BrigaVSCrizo ALK+

The n e w e ng l a n d j o u r na l of m e dic i n e
Original Article
Brigatinib versus Crizotinib in ALK-Positive

Non–Small-Cell Lung Cancer
D.R. Camidge, H.R. Kim, M.-J. Ahn, J.C.-H. Yang, J.-Y. Han, J.-S. Lee,
M.J. Hochmair, J.Y.-C. Li, G.-C. Chang, K.H. Lee, C. Gridelli, A. Delmonte,
R. Garcia Campelo, D.-W. Kim, A. Bearz, F. Griesinger, A. Morabito, E. Felip,
R. Califano, S. Ghosh, A. Spira, S.N. Gettinger, M. Tiseo, N. Gupta,
J. Haney, D. Kerstein, and S. Popat
A BS T R AC T
BACKGROUND
Brigatinib, a next-generation anaplastic lymphoma kinase (ALK) inhibitor, has The authors’ full names, academic de-
robust efficacy in patients with ALK-positive non–small-cell lung cancer (NSCLC) grees, and affiliations are listed in the
Appendix. Address reprint requests to Dr.
that is refractory to crizotinib. The efficacy of brigatinib, as compared with crizo- Camidge at the University of Colorado
tinib, in patients with advanced ALK-positive NSCLC who have not previously re- Cancer Center, Anschutz Cancer Pavilion,
ceived an ALK inhibitor is unclear. Rm. 5327, 1665 North Aurora Ct., Aurora,
CO 80045, or at ross.camidge@ucdenver
.edu.
METHODS
In an open-label, phase 3 trial, we randomly assigned, in a 1:1 ratio, patients with This article was published on September
25, 2018, at NEJM.org.
advanced ALK-positive NSCLC who had not previously received ALK inhibitors to
receive brigatinib at a dose of 180 mg once daily (with a 7-day lead-in period at 90 N Engl J Med 2018;379:2027-39.
DOI: 10.1056/NEJMoa1810171
mg) or crizotinib at a dose of 250 mg twice daily. The primary end point was Copyright © 2018 Massachusetts Medical Society.
progression-free survival as assessed by blinded independent central review. Sec-
ondary end points included the objective response rate and intracranial response.
The first interim analysis was planned when approximately 50% of 198 expected
events of disease progression or death had occurred.
RESULTS
A total of 275 patients underwent randomization; 137 were assigned to brigatinib
and 138 to crizotinib. At the first interim analysis (99 events), the median follow-
up was 11.0 months in the brigatinib group and 9.3 months in the crizotinib
group. The rate of progression-free survival was higher with brigatinib than with
crizotinib (estimated 12-month progression-free survival, 67% [95% confidence
interval {CI}, 56 to 75] vs. 43% [95% CI, 32 to 53]; hazard ratio for disease progres-
sion or death, 0.49 [95% CI, 0.33 to 0.74]; P<0.001 by the log-rank test). The con-
firmed objective response rate was 71% (95% CI, 62 to 78) with brigatinib and 60%
(95% CI, 51 to 68) with crizotinib; the confirmed rate of intracranial response
among patients with measurable lesions was 78% (95% CI, 52 to 94) and 29%
(95% CI, 11 to 52), respectively. No new safety concerns were noted.
CONCLUSIONS
Among patients with ALK-positive NSCLC who had not previously received an ALK
inhibitor, progression-free survival was significantly longer among patients who
received brigatinib than among those who received crizotinib. (Funded by Ariad
Pharmaceuticals; ALTA-1L ClinicalTrials.gov number, NCT02737501.)
n engl j med 379;21 nejm.org November 22, 2018 2027

