) was not reached.
Additionally, an integrated analysis of the ALKA-372-001, STARTRK-1 and STARTRK-2 trials evaluated the activity of entrectinib in patients with locally advanced or metastaticROS1fusionpositive NSCLC, which was also presented at the 2019 ELCC. Results showed that the multikinase inhibitor is highly active and well tolerated in this patient population, including those who have CNS metastases.2
Based on these findings, the FDA granted a priority review designation to a new drug application for entrectinib as a treatment for select adult and pediatric patients withNTRKfusionpositive locally advanced or metastatic solid tumors, as well as those with metastaticROS1-positive NSCLC in February 2019.
“The takeaway message is that entrectinib is an active drug forROS1fusionpositive lung cancers. The hope is that at some point in the future, maybe we’ll get the drug approved for this subset of patients. The question then becomes, “When do I use entrectinib versus crizotinib (Xalkori)?,” said Alexander Drilon, MD, clinical director, Early Drug Development Service, associate professor of thoracic oncology service, Memorial Sloan Kettering Cancer Center.
In an interview withTargeted Oncologyduring the 2019 ELCC, Drilon discussed these findings with larotrectinib inNTRKfusionpositive tumors and entrectinib inROS1fusionpositive NSCLC.
TARGETED ONCOLOGY:Could you give some background to TRK inhibition and the findings presented at this year’s meeting?
Drilon:First off,TRKfusions are a bona fide driver of both lung cancers and many other different cancers. In fact, larotrectinib was approved late last year, in November 2018, for anyTRKfusionpositive cancer, regardless of what the cancer looks like under the microscope. That was really a landmark approval, because it was the first targeted therapeutic that was approved for a genomic signature and really agnostic of tumor type.
At the 2019 ELCC, we specifically pulled out the outcomes of NSCLC patients who harborTRKfusion in their cancers, and there were 11 patients who were part of the larotrectinib series, and we found that there was a high response rate to therapy. It was a 71% ORR, considering that the denominator isn’t large. We saw a disease reduction in all patients who had therapy, there were deep responses to treatment, and 1 had a CR to larotrectinib.
In terms of durability, the median DOR was actually not reached yet. Most patients who received larotrectinib for theirTRKfusionpositive lung cancers remain on therapy. The patient who had been on [treatment] the longest is already pushing 2.5 years.
The last thing we featured was the intracranial activity of larotrectinibrecognizing that there were 3 out of 11 patients who had brain metastases on this trial. This has been a question that’s come up for theTRKfield in general, on whether or not these drugs have substantial activity in the CNS. What we saw was that patients had global overall disease control, meaning that patients had disease regression both extracranially and intracranially.
Thus far, we weren’t able to calculate intracranial response rate, but we featured a case of an older woman in her mid-70s who was found to have aTRKfusion in her stage IV lung cancer. She, unfortunately, had multiple asymptomatic brain metastases, and she refused standard-of-care platinum-doublet therapy after hearing the phenomenal results that we’ve achieved with larotrectinib across different cancers. She had a confirmed PR to therapy with a near complete intracranial response. By volumetric analysis, this was almost a 95% reduction in her tumor burden in the CNS.
The punchline here is that larotrectinib is not only an active drug if you have a lung cancer that harbors aTKRfusion, but also if you have brain metastasesthis can be a very active therapy. The corollary to that, hopefully, is that it also helps prevent the acquisition of CNS disease.
TARGETED ONCOLOGY:How do these findings affect screening decisions?
Drilon:The take-home message is that you should screen forTRKfusions. Coming from a lung cancer paradigm where we’re screening for many other drivers, you just need to be cognizant of whether you’re using an assay that meaningfully looks for theseTRKfusions.
Diagnostically, this can be a little bit more challenging than detecting other fusions, such asALK, because we found that the best way to look for aTRKfusion is to do next-generation sequencing (NGS), preferably with a hybrid capture-based platform. Also, we found that even when you use a very good DNA-based NGS test, you can still miss some of theseTRKfusions. That’s due to the length of the introns, for example; it’s hard to find these fusions.
