GSK2118436

Dabrafenib plus trametinib in patients with previously untreated BRAFV600E-mutant metastatic non-small-cell lung cancer: an open-label, phase 2 trial
David Planchard, Egbert F Smit, Harry J M Groen, Julien Mazieres, Benjamin Besse, Åslaug Helland, Vanessa Giannone, Anthony M D’Amelio Jr, Pingkuan Zhang, Bijoyesh Mookerjee, Bruce E Johnson
Summary
Background BRAFV600E mutation occurs in 1–2% of lung adenocarcinomas and acts as an oncogenic driver. Dabrafenib, alone or combined with trametinib, has shown substantial antitumour activity in patients with previously treated BRAFV600E-mutant metastatic non-small-cell lung cancer (NSCLC). We aimed to assess the activity and safety of dabrafenib plus trametinib treatment in previously untreated patients with BRAFV600E-mutant metastatic NSCLC.

Methods In this phase 2, sequentially enrolled, multicohort, multicentre, non-randomised, open-label study, adults (≥18 years of age) with previously untreated metastatic BRAFV600E-mutant NSCLC were enrolled into cohort C from 19 centres in eight countries within North America, Europe, and Asia. Patients received oral dabrafenib 150 mg twice per day plus oral trametinib 2 mg once per day until disease progression, unacceptable adverse events, consent withdrawal, or death. The primary endpoint was investigator-assessed overall response, defined as the percentage of patients who achieved a confirmed complete response or partial response per Response Evaluation Criteria In Solid Tumors version 1.1. The primary and safety analyses were by intention to treat in the protocol-defined population (previously untreated patients). The study is ongoing, but no longer recruiting patients. This trial is registered with ClinicalTrials.gov, number NCT01336634.

Findings Between April 16, 2014, and Dec 28, 2015, 36 patients were enrolled and treated with first-line dabrafenib plus trametinib. Median follow-up was 15·9 months (IQR 7·8–22·0) at the data cutoff (April 28, 2017). The proportion of patients with investigator-assessed confirmed overall response was 23 (64%, 95% CI 46–79), with two (6%) patients achieving a complete response and 21 (58%) a partial response. All patients had one or more adverse event of any grade, and 25 (69%) had one or more grade 3 or 4 event. The most common (occurring in more than two patients) grade 3 or 4 adverse events were pyrexia (four [11%]), alanine aminotransferase increase (four [11%]), hypertension (four [11%]), and vomiting (three [8%]). Serious adverse events occurring in more than two patients included alanine aminotransferase increase (five [14%]), pyrexia (four [11%]), aspartate aminotransferase increase (three [8%]), and ejection fraction decrease (three [8%]). One fatal serious adverse event deemed unrelated to study treatment was reported (cardiorespiratory arrest).

Interpretation Dabrafenib plus trametinib represents a new therapy with clinically meaningful antitumour activity and a manageable safety profile in patients with previously untreated BRAFV600E-mutant NSCLC.

Funding Novartis.

Lancet Oncol 2017
Published Online September 11, 2017 http://dx.doi.org/10.1016/ S1470-2045(17)30679-4
See Online/Comment http://dx.doi.org/10.1016/ S1470-2045(17)30678-2
Gustave Roussy, Villejuif, France (D Planchard MD,
B Besse MD); Vrije Universiteit VU Medical Centre, Amsterdam, Netherlands (Prof E F Smit MD); University of Groningen and University Medical Center Groningen,
Groningen, Netherlands (Prof H J M Groen MD); Rangueil-Larrey Hospital and Paul Sabatier University,
Toulouse, France
(Prof J Mazieres MD); Oslo University Hospital, Department of Oncology, Norwegian Radium Hospital, Oslo, Norway (Å Helland MD); Novartis Pharmaceuticals Corporation, East Hanover, NJ, USA (V Giannone MPA,
A M D’Amelio Jr PhD,
P Zhang MD, B Mookerjee MD); and Dana-Farber Cancer Institute, Boston, MA, USA (Prof B E Johnson MD)
Correspondence to:
Prof Bruce E Johnson, Department of Medical Oncology,

Introduction
Non-small-cell lung cancer (NSCLC) remains a leading cause of cancer-related deaths worldwide. Progress has been made in characterisation of oncogenic driver mutations that contribute to the molecular pathogenesis of lung cancer. Activating mutations in EGFR plus rearrangements in ALK and ROS1 have been identified as contributing oncogenic drivers of molecular pathogenesis in lung cancer, which has led to a rapid development of targeted therapies and more individualised treatment strategies for patients with NSCLC.1–3
Activating BRAF mutations are considered an alternative oncogenic driver in NSCLC that are almost never observed in tumours harbouring EGFR mutations, ALK rearrangements, or ROS1 rearrangements. The most common BRAF mutation, V600E (Val600Glu), is observed
in 1–2% of lung adenocarcinomas.4–6 The prognostic value of BRAFV600E in NSCLC remains unclear; however, some studies have associated this mutation with poor outcomes and lower response rates to platinum-based chemotherapy in patients with this disease compared with those without BRAF mutations.7,8 Among patients with driver mutations, outcomes for patients treated with targeted therapies are prolonged compared with those who do not receive genotype-directed treatment.5 Immune checkpoint inhibitors as monotherapy9 or in combination with chemotherapy10 have demonstrated significantly prolonged survival for patients with NSCLC, especially those with high PD-L1 overexpression. However, nearly 50% of patients do not respond at all to these therapies. Therefore, an unmet need remains for effective targeted agents in this patient population.
Dana-Farber Cancer Institute, Boston, MA 02215, USA
[email protected]. edu