The New England Journal of Medicine is produced by NEJM Group, a division of the Massachusetts Medical Society.
Downloaded from nejm.org on October 26, 2024. For personal use only.
No other uses without permission. Copyright © 2018 Massachusetts Medical Society. All rights reserved.
R
earrangements of the oncogenic ALK inhibitor. Here we report the results of the
anaplastic lymphoma kinase (ALK) gene first prespecified interim analysis.
occur in 3 to 5% of patients with non–
small-cell lung cancer (NSCLC).1-3 A previous Me thods
phase 3 trial showed that progression-free sur-
vival was significantly longer among patients who Patients
received crizotinib, a first-generation ALK inhibi- Eligible patients were at least 18 years of age,
tor, than among those who received platinum- had locally advanced or metastatic NSCLC with
based, double-agent chemotherapy (median pro- at least one measurable lesion according to the
gression-free survival, 10.9 months vs. 7.0 months).4 Response Evaluation Criteria in Solid Tumors
Disease progression in the central nervous sys- (RECIST), version 1.1,25 and had not previously
tem (CNS) in patients receiving crizotinib is received ALK-targeted therapy. Patients with
common, probably because of its poor brain asymptomatic, untreated CNS metastases were
penetration.5-7 Progression beyond the CNS in not excluded. Patients were eligible for trial entry
patients receiving crizotinib and other ALK in- on the basis of locally determined ALK testing.
hibitors can occur through the emergence of Patients were excluded if they had previously re-
ALK mutations, which are detectable in 20% of ceived more than one systemic anticancer ther-
patients who have received crizotinib and in 56% apy regimen for advanced disease or chemother-
of patients who have received next-generation apy or radiation therapy (other than stereotactic
inhibitors. Progression also occurs in patients radiosurgery or stereotactic body radiation ther-
through other mechanisms that are not related apy) within 14 days before the first dose of the
to ALK mutations.8-10 trial drug. Complete inclusion and exclusion
Brigatinib (Ariad Pharmaceuticals) is a next- criteria are provided in the trial protocol, avail-
generation ALK inhibitor that targets a broad able with the full text of this article at NEJM.org.
range of ALK mutations and ROS1 rearrange-
ments. It is also the only ALK inhibitor with Trial Oversight
activity in cell lines with mutations in the gene This trial was conducted in accordance with the
encoding epidermal growth factor receptor ethical standards of the Declaration of Helsinki
(EGFR).11-14 Step-up dosing of brigatinib over a and the International Council for Harmonisation
period of 7 days is used to abrogate the risk of guidelines for Good Clinical Practice. All patients
uncommon early-onset pulmonary events.13 In provided written informed consent. The protocol
the ALK in Lung Cancer Trial of AP26113 (ALTA) and informed-consent documents were approved
involving 222 patients with disease that was re- by the local institutional review board or ethics
fractory to crizotinib, brigatinib administered at committee at each site. The trial was designed by
the recommended dosing regimen of 180 mg the sponsor, Ariad Pharmaceuticals, in collabora-
once daily (with a 7-day lead-in period at 90 mg tion with the first author. Data were collected
in 110 patients) was associated with high sys- and trial procedures were overseen by the trial
temic and CNS response rates and a median investigators (listed in the Supplementary Ap-
progression-free survival of 16.7 months.15-18 The pendix, available at NEJM.org). The sponsor
same regimen was associated with similar pro- analyzed the data, and all the authors had full
gression-free survival (16.3 months) among pa- access to the data and participated in the inter-
tients who had received crizotinib in the phase pretation of the data. The manuscript was writ-
1–2 trial.13,15,19 These median rates of progres- ten by the authors with medical writing assis-
sion-free survival are higher than those associ- tance paid for by the sponsor. All the authors
ated with other next-generation ALK inhibitors vouch for the completeness and accuracy of the
(including alectinib, ceritinib, ensartinib, and data reported and for the adherence of the trial
lorlatinib) among such patients.17,19-24 to the protocol and statistical analysis plan.
The ALK in Lung Cancer Trial of Brigatinib in
1st Line (ALTA-1L) is a phase 3 trial comparing Trial Design
the efficacy and safety of brigatinib with those ALTA-1L is an open-label, multicenter, random-
of crizotinib in patients with ALK-positive NSCLC ized, international, phase 3 trial conducted at 124
