According to Tony S. Mok, MD, first-line treatment with tyrosine kinase inhibitors for patients with non–small cell lung cancers harboring uncommon driver mutations can be controversial. Instead, there is more evidence supporting the use of TKIs in the second-line for mutations such as <em>ROS1 </em>and <em>BRAF</em>, he explained during a presentation at the <em>19th Annual </em>International Lung Cancer Congress.<br />
Tony S. Mok, MD
According to Tony S. Mok, MD, first-line treatment with tyrosine kinase inhibitors (TKIs) for patients with nonsmall cell lung cancers (NSCLCs) harboring uncommon driver mutations can be controversial. Instead, there is more evidence supporting the use of TKIs in the second-line for mutations such asROS1andBRAF, he explained during a presentation at the19th AnnualInternational Lung Cancer Congress.
Many options currently exist for treating patients withEGFR-mutant andALK-positive advanced NSCLC with a targeted first-line therapy; however, not as much data currently exist for treating patients harboringROS1,BRAF,or other less common mutations, with targeted therapies upfront, Mok, the Li Shu Fan Medical Foundation Professor of Clinical Oncology at the Chinese University of Hong Kong, said.
Frontline EGFR-Targeted Therapies
Many first-line treatment options currently exist for treating patients withEGFR-mutant NSCLC with single-agent targeted therapy, and combinations with EGFR-targeted therapies are also making their way into clinical practice internationally, Mok said. “How then are you going to choose [which treatment regimen to give], I really don’t know,” he added.
He raised the question as to whether next-generation TKIs could be better for frontline treatment than first-generation TKIs.
First-generation EGFR TKIs gefitinib (Iressa) and erlotinib (Tarceva) have shown similar efficacy, but second-generation TKIs, including afatinib (Gilotrif) and dacomitinib, and the third-generation osimertinib (Tagrisso) have demonstrated improved survival rates over first-line agents in head-to-head trials.
The phase III ARCHER 1050 trial compared frontline dacomitinib with gefitinib in patients with advancedEGFR-mutant NSCLC, and updated results from the trial were presented at the 2018 ASCO Annual Meeting.1The median PFS with dacomitinib (n = 227) was 14.7 months (95% CI, 11.1-16.6) compared with 9.2 months (95% CI, 9.1-11.0) with gefitinib (n = 225; HR, 0.59; 95% CI, 0.47-0.74;P<.0001). At 2 years, the PFS rate with dacomitinib was 30.6% versus 9.6% with gefitinib.
The median overall survival (OS) was 34.1 months (95% CI, 29.5-37.7) with dacomitinib and 26.8 months (95% CI, 23.7-32.1) with gefitinib (HR, 0.760; 95% CI, 0.582-0.993; 2-sidedP= .0438). At 30 months, the OS probability was 56.2% in the dacomitinib arm compared with 46.3% in the gefitinib arm. Additionally, central nervous system metastases were noted in 11 patients on the gefitinib arm at progression compared with 1 on the dacomitinib arm.
“This is the first study ever to prove the presence of an overall survival benefit when you compare TKIs,” Mok said. “So, in a sense, we can say ‘yes’, a second-generation TKI may be better than a first-generation TKI; however, the toxicity is also high.”
Dacomitinib was granted a priority review designation by the FDA in April 2018 for the treatment of patients with previously untreatedEGFR-positive locally advanced or metastatic NSCLC based on the results of the ARCHER 1050 trial.
The phase III FLAURA study looked at the third-generation EGFR inhibitor osimertinib in comparison with erlotinib or gefitinib in patients withEGFR-mutant NSCLC.2Patients treated with osimertinib showed a median PFS of 18.9 months (95% CI, 15.2-21.4) compared with 10.2 months (95% CI, 9.6-11.1) with the first-generation EGFR TKIs (HR, 0.46; 95% CI, 0.37-0.57;P<.0001). Mok explained that this trial was very important as the survival rates exceeded expectations.
The question with all of these agents is not only which agent should be given in the frontline setting, but also what the optimal sequencing of these EGFR TKIs is, Mok said.
He also highlighted combination regimens with EGFR TKIs as first-line treatments, although none of these combinations have yet been approved for use in the United States.