See Online for appendix
In this Article we discuss the third (cohort C) of three sequentially enrolled cohorts in this phase 2 study of patients with BRAFV600E-mutant metastatic NSCLC. The results of the first two cohorts of this trial (cohort A and cohort B) were previously reported,11,12 demonstrating a response of 67% and duration of response of 9·8 months in patients with previously treated BRAFV600E-mutant metastatic NSCLC who received the combination of dabrafenib plus trametinib.13 Indirect comparison across cohorts showed that the antitumour activity of the combination was higher than that observed with dabrafenib monotherapy (the proportion of patients achieving an overall response was 67% with the combination vs 33% with dabrafenib monotherapy). The safety profile of the combination was manageable and similar to that previously reported for the combination in patients with BRAFV600-mutant metastatic melanoma.14,15 In cohort C, we aimed to assess the antitumour activity of dabrafenib plus trametinib in previously untreated patients with BRAFV600E-mutant metastatic NSCLC.

Methods
Study design and participants
As part of a phase 2, sequentially enrolled, multicohort, multicentre, non-randomised, open-label trial, this study enrolled adults (≥18 years) with metastatic BRAFV600E- mutant NSCLC without previous systemic treatment for metastatic disease from 19 centres in eight countries within North America, Europe, and Asia (appendix p 4). Local testing methods chosen by each study site were
implemented in laboratories approved by the Clinical Laboratory Improvement Amendments (or its equivalent outside the USA) to determine BRAFV600E mutational status for purposes of enrolment. Tumour tissue samples (archival or pretreatment biopsy) were also collected from all enrolled patients at screening to centrally confirm BRAFV600E mutation status using the Oncomine Dx Target Test (ThermoFisher Scientific, Waltham, MA, USA; additional details about tissue sampling are included in appendix p 3). Patients needed to have measurable disease per Response Evaluation Criteria In Solid Tumors (RECIST) version 1.1, an Eastern Cooperative Oncology Group performance status of 2 or less, and an estimated life expectancy of at least 3 months. Laboratory assessments for eligibility were: haematology (absolute neutrophil count ≥1·5 × 10⁹ cells per L, haemoglobin
≥90 g/L, platelet count ≥100 × 10⁹ per L, prothrombin time/
international normalised ratio and partial thromboplastin time of ≤1·5 × upper limit of normal [ULN]), hepatic (total bilirubin ≤1·5 × ULN, aspartate aminotransferase and alanine aminotransferase ≤2·5 × ULN), renal (at least one of the following: serum creatinine ≤1·5 mg/dL [132⋅6 μmol/L] or creatinine clearance of ≥50 mL/min), and cardiac (left ventricular ejection fraction of at least the lower limit of normal by echocardiography). Patients were required to have previous testing for EGFR and ALK mutations in lung cancer tissue that was confirmed in a Clinical Laboratory Improvement Amendments-certified laboratory. Patients who tested positive for these mutations were not excluded; however they had to have received

previous EGFR or ALK inhibitor therapy (although patients in cohort C could have no previous systemic therapy for metastatic disease, patients who tested positive could have received these agents as adjuvant therapy). Patients who had received previous BRAF or MEK inhibitor therapy were ineligible. Brain metastases were not permitted unless they were asymptomatic, untreated, and measured less than 1 cm, or, if treated, were clinically and radiographically stable 3 weeks after local therapy (which could include surgery or radiotherapy). Pregnant women, patients with confirmed hepatitis B or C virus infection, or patients with a history or signs of cardiovascular risk were ineligible. Patients were also excluded if they had a history of interstitial lung disease, pneumonitis, or a history of or current evidence of retinal vein occlusion.
This study was done in accordance with the provisions of the Declaration of Helsinki and Good Clinical Practice guidelines, and the protocol was approved by the institutional review board at each study site. All patients provided written informed consent.

Procedures
Patients were treated with oral dabrafenib 150 mg twice per day plus oral trametinib 2 mg once per day until disease progression, unacceptable adverse events, consent withdrawal, or death, as previously reported.12 Treatment beyond progression was allowed in patients who had a confirmed response per RECIST version 1.1 or stable disease for 12 weeks or more during study treatment and were considered by the investigator to be clinically benefiting from therapy. Dose modifications were permitted to manage intolerable adverse events of grade 2 or higher. When a dose adjustment was required, both treatment doses were reduced simultaneously, except in the case of pyrexia or uveitis, in which only the dabrafenib dose was reduced, or retinal vein occlusion, retinal pigment epithelial detachment, left ventricular ejection fraction reduction, pneumonitis, or interstitial lung disease, in which only the trametinib dose was reduced. Dabrafenib was reduced to 100 mg on first reduction, 75 mg on second reduction, and 50 mg on third reduction, with no doses less than 50 mg administered. Trametinib was reduced to 1·5 mg on first reduction and 1 mg on second reduction, with no doses less than 1 mg given.
Radiologically detected disease per RECIST version 1.1 was assessed with CT at baseline, at week 6, every 6 weeks until week 36, and then every 12 weeks. Responses were confirmed by repeat assessments 4–7 weeks after an initial response. RECIST scans were also reviewed by an independent review committee. Patient follow-up for subsequent treatment and survival every 12 weeks was continued for those who discontinued study treatment until death or study completion. Adverse events, laboratory values (ie, haematology, clinical chemistry), and vital signs, graded according to the Common Terminology Criteria for Adverse Events version 4.0, were

evaluated at least once every 3 weeks. The protocol required reporting of cutaneous squamous -cell carcinoma for grade 3 or higher events, pyrexia with symptoms and left ventricular ejection fraction decreases of grade 2 or higher, and other protocol-specified serious adverse events, regardless of whether they met the standard definition of a serious adverse event. Complete details on study assessments are included in the appendix (p 3).
Optional biopsy samples could be collected at week 6 and at the end of study per protocol. The method of optional biopsy collection was not mandated in the protocol and any clinically available tumour or cytological specimen could be accepted.