who had not received previous treatment with an centers in 20 countries. Patients were stratified
2028 n engl j med 379;21 nejm.org November 22, 2018
Brigatinib or Crizotinib in Lung Cancer
according to baseline brain metastases (present all 198 expected events (disease progression or
or absent) and completion of at least one full cycle death) were observed. A Lan–DeMets alpha spend-
of previous chemotherapy for locally advanced or ing function with O’Brien–Fleming boundaries26
metastatic disease (yes or no). They were ran- was used to control the overall alpha level at 0.05
domly assigned (in a 1:1 ratio) to receive oral (two-sided). Assuming an estimated median
brigatinib at a dose of 180 mg once daily after a progression-free survival of 10 months among
7-day lead-in period of 90 mg once daily or oral patients who received crizotinib in sample-size
crizotinib at a dose of 250 mg twice daily. calculations,4 a total of 198 events (disease pro-
Patients continued treatment until they had gression or death among approximately 270 pa-
progressive disease as assessed by an indepen- tients who underwent randomization) was re-
dent review committee whose members were quired for the trial to have approximately 90%
unaware of the trial-drug assignments, had un- power at the final analysis of the primary end
acceptable toxic effects, or had another discon- point to detect a 6-month increase in progression-
tinuation criterion. In the crizotinib group, cross- free survival (hazard ratio for disease progres-
over to brigatinib was permitted after progression sion or death, 0.625).
assessed by means of blinded independent re- For the first interim analysis, the primary end
view (with a 10-day washout period from crizo- point of progression-free survival was tested at
tinib). At the investigator’s discretion, treatment a two-sided alpha level of 0.0031. Efficacy was
in the brigatinib group could be continued after evaluated in the intention-to treat population.
disease progression. Dose interruptions or reduc- Patients who received at least one dose of a trial
tions were permitted for treatment-related adverse drug constituted the safety population. For time-
events. Adverse events were categorized with the to-event efficacy analyses, median values and
use of the National Cancer Institute Common Ter- two-sided 95% confidence intervals were esti-
minology Criteria for Adverse Events, version 4.0. mated with the use of Kaplan–Meier methods.
Disease assessment (according to RECIST, The primary end point was compared between
version 1.1) included imaging of the chest and the brigatinib and crizotinib groups with the use
abdomen with the use of computed tomography of a two-sided log-rank test stratified according
or magnetic resonance imaging (MRI) with con- to the presence or absence of baseline brain
trast material and imaging of the head with the metastases and the use or nonuse of previous
use of MRI with contrast material. Assessment chemotherapy in patients with locally advanced
was performed at screening, every 8 weeks or metastatic disease.
through cycle 14 (28 days per cycle), and then Efficacy and safety data are reported as of
every 12 weeks until the end of treatment. Two February 19, 2018. Statistical analyses were per-
independent review committees whose members formed with the use of Base 9.4 SAS/STAT soft-
were unaware of the trial-drug assignments per- ware, version 13.1. Statistical methods are de-
formed disease assessments: one for all disease scribed further in the statistical analysis plan
according to RECIST, version 1.1,25 and one ex- and in the Supplementary Methods section in
clusively for the evaluation of intracranial CNS the Supplementary Appendix.
end points. Responses were confirmed at least
4 weeks after the initial response. R e sult s
Outcomes Patients
The primary end point was progression-free sur-Between April 2016 and August 2017, a total of
vival as assessed by blinded independent review275 patients were enrolled; 137 patients were
according to RECIST, version 1.1. Secondary endrandomly assigned to brigatinib and 138 were
points included the objective response rate andrandomly assigned to crizotinib (Fig. 1). Two
intracranial response. A complete list of second-
patients (1 per group) did not receive treatment
ary end points is provided in the trial protocol.
but were included in the intention-to-treat analy-
ses. Baseline factors, including sex, Eastern Coop-
Statistical Analysis erative Oncology Group performance-status score,
Interim analyses were planned after approxi- use of a Food and Drug Administration–approved
mately 50% (99 events) and 75% (149 events) of ALK diagnostic test, and the use of previous