The punchline here is that, if you’re looking at an assay, it might be best to look for an assay that has RNA on top of DNA in order to maximize the likelihood of picking up these fusions. In our experience with using an RNA-based assay, such as anchored multiplex PCRfollowing a good DNA-based NGA assay—is that you miss a driver 15% of the time, even when you do a good DNA test, that’s subsequently picked up by an RNA assay.
TARGETED ONCOLOGY:What are the most pressing research questions that remain to be answered about larotrectinib in this setting?
Drilon:Now that the drug is FDA approved, we would obviously like to see how the “chips fall” in terms of long-term tolerability. Thankfully, we’ve seen that with the safety data presented so far; there is a very good tolerability profile. The rate of dose reductions was very lowit was 9%—of the more than 120 patients who received larotrectinib, only 1 discontinued for a drug-related adverse event. However, it would be nice to see how things march out when patients are on this therapy for a very long time—knowing that they could be on this for years given the durability of disease control.
The second major challenge would be helping to characterize the profile of resistance to larotrectinib. We know that some of these cancers can acquire on-target resistance, meaning the cancers are still reliant on theTRKpathway.
Usually, when you see a report, when you re-sequence these cancers or do a plasma test, you find a kinase-domain. Thankfully, we now have next-generation TRK inhibitors, such as repotrectinib, formerly known as TPX-0005, and LOXO-195. These are drugs that that have been designed to address these resistance mutations. There are ongoing phase I trials for both agents. Therefore, if you have a patient who progresses on a first-generation TRK inhibitor like larotrectinib, [these options] would be available for patients. There have been data presented that we can reestablish disease control with a second pill after an earlier-generation pill fails.
TARGETED ONCOLOGY:You are also an author on an integrated analysis of ALKA-372-001, STARTRK-1, and STARTRK-2, which looked at entrectinib in locally advanced or metastaticROS1fusionpositive NSCLC. What did your analysis find?
Drilon:We know that crizotinib (Xalkori) is already approved by 1 or more regulatory agencies for the treatment ofROS1fusionpositive lung cancers. The question then becomes, “Are there agents that are able to improve outcomes for patients in this setting?” Entrectinib is a very potent ROS1 inhibitor that was designed to cross the blood-brain barrier. The question is, “Are we able to use a CNS-active agent in the first-line setting and hopefully improve outcomes for patients?” We know that crizotinib, while it can get into the CNS and we can see CNS responses, there can be a high failure rate in the CNS; about half of patient withROS1-rearranged lung cancersimilar toALKwill have some type of failure intracranially when they progress on crizotinib.
The entrectinib trial took ROS1 TKI-naïve patients. This is mainly because we know that, with our early patients on entrectinib who received prior ROS1-targeted therapies, we did not see any responses in that group. Therefore, the first important point is that entrectinib is not only a CNS-active agent, but it is really considered more of an early-generation ROS1 inhibitor.
Looking now at the treatment-naïve,ROS1fusionpositive lung cancers, this was a data set that had north of 50 patients, all of whom obviously had aROS1fusion that was detected by a local platform. Some did fluorescence in situ hybridization (FISH), some did NGS, but what we saw in the topline results in terms of activity was that the ORR was high at 77%this is something that you like to see with a targeted therapy that’s highly active, with a very nice waterfall. The majority of patients had disease regression with entrectinib, and certainly there were deep and durable responses to therapy.
The data set is notable, in that entrectinib actually achieved the longest median duration of response of any ROS1 TKI at approximately 25 months. That, to me, is a meaningful endpoint. If you have almost 80% of patients responding to therapy, it’s comforting to know that these responses can go on for a very long period of time. In terms of other endpoints, the median progression-free survival for the data set was around 19 months.
TARGETED ONCOLOGY:What should community oncologists take away from these findings?
Drilon:Certainly, if you had a patient with brain metastases, given the substantial number of patients with brain metastases on the entrectinib series, I would feel very comfortable giving the drug to those patients. We’re not only seeing an ORR of approaching 80%, but there’s also an intracranial response rate that was north of 50% in the entrectinib data set.
Corollary to that, if you assume that if the drug has meaningful penetration of the blood-brain barrier, then even in patients who do not have CNS metastases, you hope that with the use of entrectinib, you might prevent the emergence of brain metastases.