Outcomes
The primary endpoint was investigator-assessed overall response, defined as the percentage of patients who achieved a confirmed complete response or partial response per RECIST version 1.1. Secondary endpoints were the proportion of patients achieving a response as assessed by independent reviewer; progression-free survival (defined as the interval between the first dose of study drug and the earliest date of disease progression or death from any cause); and duration of response (defined as the time from first documented evidence of complete or partial response until the time of first documented disease progression or death from any cause, whichever occurred first) assessed by investigator and independent reviewer; overall survival (defined as the time from first dose of study drug to death from any cause); safety; and pharmacokinetic assessment (the pharmacokinetic analysis has not been completed and will be reported separately).

Statistical analysis
Patients with previously untreated BRAFV600E-mutant NSCLC were monitored during the study for clinical response with a one-stage binomial design, and no formal interim analysis was done in this cohort. The null hypothesis was that overall response was not clinically meaningful (≤30% of patients achieving an overall response), and the alternative hypothesis was that the proportion of patients achieving an overall response was at least 60% for first-line dabrafenib plus trametinib in patients with BRAFV600E-mutant NSCLC. Sample size considerations were based on a targeted response of at least 60% and the incorporation of an exact binomial test, which corresponded to a planned enrolment of 25 patients, a type I error of 0·044, and power of 92·2% for the one-stage binomial design. Additional patients were permitted to enrol per protocol to ensure an adequate number of evaluable patients with centrally confirmed response assessments and BRAF mutation status, which did not change our statistical assumptions. Activity and safety evaluations were based on the intention-to-treat principle in the protocol-defined first-line population. Patients considered non-evaluable had either no

Figure 1: Trial profile

post-baseline CT scan or discontinued after less than 12 weeks without documented progression. 95% CIs for overall response were calculated with the Clopper-Pearson method. Kaplan-Meier methods were used to summarise duration of response, progression-free survival, and overall survival. Medians and 95% CIs (calculated with the Brookmeyer-Crowley method16) are presented. Sensitivity analyses were done for overall response, duration of response, and progression-free survival with scan data reviewed by an independent review committee. A post-hoc analysis was performed to evaluate investigator-assessed response based on patient smoking history. All statistical analyses were done with SAS version 9.3 or higher.
This trial is registered with ClinicalTrials.gov, number NCT01336634.

Role of the funding source
This study was sponsored by GlaxoSmithKline; dabra- fenib and trametinib are assets of Novartis AG as of March 2, 2015. The study was designed by the authors and the sponsor. Data were collected by the study site staff and monitored by GlaxoSmithKline and Novartis AG, and Novartis AG was involved in the data analysis, data interpretation, and writing of the report. All authors had full access to all data in the study and had final responsibility for the decision to submit for publication.
Results
Between April 16, 2014, and Dec 28, 2015, 36 patients were enrolled, including 34 patients in cohort C (figure 1). The cutoff date for safety and activity data was April 28, 2017. As previously reported, two patients initially enrolled in cohort B owing to protocol deviation had received no previous systemic anticancer therapy for metastatic disease, and were thus excluded from the analysis of cohort B.12 Since the requirement for previous treatment was the only major difference
Adenocarcinoma 32 (89%)
Adenosquamous carcinoma (predominantly adenocarcinoma) 1 (3%)
Adenosquamous carcinoma (predominantly squamous-cell carcinoma) 1 (3%)
Large-cell carcinoma 1 (3%)
NSCLC not otherwise specified 1 (3%)
Smoking history*
Never 10 (28%)
Current 5 (14%)
Former 21 (58%)

between the eligibility criteria used for enrolment in cohorts B and C, these two patients were combined with cohort C patients for the purpose of the present report so that this patient information is not lost. Therefore, 36 patients were enrolled and included in all analyses reported here. Because this study enrolled patients with BRAFV600E-mutant disease based on local laboratory testing, typically using multiplex genotyping, the exact number of patients with NSCLC evaluated for BRAFV600E mutation at participating study sites was not recorded. However, based on the frequency of BRAFV600E mutations in NSCLC (1–2%),4–6 it can be estimated that screening of approximately 2000–4000 patients was needed to identify the 36 patients included in this analysis.

Baseline characteristics of the 36 patients enrolled are reported in table 1. Nearly all patients had non- squamous histology (except for one patient who had adenosquamous histology with predominant squamous- cell carcinoma). No patients tested positive for EGFR or ALK alterations. At data cutoff, 11 (31%) patients remained on study treatment and 25 (69%) had discontinued because of adverse events (eight [22%]), disease progression (14 [39%]), investigator discretion
(two [6%]), or patient decision (one [3%]) (figure 1; appendix p 5). At data cutoff, 19 (53%) patients were alive and 17 (47%) had died.