311 Patients were assessed for eligibility
36 Were not eligible

4 Were not willing to comply with trial
procedures
3 Had insufficient tumor tissue available
3 Had symptomatic CNS metastases
at screening or asymptomatic disease
leading to an increased dose
of glucocorticoids within 7 days after
enrollment
26 Did not meet other eligibility criteria
275 Underwent randomization
137 Were assigned to receive brigatinib, 180 mg 138 Were assigned to receive crizotinib, 250 mg
once daily with a 7-day lead-in at 90 mg twice daily
136 Received assigned treatment 137 Received assigned treatment
1 Did not receive assigned treatment 1 Did not receive assigned treatment
41 Discontinued treatment 78 Discontinued treatment

22 Had disease progression 61 Had disease progression
10 Had adverse event 5 Had adverse event
5 Died 4 Died
2 Were withdrawn from trial by physician 5 Were withdrawn from trial by physician
2 Withdrew consent 2 Withdrew consent
1 Had other reason
35 Crossed over to brigatinib
Patient status at date of data cutoff: Patient status at date of data cutoff:
95 Continued to receive brigatinib 59 Continued to receive crizotinib
31 Were being followed for survival after 33 Were being followed for survival after
discontinuation of trial treatment discontinuation of trial treatment
10 Were alive on date of follow-up 20 Were alive on date of follow-up
10 Died 10 Died
10 Were not followed for survival after 3 Were not followed for survival after
discontinuation of trial treatment discontinuation of trial treatment
1 Withdrew consent
137 Were evaluated for primary end point 138 Were evaluated for primary end point
136 Were evaluated for safety 137 Were evaluated for safety
Figure 1. Screening, Enrollment, Randomization, and Follow-up.

Data reported as of the cutoff date for the first interim analysis (February 19, 2018) are shown. In the brigatinib
group, 18 patients had documented disease progression according to Response Evaluation Criteria in Solid Tumors
(RECIST), version 1.1, and 4 had clinical disease progression. In the crizotinib group, 54 patients had documented
disease progression according to RECIST, version 1.1, and 7 had clinical disease progression. Crossover from crizo-
tinib to brigatinib was permitted for patients who had objective progression as assessed by blinded independent
review. Patients who discontinued crizotinib for other reasons (e.g., progression according to investigator assess-
ments) and then began to receive brigatinib were not included in the number of crossover patients. CNS denotes
central nervous system.
Table 1. Baseline Patient Characteristics in the Intention-to-Treat Population.*
Characteristic Brigatinib (N = 137) Crizotinib (N = 138) Total (N = 275)

Age — yr
Median 58 60 59
Range 27–86 29–89 27–89
Female sex — no. (%) 69 (50) 81 (59) 150 (55)
Race — no. (%)†
Non-Asian 78 (57) 89 (64) 167 (61)
Asian 59 (43) 49 (36) 108 (39)
ECOG performance-status score — no. (%)‡
0 or 1 131 (96) 132 (96) 263 (96)
2 6 (4) 6 (4) 12 (4)
History of tobacco use — no. (%)
Never smoked 84 (61) 75 (54) 159 (58)
Former smoker 49 (36) 56 (41) 105 (38)
Current smoker 4 (3) 7 (5) 11 (4)
Stage of disease at trial entry — no. (%)
IIIB 8 (6) 12 (9) 20 (7)
IV 129 (94) 126 (91) 255 (93)
Histologic type — no. (%)
Adenocarcinoma 126 (92) 137 (99) 263 (96)
Adenosquamous carcinoma 3 (2) 1 (1) 4 (1)
Squamous-cell carcinoma 4 (3) 0 4 (1)
Large-cell carcinoma 2 (1) 0 2 (1)
Other 2 (1) 0 2 (1)
ALK status assessed locally with the use of FDA-approved test — no. %§ 123 (90) 112 (81) 235 (85)
Brain metastases — no. (%)¶ 40 (29) 41 (30) 81 (29)
Previous radiotherapy to brain — no. (%) 18 (13) 19 (14) 37 (13)
Previous chemotherapy in patients with locally advanced 36 (26) 37 (27) 73 (27)
or metastatic disease — no. (%)‖
* Percentages may not sum to 100 because of rounding. FDA denotes Food and Drug Administration.
† Race was reported by the investigator.
‡ Eastern Cooperative Oncology Group (ECOG) performance-status scores range from 0 to 5, with higher numbers indicating increasing im-
pairment in activities of daily living.
§ ALK-positive status was confirmed locally by fluorescence in situ hybridization (Vysis) or immunohistochemical assay (Ventana).
¶ The presence of brain metastases was assessed by the investigator.
‖ Previous chemotherapy was defined as completion of at least one full cycle of chemotherapy in patients with locally advanced or metastatic
disease. Among 36 patients who received previous chemotherapy in the brigatinib group, 2 (6%) had a complete response, 9 (25%) had a
partial response, 10 (28%) had stable disease, and 8 (22%) had progressive disease; the best response to previous chemotherapy was “other
or unknown” in 7 patients (19%). Among 37 patients who received previous chemotherapy in the crizotinib group, 2 (5%) had a complete
response, 8 (22%) had a partial r esponse, 13 (35%) had stable disease, and 7 (19%) had progressive disease; the best response to previous
chemotherapy was “other or unknown” in 7 patients (19%).
chemotherapy (and best response to chemo- Appendix). Of 275 patients, 81 (29%) had brain
therapy) were well balanced between the groups metastases at baseline (29% in the brigatinib
(Table 1, and Table S1 in the Supplementary group and 30% in the crizotinib group), with

