Corey J. Langer, MD, outlines the prevalence of BRAF mutations in NSCLC and looked ahead to where the development pipeline of agents for the treatment of patients with BRAF-mutated NSCLC is headed.
Corey J. Langer, MD
Novel combinations are in development for the treatment of patients with BRAF V600-muated non–small cell lung cancer (NSCLC), and although resistance remains a challenge and questions regarding how to best sequence targeted agents in the treatment paradigm, encorafenib (Braftovi) plus binimetinib (Mektovi) has shown promising efficacy in this patient population, according to Corey J. Langer, MD.
“The big problem with BRAF V600 mutations, and BRAF mutations in general, is the development of resistance; it’s generally inevitable,” Langer said. “There are multiple mechanisms including KRAS and NRAS, CRAF and ARAF overexpression, increased levels of BRAF homodimers, MEK mutations, and PTEN loss—all of these are at least potentially targetable. This has led to what I call the next generation of RAF inhibitors. Ideally, [these agents] not only inhibit class I, but class II and III. There’s this phenomenon of paradoxical activation where the drugs work, and then after a while they seem to stimulate tumors that harbor this mutation. If we can inhibit that, we’ll be ahead of the game.”
During a presentation at the 24th Annual International Lung Cancer Congress,1 Langer highlighted recently presented data from the phase 2 PHAROS trial (NCT03915951), which evaluated the TKI combination of encorafenib plus binimetinib in patients with BRAF V600–mutated NSCLC. In the treatment-naive cohort (n = 59) and the previously treated cohort (n = 39), the objective response rates (ORRs) were 75% (95% CI, 62%-85%) vs 46% (95% CI, 30%-63%), respectively. The median duration of response (DOR) was not estimable (NE; 95% CI, 23.1-NE) and 16.7 months (95% CI, 7.4-NE), respectively.
In an interview with OncLive, Langer, the director of thoracic oncology at the Abramson Cancer Center and a professor of medicine at the Hospital of the University of Pennsylvania in Philadelphia, outlined the prevalence of BRAF mutations in NSCLC and looked ahead to where the development pipeline of agents for the treatment of patients with BRAF-mutated NSCLC is headed.
Langer: BRAF mutations generally occur in exon 11 or 15. Approximately 50% of NSCLC BRAF mutations are actionable, [meaning] they occur in V600E, for which we have targeted therapies. Unfortunately, we have no approved drugs for the other mutations.
[BRAF mutations] may or may not be related to nonsmoking history, particularly V600E. The V600E mutation is more likely to occur in women and independently, without treatment, is associated with shorter disease-free and overall survival [OS]. There have been surveys in the United States where the incidence of V600E is approximately 1.5% to 2.0%. A much larger survey from Europe—France specifically—showed a very similar percentage.
[Investigators] looked at 236 patients with BRAF-mutant NSCLC, and roughly 45% fell into what we call class I, which includes the actionable group, 32% were class II, and 23% were class III. By and large, most of these individuals had a smoking history, but class 2 and 3 were associated almost exclusively with smoking history. Ninety-seven percent of those in class 2 and 94% of those in class III had at least some smoking history compared with 78% in class I. Class II and III were more likely to have KRAS co-mutations, brain metastases, poorer clinical outcomes, and shorter progression-free survival [PFS] with conventional platinum-based treatments.
When we look at how these mutations activate and signal, class I is RAS independent. It’s associated with BRAF monomers; class II is similar, except BRAF dimers are the feature; and class III is dependent on RAS and dimer dependent, and unlike class I, has impaired or virtually no BRAF kinase activity. That in turn may be the reason it’s relatively resistant to treatment.
Patients [enrolled received] 450 mg daily of encorafenib and 45 mg twice daily of binimetinib. Prior brain metastases were allowed as long as patients were asymptomatic. They had to have good performance [ECOG] performance status of 0 or 1, no other actionable mutations or alterations, and no more than 1 prior line of treatment for metastatic or recurrent disease.
The disease control rate at 24 weeks was 64% in the treatment-naïve group—this includes patients with stable disease—and it was 41% for the previously treated group. The waterfall plots for these cohorts do show a number of complete remissions in both groups, which is quite gratifying. The median PFS for this combination has not yet been reached [in the treatment-naïve cohort]. The lower end of the confidence interval was 15.7 months, so it looks like it may be better than the approved dabrafenib [Taflinar] and trametinib [Mekinist] combination. In the previously treated cohort, it’s very similar at 9.3 months.