Overall response (complete and partial responses) 23 (64%; 46–79) 23 (64%; 46–79)
Disease control (complete responses, partial responses, and stable disease) 27 (75%; 58–88) 26 (72%; 55–86)
Complete response 2 (6%) 2 (6%)
Partial response 21 (58%) 21 (58%)
Stable disease 4 (11%) 3 (8%)
Progressive disease 5 (14%) 7 (19%)
Not evaluable 4 (11%) 3 (8%)

At a median follow-up of 15·9 months (IQR 7·8–22·0; appendix p 5), the proportion of patients with inves- tigator-assessed confirmed overall response was 23 (64%, 95% CI 46–79), including two (6%) who had a complete response and 21 (58%) who had a partial response (table 2, figure 2). Four (11%) patients had stable disease, resulting in 27 (75%, 95% CI 58–88) patients with disease control.
Maximum reduction from baseline measurement (%)
The investigator-assessed median duration of response was 10·4 months (95% CI 8·3–17·9; 16 [70%] of 23 responders progressed or died) with a Kaplan-Meier estimated 6-month duration of response of 86% (95% CI 62–95; figure 3A). With 24 (67%) of 36 total patients having progressed or died at the time of the analysis, the median investigator-assessed progression-free survival was 10·9 months (95% CI 7·0–16·6), and the 6-month progression-free survival was 72% (53–84; figure 3B). Based on independent review committee assessment in a sensitivity analysis, the median duration of response was 15·2 months (95% CI 7·8–23·5) and median progression-free survival was 14·6 months (7·0–22·1; appendix p 9). Differences in median duration of response and progression-free survival between investi- gator and independent review committee assessments were mainly driven by censored observations for the committee (five patients who were assessed by the investigators as having progressive disease had values for progression-free survival or duration of response close to the medians). Because the independent review committee did not assess the last scans for these patients as progressive disease, and the patients did not have any

Figure 2: Confirmed maximum reduction in target lesions according to investigator assessment
Dashed line at −30% represents the threshold for partial response, according to Response Evaluation Criteria In Solid Tumors version 1.1. Bars show maximum reduction from baseline sum of diameters by best confirmed response. Two patients were not included because they did not have a post-baseline assessment of target lesions.

Progression-free survival (%)
Patient

Overall survival (%)
Figure 3: Investigator-assessed duration of response (A), investigator-assessed estimated progression-free survival (B), and estimated overall survival (C)
Arrows indicate censored patients with follow-up ongoing. Tick marks represent censored patients. Dotted lines in Kaplan-Meier estimated curves represent 95% CIs.

further tumour assessment scans to assess, they were not counted as having progressive disease by the committee and their progression-free survival and duration of response were censored when they started to receive subsequent anticancer therapy (these therapies

were received after the investigator noted progressive disease).
At data cutoff, 17 (47%) patients had died, with a median overall survival of 24·6 months (95% CI 12·3–not estimable; figure 3C). The 2-year overall survival was 51% (95% CI 33–67). Of 25 patients who discontinued study treatment, nine (36%) went on to receive at least one subsequent therapy following discontinuation of study drug, one (4%) was lost to follow-up before information on subsequent anticancer therapy was recorded, two (8%) withdrew consent precluding collection of further data, three (12%) remained in follow- up with subsequent therapy data continued after the data cutoff, and ten (40%) died without receiving any subsequent anticancer therapy. In the nine patients who went on to receive further therapy, subsequent anticancer therapies included chemotherapy (n=6), biological therapy (n=4), radiotherapy (n=4), small molecule targeted therapy (n=2), surgery (n=1), and immuno- therapy (n=1). The median time from discontinuation of study treatment to the start of subsequent therapy was 43 days (IQR 13·5–113·0).
The median duration of treatment for dabrafenib was 9·0 months (IQR 3·1–20·5) and that of trametinib was 9·5 months (3·2–19·3; appendix p 6). Dose reductions of dabrafenib occurred in 17 (47%) patients; 11 (65%) of 17 required one dose reduction, and three (18%) each required two and three or more dose reductions. Trametinib dose reductions occurred in ten (28%) of 36 patients, with a single dose reduction in four (40%) of these patients, and two dose reductions in six (60%).
All patients had at least one adverse event of any grade, most commonly (ie, in >30% of patients) pyrexia, nausea, diarrhoea, fatigue, peripheral oedema, decreased appetite, dry skin, and vomiting (table 3). 25 (69%) patients had at least one grade 3 or 4 adverse event, most commonly (ie, in more than two patients) pyrexia (four [11%]), alanine aminotransferase increase (four [11%]), hypertension (four [11%]), and vomiting (three [8%]). Serious adverse events occurring in more than two patients included alanine aminotransferase increase (five [14%]), pyrexia (four [11%]), aspartate aminotransferase increase (three [8%]), and ejection fraction decrease (three [8%]; appendix p 7). Adverse events led to permanent discontinuation, dose interruption or delay, and dose reduction in eight (22%), 27 (75%), and 14 (39%) patients, respectively. One patient died from a serious adverse event (cardiorespiratory arrest) considered unrelated to study treatment by the investigators; no radiological disease progression was observed before death. One patient died of a stroke more than 1 year after discontinuation of study drug (death was not believed to be related to study treatment). The remaining 15 deaths were attributed to the disease under study (appendix p 8). A post-hoc subgroup analysis by baseline smoking history showed similar responses in patients who reported no history of smoking (response in five [50%] of