similar rates of CNS radiotherapy before en- Figure 2 (facing page). Efficacy of Brigatinib
rollment. and Crizotinib in Patients with ALK-Positive NSCLC
As of February 19, 2018, a total of 95 patients Who Had Not Previously Received an ALK Inhibitor.
(69%) in the brigatinib group and 59 patients Panel A shows Kaplan–Meier estimates of blinded in-
(43%) in the crizotinib group continued to re- dependent review–assessed progression-free survival
ceive trial treatment, with a median follow-up of among patients in the intention-to-treat population.
Of the 137 patients in the brigatinib group, 36 (26%)
11.0 months (range, 0 to 20.0) and 9.3 months had an event; of the 138 patients in the crizotinib group,
(range, 0 to 20.9), respectively. The median dura- 63 (46%) had an event. Tick marks indicate censored
tion of treatment was 9.2 months (range, 0.1 to data. NR denotes not reached. Panel B shows hazard
18.4) in the brigatinib group and 7.4 months ratios for disease progression or death across prespec-
(range, 0.1 to 19.2 months) in the crizotinib ified patient subgroups. The hazard ratio was not calcu-
lated for patients who were current smokers (4 in the
group. A total of 35 patients who discontinued brigatinib group and 7 in the crizotinib group) because
crizotinib because of disease progression crossed of insufficient patient numbers according to the statis-
over to brigatinib treatment as part of the trial tical analysis plan. Values for the Eastern Cooperative
(see the Supplementary Results section in the Oncology Group (ECOG) performance-status score
Supplementary Appendix). are on a 5-point scale, with higher numbers reflecting
greater disability. The hazard ratio was not calculated
for patients who had an ECOG performance-status
Efficacy score of 2 (6 in the brigatinib group and 6 in the crizo-
At the first interim data cutoff, a total of 99 tinib group) because of insufficient patient numbers
events for the primary end point (disease pro- according to the statistical analysis plan. The presence
gression or death) had occurred in the intention- of brain metastases at baseline was assessed by the
investigator. The “previous chemotherapy” category
to-treat population (36 of 137 patients [26%] in is previous chemotherapy in patients with locally ad-
the brigatinib group and 63 of 138 patients vanced or metastatic disease. Panel C shows the best
[46%] in the crizotinib group). For blinded inde- percentage change from baseline in the sum of the
pendent review–assessed progression-free sur- longest diameters of target lesions in patients who
vival, brigatinib met the prespecified threshold underwent follow-up imaging and could be evaluated
for a response (121 patients in the brigatinib group
for statistical superiority over crizotinib (esti- and 122 patients in the crizotinib group). Assessments
mated 12-month progression-free survival, 67% were based on blinded independent review. All trial as-
[95% confidence interval {CI}, 56 to 75] in the sessments were used in these calculations. The solid
brigatinib group and 43% [95% CI, 32 to 53] in line at −30% indicates the threshold for partial response
the crizotinib group; hazard ratio for progres- according to RECIST, version 1.1. Panel D shows survival
without intracranial disease progression among patients
sion or death, 0.49 [95% CI, 0.33 to 0.74]; with brain metastases at baseline. Of the 43 patients in
P<0.001 by the log-rank test) (Fig. 2A). Accord- the brigatinib group, 11 (26%) had an event; of the 47