Gefitinib in combination with carboplatin and pemetrexed chemotherapy was explored in patients with treatment-naïve stage IIIb/IVEGFR-mutant NSCLC in the Japanese NEJ009 trial.3Patients were randomized 1:1 to either the combination or gefitinib alone. The combination arm went on to receive gefitinib and pemetrexed maintenance, and the monotherapy arm went on to receive a platinum-based regimen.
The median PFS in the combination arm was 20.9 months (95% CI, 18.0-24.0) compared with 11.2 months (95% CI, 9.0-13.4) in the single-agent treatment arm (HR, 0.494; 95% CI, 0.391-0.625;P<.001). The combination regimen demonstrated a median OS of 52.2 months (95% CI, 44.0-not available) versus 38.8 months (95% CI, 31.1-50.8) with gefitinib monotherapy (HR, 0.695; 95% CI, 0.520-0.927;P= .013).
“Usually, even in control arms, an OS of 38 months is higher than what we would expect,” Mok said. “So, the question to us is…‘Should we do a similar study outside Japan to validate this interesting finding?’”
Frontline ALK-Targeted Therapies
Mok suggested that physicians should try to apply the survival benefits that have been reported in patients withALK-positive disease to other areas of lung cancer. In the final primary OS analysis from the PROFILE 1014 trial, for example, patients in the control arm had a median OS of 47.5 months, and the median OS was not reached with frontline crizotinib (Xalkori; HR, 0.760; 95% CI, 0.548-1.053;P= .0489).4
Patients in the chemotherapy arm who went on to receive an ALK TKI had a median OS of 49.5 months compared with 12.1 months for patients who received a treatment other than an ALK TKI. In the crizotinib arm, the median OS after subsequent ALK TKI treatment was not reached and was 20.8 months after other subsequent therapy. “In other words, more exposure to TKI is associated with longer survival,” Mok said.
In the global phase III ALEX trial, alectinib (Alecensa) was compared with crizotinib in the frontline setting. The median PFS with alectinib, which was approved for use in the frontline setting inALK-positive patients with NSCLC in November 2017, was 34.8 months compared with 10.9 months with crizotinib (stratified HR, 0.43).5Mok commented that he expected the median OS to be more than 5 years in this study.
He mentioned that several phase III trials are currently ongoing comparing next-generation ALK inhibitors to crizotinib in the frontline setting.
Approved Therapies forROS1andBRAF
Mok noted that while there are often targeted therapies to treat patients with less common mutations, there are few data to support the use of these agents in the frontline setting, and currently there are only FDA-approved agents for treating patients withROS1andBRAFV600E mutations, and not other targetable mutations, such asMET, RET, HER2,andNTRK.
WithROS1rearrangements, for example, which occur in approximately 2.6% of lung adenocarcinomas, many of the agents being investigated for treatment ofALK-positive patients could also work for patients withROS1.However, in the study that led to the approval of crizotinib inROS1-rearranged NSCLC, very few patients received the treatment in the first-line setting. Of 50 patients in the study, only 7 were treatment naïve. Nonetheless, treatment with crizotinib demonstrated an overall response rate of 72% and a median PFS of 19.2 months. At 1 year, the OS rate was 85%.6
“You can argue that you want to treat everyROS1-positive patient with first-line crizotinib, but the data are based on 21 patients [across these studies],” Mok said. There is greater rationale instead, he suggested, for ROS1-targeted therapies to be used in the second-line setting.
WithBRAFmutations, dabrafenib (Tafinlar), a BRAF inhibitor, in combination with trametinib (Mekinist), a MEK inhibitor, have shown similar PFS and response rates across treatment lines. In a phase II study of treatment-naïve patients with metastaticBRAFV600Emutant NSCLC, the investigator-assessed median PFS was 10.9 months (95% CI, 7.0-16.6), and the response rate was 62%.7In previously treated patients, another phase II trial demonstrated a median PFS of 9.7 months with the combination and a response rate of 63.2% in the second line and beyond.8
“Which one is better [forBRAFmutations], first line or second line? Based on these data, I can’t say there’s much difference,” Mok said.