ten patients), those with 30 pack-years of smoking history or less (response in 13 [76%] of 17 patients), and more than 30 pack-years of smoking history (response in four [57%] of seven patients). Patient numbers were small and interpretations of this exploratory analysis should be made with caution. Two patients had baseline brain metastases that were non-target lesions. For both patients, a best response of non-complete response or non- progressive disease in the brain lesions was reported, with an overall best response of partial response.
Discussion
The results from this phase 2 trial showed substantial antitumour activity of dabrafenib plus trametinib therapy in patients with treatment-naive BRAFV600E-mutant metastatic NSCLC. The protocol-defined primary objective was met (target response rate ≥60%), with 64% of patients achieving a confirmed overall response Responses were durable, with a median response duration of 10·4 months. This study represents the first evaluation, to our knowledge, of combination BRAF and MEK inhibition in patients with previously untreated NSCLC.
Because data are scarce in patients with BRAFV600E- mutant NSCLC and effective targeted therapeutic options remain a crucial unmet need for this patient population, results from this study continue to be of clinical importance. Notably, available BRAF-mutant inhibitor regimens, including dabrafenib and trametinib combination therapy, have shown efficacy in patients with BRAF mutations that activate BRAF in its monomeric state, which are confined to mutations in codon 600, and are sensitive to BRAF mutant-specific inhibitors. Therefore, the combination is not expected to be active in patients with non-V600 mutations, despite a small amount of preclinical evidence to the contrary.17 However, the concentration of dabrafenib (2·5 μmol/L) plus trametinib (25 nmol/L) and duration of treatment (cells treated continously for 3–5 days) used in this preclinical study are unlikely to be achieved in human beings at the approved dose of both agents.
The activity of dabrafenib plus trametinib in previously untreated patients in this study was improved compared with previous observations in patients with metastatic NSCLC and activating BRAF mutations who primarily received standard-of-care chemotherapy as first-line treatment in a real-world setting, in which 23% of the
70 patients with data available achieved an overall response and median progression-free survival was 5·6 (95% CI 4·2–7·1) months in the 81 patients with data available.4 The response, duration of response, and progression-free survival of treatment-naive patients with combination dabrafenib and trametinib in this study were similar to those reported for the previously treated cohort (cohort B) of this trial12 (overall response 64% vs 63%; median duration of response 10·4 vs 9·0 months; median progression-free survival 10·9 vs

Grade 1–2 Grade 3 Grade 4 Grade 5
Total 10 (28%) 23 (64%) 2 (6%) 1 (3%)
Pyrexia 19 (53%) 4 (11%) 0 0
Nausea 20 (56%) 0 0 0
Diarrhoea 12 (33%) 1 (3%) 0 0
Fatigue 13 (36%) 0 0 0
Peripheral oedema 13 (36%) 0 0 0
Vomiting 9 (25%) 3 (8%) 0 0
Dry skin 12 (33%) 0 0 0
Decreased appetite 12 (33%) 0 0 0
Chills 9 (25%) 0 0 0
Headache 9 (25%) 0 0 0
Rash 7 (19%) 1 (3%) 0 0
Dizziness 8 (22%) 0 0 0
Cough 8 (22%) 0 0 0
Alanine aminotransferase increase 2 (6%) 4 (11%) 0 0
Dyspnoea 4 (11%) 2 (6%) 0 0
Hypotension 4 (11%) 2 (6%) 0 0
Back pain 6 (17%) 0 0 0
Weight decrease 6 (17%) 0 0 0
Abdominal pain 4 (11%) 1 (3%) 0 0
Anaemia 4 (11%) 1 (3%) 0 0
Arthralgia 4 (11%) 1 (3%) 0 0
Constipation 5 (14%) 0 0 0
Insomnia 5 (14%) 0 0 0
Myalgia 5 (14%) 0 0 0
Hypertension 0 4 (11%) 0 0
Hyponatraemia 2 (6%) 2 (6%) 0 0
Aspartate aminotransferase increase 3 (8%) 1 (3%) 0 0
Asthenia 3 (8%) 1 (3%) 0 0
Pruritus 3 (8%) 1 (3%) 0 0
Pain in extremity 3 (8%) 1 (3%) 0 0
Erythema 4 (11%) 0 0 0
Dysphonia 3 (8%) 0 0 0
Malaise 4 (11%) 0 0 0
Musculoskeletal chest pain 4 (11%) 0 0 0
Urinary tract infection 4 (11%) 0 0 0
Dehydration 2 (6%) 1 (3%) 0 0
Ejection fraction decrease 1 (3%) 2 (6%) 0 0
Pulmonary embolism 1 (3%) 2 (6%) 0 0
Weight increase 2 (6%) 1 (3%) 0 0
(Table 3 continues on next page)

9·7 months, respectively). These findings indicate that physicians have some flexibility to treat patients with this targeted therapy combination in either the first-line setting or following chemotherapy, and thus provide several strategies that can be considered to accommodate individual patient needs and the logistical challenges of BRAF mutation testing. The activity of first-line dabrafenib plus trametinib also seemed to be similar to the efficacy shown in randomised phase 3 trials for targeted agents for other genomic aberrations in NSCLC used as first-line treatment, such as EGFR tyrosine

Grade 1–2 Grade 3 Grade 4 Grade 5
(Continued from previous page)
Hyperglycaemia
1 (3%)
1 (3%)
0
0
Hypophosphataemia 1 (3%) 1 (3%) 0 0
Lymphocyte count decrease 0 2 (6%) 0 0
Blood creatine phosphokinase increase 0 1 (3%) 0 0
Blood phosphorus decrease 0 1 (3%) 0 0
Gastric haemorrhage 0 1 (3%) 0 0
Gastrointestinal toxicity 0 1 (3%) 0 0
Neutropenia 0 1 (3%) 0 0
Pancytopenia 0 1 (3%) 0 0
Peripheral sensory neuropathy 0 1 (3%) 0 0
Rash maculopapular 0 1 (3%) 0 0
Renal artery thrombosis 0 1 (3%) 0 0
Splenic thrombosis 0 1 (3%) 0 0
Vertigo positional 0 1 (3%) 0 0
White blood cell count decrease 0 1 (3%) 0 0
Lymphopenia 0 0 1 (3%) 0
Multiple injuries 0 0 1 (3%) 0
Cardiac arrest 0 0 0 1 (3%)
Respiratory arrest 0 0 0 1 (3%)