Encorafenib plus binimetinib has not yet been FDA approved, but I suspect its approval may be imminent.
The incidence of grade 3 and 4 toxicity [with] encorafenib and binimetinib was 41% [in PHAROS]. For the traditional dabrafenib/trametinib combination the incidence of grade 3 or higher toxicity is approximately 75%. The problem is we have long term follow up now for dabrafenib/trametinib, 4 to 5 years, but approximately a year and a half to 2 years for encorafenib and binimetinib [which is] still quite short.
When I treat patients with these agents, for the most part, they do respond. [However,] I’ve asked this question, and I haven't gotten a full answer yet, but the responses that we're seeing include patients with dose reductions. My patients, with 1 exception, are all dose reduced, and 3 out of 4 have sustained phenomenal responses. Intriguingly, they all presented with malignant pleural effusions.
At least 2 of my patients are in complete response. I had 1 patient, who after a 4.5-year response, did have disease progression, and she's now getting the [phase 3] KEYNOTE-189 trial [NCT02578680] combination [of pembrolizumab plus chemotherapy]. [Prior to this], I had to stop the BRAF drugs because she was having a fair amount of toxicity, and I did not think she could tolerate both chemotherapy and the TKIs together. However, if she does have disease progression at some point on chemotherapy, I may re-invoke those agents or maybe mix and match at some point.
What should be the focus as the next generation of agents for BRAF-mutant NSCLC are developed? Are there any agents in development you’re particularly excited about?
The next generation ideally should probably be active in KRAS-mutant tumors since this is a mechanism of resistance. [They should also] be able to overcome resistance mechanisms to the current BRAF-MEK inhibitor combination. In addition, we need better documentation and realization of central nervous system (CNS) activity. There are several strategies [that are currently being examined]; BRAF dimer inhibitors including lifirafenib [BGB-283)] and LXH254, [and] RAF dimer breakers such as PLX8394, [are of interest]. These agents, particularly in class II BRAF-mutant tumors seem to be much more active in preclinical models than our conventional drugs, including dabrafenib and encorafenib.
Lifirafenib is an example of a monomer dimer inhibitor. It’s reversible and inhibits not just mutant BRAF but also wild-type BRAF, ARAF, CRAF, and EGFR. In early phase 1 studies that included multiple different tumors driven by BRAF and KRAS, it clearly showed activity; response rates in lung cancer looked to be close to 50%. Toxicities were tolerable with hypertension and fatigue being the most common grade 3 AEs.
Another example is belvarafenib (HM95573), which is a dimer-selective inhibitor. It seems to circumvent resistance that can develop with dabrafenib and trametinib. BDTX4933 is a dimer inhibitor with CNS activity and particular activity in those class II and III resistant mutations, certainly in preclinical models with much lower IC50s compared with encorafinib and belvarafenib.
The big question is whether patients should receive targeted therapy up-front or immunotherapy as first-line treatment. The argument for giving dabrafenib and trametinib up-front is you get much better response rates and PFS compared with either chemotherapy or chemoimmunotherapy. It’s orally administered, so patients don’t have to come to the infusion center every 3 weeks. The dosing is malleable, [and] the majority of patients do require dose reductions.
[Additionally], we have the potential benefit of rechallenging patients if they do go on to chemoimmunotherapy and it doesn’t work—we’ve seen instances where patients can respond again. The cons [are] a high percentage of patients with BRAF mutations have a smoking history, [and] they may be particularly sensitive to checkpoint inhibitors. We are now seeing long-term survival, which is not terribly likely with TKIs. In addition, at least for some patients, the toxicity profile of checkpoint inhibitors, even though [they are] given intravenously, may be superior; the drug combinations are fairly toxic.
This is a [disease] class that we recognized 10 or 12 years ago, it’s only in the last 6 or 7 years, we’ve had drugs for BRAF [mutations], specifically for V600E. Dabrafenib and trametinib remain the standard of care both in the first and second line. Encorafenib plus binimetinib have yielded similar response and PFS data to date, although [results] may be better [in patients who are] treatment naïve. There may also be potential improvements in the toxicity profile with far less fever and far fewer treatment discontinuations.
Whether to proceed first with checkpoint inhibitors or TKIs is a point of controversy. We need to see improvements in tolerability, DOR, and CNS penetration. The next generation of RAF inhibitors need to show absence of paradoxical activation, they need to inhibit class II and III BRAF mutations, they ideally should show activity in KRAS-mutant NSCLC, and they should also have CNS activity.