kinase inhibitors in patients with activating EGFR mutations (overall response approximately 70%; progression-free survival 9·7–14·7 months)3,18,19 and ALK and ROS1 inhibitors in patients with ALK and ROS1 rearrangements (overall response 62–83%; progression- free survival 10·4–25·7 months).1,2,20–22
Although the presence of brain metastases is fairly common in patients with NSCLC at diagnosis (around 10% of patients) or at some point during the course of their disease (up to 40%), similar to most initial trials of therapies for advanced cancer,23 patients with active brain metastases were excluded from this study. Patients with NSCLC with symptomatic brain metastases were excluded because these patients have poor prognosis and are unlikely to meet the minimal life expectancy (>3 months) included as part of the inclusion criteria. Additionally, patients with asymptomatic untreated brain metastases larger than 1 cm might be candidates for other treatments to manage brain metastases (such as local therapies including surgical intervention or radiotherapy), which could have potentially jeopardised the assessment of the study drugs. Moreover, when this study was designed, no data were available to provide evidence for whether the combination of dabrafenib and trametinib has intracranial activity in patients with brain metastases with either BRAFV600E-mutant melanoma or NSCLC, therefore providing another reason to exclude patients with brain metastases from this study. However, recently reported results from the phase 2 COMBI-MB

trial24 showing activity of dabrafenib plus trametinib in patients with BRAFV600-mutant melanoma with brain metastases might support further exploration of this combination in patients with BRAFV600-mutant NSCLC with brain metastases.
Checkpoint inhibitor immunotherapies (ie, anti-PD-1 or anti-PD-L1 agents) also represent emerging front-line treatment options for patients with NSCLC. In the phase 3 KEYNOTE-024 study9 of previously untreated patients with NSCLC tumours that had 50% or more PD-L1-positive cells, pembrolizumab was superior to chemotherapy (progression-free survival 10·3 months, 95% CI 6⋅7–not reached vs 6·0 months, 4⋅2–6⋅2),9 resulting in the US Food and Drug Administration approval of pembrolizumab for this patient population. Results from the phase 1 CheckMate 012 study25 indicated that the combination of nivolumab plus ipilimumab also provided clinical benefit in previously untreated patients with NSCLC (median duration of response not reached after 12·8 months of follow-up). Notably, responses were observed in a smaller subset of patients with these agents (40–45%) than recorded in this report.9,25 The combination of pembrolizumab plus pemetrexed and carboplatin showed a higher response (55%) when compared indirectly with pembrolizumab monotherapy (45% in patients with PD-L1 expression in at least 50% of tumour cells9) and 13·0-month median progression-free survival (95% CI 8⋅3–not estimable) in patients with metastatic NSCLC.10 However, whether the enhanced response with pembrolizumab plus chemotherapy will translate into improved long-term outcomes for these patients is not known, and no data are yet available characterising the effect of these checkpoint inhibitor regimens specifically in patients with BRAFV600E- mutant NSCLC. Furthermore, nearly 50% of patients did not have a clinical response to this regimen, suggesting that targeted agents for patients with actionable mutations address an unmet medical need and remain an important component in the clinical management of these patients, with high responses achievable in selected patients. Notably, although some recent NSCLC studies have reported poor responses to checkpoint inhibitor therapies in patients with driver mutations (eg, cMET, EGFR, and ALK),26,27 further research will be needed to assess the efficacy of immunotherapeutic agents in these patients with BRAFV600E mutations. Overall, these findings indicate that patients with NSCLC tumours harbouring actionable driver mutations, such as EGFR, ALK, ROS1,1–3 and now BRAFV600E, might be most amenable to management with precision oncology-based treatment strategies, whereas patients with tumours characterised by other clinical features might be better suited for immunotherapy-based regimens.
Acquired resistance to BRAF and MEK inhibitors in patients with melanoma has been well characterised, and is associated with mechanisms that result in reactivation of the MAPK pathway (ie, receptor tyrosine kinase upregulation, NRAS mutations, activating MEK1/2

mutations, and other BRAF alterations) and changes in PI3K–PTEN–AKT pathway signalling.28 Preclinical studies have indicated that, in addition to BRAF alterations, constitutive autocrine EGFR signalling might also have a role in the development of resistance to targeted therapy in BRAFV600E-mutant NSCLC;17,29,30 however, further clinical investigation is needed to confirm these results in patients with BRAFV600E-mutant NSCLC treated with dabrafenib and trametinib. Per the protocol of the study described here, optional biopsies could be collected at week 6 and at the end of the trial. Acquisition of post-progression biopsies and follow-up across all cohorts of this study are ongoing and await adequate numbers of samples for analyses as the patients progress.
Consistent with that reported for patients in the previously treated cohort in this trial and in melanoma studies,11,14,15 the safety profile of dabrafenib plus trametinib in the current study was manageable, with no unexpected toxicities observed. Eight (22%) patients discontinued study treatment because of adverse events; however, we note that this proportion of patients is based on a small sample size. Therefore, any comparisons of these data against those reported for phase 3 melanoma trials evaluating dabrafenib plus trametinib (discontinuation rate owing to adverse events of 11–13%)14,15 should be interpreted with caution. Similar to the clinical experience with dabrafenib plus trametinib in melanoma,14,15 pyrexia was the most frequently observed adverse event in this study. Notably, experience with the combination in patients with BRAFV600-mutant melanoma has shown that most grade 3 or 4 adverse events can be managed through dose modification, which provides a framework for physicians to manage and mitigate risks of unacceptable toxicity.
Although the prognostic implications of BRAFV600E mutations in NSCLC remain unclear,4,7,8 data from the current study support an increased emphasis on BRAF mutation testing at diagnosis to help to inform personalised therapeutic decisions. Furthermore, because the BRAF mutation frequency is low in patients with NSCLC (1–2%),5 future inclusion of rare mutations such as BRAFV600E in umbrella trials in patients with NSCLC could help to enhance enrolment and allow for larger trial populations.
We acknowledge that the non-randomised design was a limitation of this study, whereby the results reported here cannot be directly compared with those for other NSCLC treatments. The small sample size in this cohort (cohort C) also limits the extent of interpretation of the outcomes for these previously untreated patients. Additionally, patient-reported outcomes and health economic analyses were not included as endpoints for this study (questionnaires for patient-reported outcomes were not administered in this trial and thus no assessment of these is planned; health economic analyses are being done outside the current study and will be reported separately). In this first report, to our knowledge, of a first-line BRAF and MEK inhibitor combination strategy in

NSCLC, dabrafenib plus trametinib elicited a high proportion of durable responses with manageable toxicity in previously untreated patients with BRAFV600E-mutant metastatic NSCLC. As such, this combination represents a promising novel treatment option for this patient population that recently received marketing authorisation from the European Commission and approval from the US Food and Drug Administration (FDA), irrespective of previous treatment history, contemporaneously with a next-generation sequencing-based companion diagnostic (the Oncomine Dx Target Test has been approved by the FDA as a companion diagnostic for identification of BRAFV600E-mutant patients). As individualised treatment algorithms continue to be refined with the emergence of optimised sequencing strategies and new targeted therapeutic options, including dabrafenib plus trametinib, outcomes in patients with NSCLC are likely to continue to improve.
Contributors
DP, AMD’A, BM, and BEJ contributed to conception and design of the study. HJMG and ÅH provided study materials or patients. DP, EFS, HJMG, JM, BB, ÅH, PZ, and BM contributed to collection and assembly of data. DP, JM, HJMG, BB, AMD’A, PZ, BM, and BEJ contributed to data analysis and interpretation. JM, PZ, and BEJ were responsible for the literature search. AMD’A and BEJ developed the tables and figures. DP, HJMG, JM, BB, ÅH, VG, AMD’A, PZ, BM, and BEJ wrote the report. DP, EFS, HJMG, JM, BB, ÅH, VG, AMD’A, PZ, BM, and
BEJ gave final approval of the report.
Declaration of interests
DP acts as an adviser for AstraZeneca, Boehringer Ingelheim, Bristol-Myers Squibb, Merck Sharp & Dohme, Novartis, Pfizer, and Roche. HJMG acts as an adviser for AbbVie, Bristol-Myers Squibb,
Eli Lilly, Merck Sharpe & Dohme, Novartis, Pfizer, and Roche and has received research funding from Eli Lilly and Roche unrelated to the current work. BB’s institution received a grant from Novartis for this study. VG, AMD’A, PZ, and BM are currently employees of Novartis. AMD’A and BM were employees of GlaxoSmithKline during a portion of the study. AMD’A and BM own stock in GlaxoSmithKline and Novartis. BEJ has received personal fees from Amgen, AstraZeneca, Boehringer Ingelheim, Chugai, Clovis, Eli Lilly, Genentech,
KEW Group, Merck Sharp & Dohme, and Novartis; has received a grant from Toshiba unrelated to the current work; and received shares of postmarket revenue for an EGFR genotyping patent paid to
Dana-Farber Cancer Institute. EFS, JM, and ÅH declare no competing
interests.
Acknowledgments
Medical writing assistance in the form of collating author comments and editorial assistance was provided by Michael Demars and
Amanda L Kauffman (ArticulateScience LLC, Hamilton, NJ, USA), and was funded by Novartis Pharmaceuticals Corporation. This study was sponsored by GlaxoSmithKline; dabrafenib and trametinib are assets of Novartis AG as of March 2, 2015.
References
⦁ Shaw AT, Kim DW, Nakagawa K, et al. Crizotinib versus chemotherapy in advanced ALK-positive lung cancer.
N Engl J Med 2013; 368: 2385–94.
⦁ Shaw AT, Ou SH, Bang YJ, et al. Crizotinib in ROS1-rearranged non-small-cell lung cancer. N Engl J Med 2014; 371: 1963–71.
⦁ Rosell R, Carcereny E, Gervais R, et al, on behalf of the Spanish Lung Cancer Group in collaboration with the Groupe Français de Pneumo-Cancérologie and the Associazione Italiana Oncologia Toracica. Erlotinib versus standard chemotherapy as first-line treatment for European patients with advanced EGFR
mutation-positive non-small-cell lung cancer (EURTAC): a multicentre, open-label, randomised phase 3 trial.
Lancet Oncol 2012; 13: 239–46.

⦁ Barlesi F, Mazieres J, Merlio JP, et al, for the Biomarkers France contributors. Routine molecular profiling of patients with advanced non-small-cell lung cancer: results of a 1-year nationwide programme of the French Cooperative Thoracic Intergroup (IFCT). Lancet 2016; 387: 1415–26.
⦁ Kris MG, Johnson BE, Berry LD, et al. Using multiplexed assays of oncogenic drivers in lung cancers to select targeted drugs.
JAMA 2014; 311: 1998–2006.
⦁ Paik PK, Arcila ME, Fara M, et al. Clinical characteristics of patients with lung adenocarcinomas harboring BRAF mutations.
J Clin Oncol 2011; 29: 2046–51.
⦁ Cardarella S, Ogino A, Nishino M, et al. Clinical, pathologic, and biologic features associated with BRAF mutations in non-small cell lung cancer. Clin Cancer Res 2013; 19: 4532–40.
⦁ Marchetti A, Felicioni L, Malatesta S, et al. Clinical features and outcome of patients with non-small-cell lung cancer harboring BRAF mutations. J Clin Oncol 2011; 29: 3574–79.
⦁ Reck M, Rodríguez-Abreu D, Robinson AG, et al. Pembrolizumab versus chemotherapy for PD-L1-positive non-small-cell lung cancer. N Engl J Med 2016; 375: 1823–33.
⦁ Langer CJ, Gadgeel SM, Borghaei H, et al, for the KEYNOTE-021 investigators. Carboplatin and pemetrexed with or without pembrolizumab for advanced, non-squamous non-small-cell lung cancer: a randomised, phase 2 cohort of the open-label KEYNOTE-021 study. Lancet Oncol 2016; 17: 1497–508.
⦁ Planchard D, Min Kim T, Mazieres J, et al. Dabrafenib in patients with BRAFV600E-positive advanced non-small-cell lung cancer:
a single arm, multicentre, open-label phase 2 trial.
Lancet Oncol 2016; 17: 642–50.
⦁ Planchard D, Besse B, Groen HJM, et al. Dabrafenib plus trametinib in patients with previously treated BRAFV600E-mutant metastatic non-small cell lung cancer: an open label, multicentre phase 2 trial. Lancet Oncol 2016; 17: 984–93.
⦁ Planchard D, Besse B, Kim TM, et al. Updated survival of patients (pts) with previously treated BRAF V600E–mutant advanced non- small cell lung cancer (NSCLC) who received dabrafenib (D) or D + trametinib (T) in the phase II BRF113928 study.
Proc Am Soc Clin Oncol 2017; 35 (suppl): abstr 9075.
⦁ Long GV, Stroyakovskiy D, Gogas H, et al. Dabrafenib and trametinib versus dabrafenib and placebo for Val600 BRAF-mutant melanoma: a multicentre, double-blind, phase 3 randomised controlled trial. Lancet 2015; 386: 444–51.
⦁ Robert C, Karaszewska B, Schachter J, et al. Improved overall survival in melanoma with combined dabrafenib and trametinib. N Engl J Med 2015; 372: 30–39.
⦁ Brookmeyer R, Crowley J. A confidence interval for the mean survival time. Biometrics 1982; 38: 29–41.
⦁ Noeparast A, Teugels E, Giron P, et al. Non-V600 BRAF mutations recurrently found in lung cancer predict sensitivity to the combination of trametinib and dabrafenib. Oncotarget 2016; published online Aug 26. DOI:10.18632/oncotarget.11635.

⦁ Maemondo M, Inoue A, Kobayashi K, et al. Gefitinib or chemotherapy for non-small-cell lung cancer with mutated EGFR. N Engl J Med 2010; 362: 2380–88.
⦁ Mok T, Cheng Y, Zhou X, et al. Dacomitinib versus gefitinib for the first-line treatment of advanced EGFR mutation positive non-small cell lung cancer (ARCHER 1050): a randomized, open-label
phase III trial. Proc Am Soc Clin Oncol 2017;
35 (suppl): abstr LBA9007.
⦁ Solomon BJ, Mok T, Kim DW, et al. First-line crizotinib versus chemotherapy in ALK-positive lung cancer. N Engl J Med 2014; 371: 2167–77.
⦁ Peters S, Camidge DR, Shaw AT, et al. Alectinib versus crizotinib in untreated ALK-positive non-small-cell lung cancer. N Engl J Med 2017; published online June 6. DOI:10.1056/NEJMoa1704795.
⦁ Soria JC, Tan DS, Chiari R, et al. First-line ceritinib versus platinum-based chemotherapy in advanced ALK-rearranged
non-small-cell lung cancer (ASCEND-4): a randomised, open-label, phase 3 study. Lancet 2017; 389: 917–29.
⦁ Venur VA, Ahluwalia MS. Targeted therapy in brain metastases: ready for primetime? Am Soc Clin Oncol Educ Book 2016;
35: e123–30.
⦁ Davies MA, Saiag P, Robert C, et al. Dabrafenib plus trametinib in patients with BRAFV600-mutant melanoma brain metastases (COMBI-MB): a multicentre, multicohort, open-label, phase 2 trial. Lancet Oncol 2017; 18: 863–73.
⦁ Hellmann MD, Rizvi NA, Goldman JW, et al. Nivolumab plus ipilimumab as first-line treatment for advanced non-small-cell lung cancer (CheckMate 012): results of an open-label, phase 1, multicohort study. Lancet Oncol 2017; 18: 31–41.
⦁ Sabari JK, Montecalvo J, Chen R, et al. PD-L1 expression and response to immunotherapy in patients with MET exon 14-alterated non-small cell lung cancers (NSCLC). Proc Am Soc Clin Oncol 2017; 35 (suppl): abstr 8512.
⦁ Gainor JF, Shaw AT, Sequist LV, et al. EGFR mutations and ALK rearrangements are associated with low response rates to
PD-1 pathway blockade in non-small cell lung cancer:
a retrospective analysis. Clin Cancer Res 2016; 22: 4585–93.
⦁ Spagnolo F, Ghiorzo P, Orgiano L, et al. BRAF-mutant melanoma: treatment approaches, resistance mechanisms, and diagnostic strategies. Onco Targets Ther 2015; 8: 157–68.
⦁ Lin L, Asthana S, Chan E, et al. Mapping the molecular determinants of BRAF oncogene dependence in human lung cancer. Proc Natl Acad Sci USA 2014; 111: E748–57.
⦁ Sen B, Peng S, Tang X, et al. Kinase-impaired BRAF mutations in lung cancer confer sensitivity to dasatinib. Sci Transl Med 2012;GSK2118436