ing to investigator assessment, progression-free patients in the crizotinib group, 28 (60%) had an event.
survival was also longer among patients who re- Tick marks indicate censored data.
ceived brigatinib (estimated 12-month progres-
sion-free survival, 69% [95% CI, 59 to 76] in the group. Changes from baseline in target lesions
brigatinib group and 40% [95% CI, 30 to 50] in are shown in Figure 2C. The estimated rate of
the crizotinib group; hazard ratio for disease the 12-month duration of response in patients
progression or death, 0.45 [95% CI, 0.30 to with a confirmed response was 78% (95% CI, 67
0.68]). Efficacy consistently favored brigatinib to 86) in the brigatinib group and 48% (95% CI,
across subgroups (Fig. 2B, and Fig. S1 in the 31 to 63) in the crizotinib group (Fig. S2 in the
Supplementary Appendix). Supplementary Appendix).
The confirmed objective response rate assessed Of 275 patients, 90 had brain metastases at
by blinded independent review was 71% (95% CI, baseline, as assessed by blinded independent
62 to 78) in the brigatinib group and 60% (95% review, and 39 had measurable brain metastases
CI, 51 to 68) in the crizotinib group (Table 2). (≥10 mm in diameter). The confirmed rate of
The overall objective response rate (objective re- intracranial objective response among patients
sponse at one or more assessments, including with measurable baseline brain metastases was
confirmed and unconfirmed responses) was 78% (95% CI, 52 to 94) (14 of 18 patients) with
76% (95% CI, 68 to 83) in the brigatinib group brigatinib and 29% (95% CI, 11 to 52) (6 of 21
and 73% (95% CI, 65 to 80) in the crizotinib patients) with crizotinib (Table 2), and the over-
A Progression-free Survival B Subgroup Analysis
Median At 1 Yr Subgroup No. of Patients Hazard Ratio for Disease Progression or Death (95% CI)
(95% CI) (95% CI) Brigatinib Crizotinib
mo % Overall 137 138 0.49 (0.33–0.74)
Brigatinib NR 67 (56–75) Age
Crizotinib 9.8 (9.0–12.9) 43 (32–53) 18–64 yr 93 95 0.44 (0.26–0.74)
≥65 yr 44 43 0.59 (0.30–1.18)
100 Sex
90 Female 69 81 0.44 (0.24–0.84)
Male 68 57 0.49 (0.28–0.85)
80 Race
70 Brigatinib
Non-Asian 78 89 0.54 (0.33–0.90)
60 Asian 59 49 0.41 (0.20–0.86)
History of tobacco use
50
Crizotinib Never smoked 84 75 0.47 (0.27–0.84)
40 Former smoker 49 56 0.51 (0.27–0.97)
(% of patients)
30 ECOG performance-status score
Hazard ratio for disease progression or death, 0 58 60 0.19 (0.06–0.55)
20 0.49 (95% CI, 0.33–0.74) 1 73 72 0.60 (0.37–0.98)
Progression-free Survival
10 P<0.001 by log-rank test Brain metastases at baseline
0 Yes 40 41 0.20 (0.09–0.46)
0 3 6 9 12 15 18 21 No 97 97 0.72 (0.44–1.18)
n engl j med 379;21

Previous chemotherapy
Months Yes 36 37 0.35 (0.14–0.85)
No 101 101 0.55 (0.34–0.88)
No. at Risk
Brigatinib 137 114 90 64 26 3 1 0.0 0.5 1.0 1.5 2.0
nejm.org
Crizotinib 138 117 75 50 18 3 2 Brigatinib Better Crizotinib Better
C Best Target-Lesion Responses D Survival without Intracranial Disease Progression among Patients
with Brain Metastases at Baseline
Progressive Stable disease Confirmed partial Confirmed complete
disease response response
Median At 1 Yr
(95% CI) (95% CI)
November 22, 2018

Brigatinib Crizotinib mo %
Brigatinib NR (11–NR) 67 (47–80)

60 60 Crizotinib 5.6 (4.1–9.2) 21 (6–42)
100
40 40 90
80
20 20 70 Brigatinib
60
0 0 50
40
−20 −20
(% of patients)
30
Hazard ratio for disease
20 Crizotinib
progression or death,
Progression-free Survival
−40 −40 10 0.27 (95% CI, 0.13–0.54)
0
−60 −60 0 3 6 9 12 15 18 21
Months
−80 −80
No. at Risk
Best Change from Baseline in Target Lesions (%)

−100 −100 Brigatinib 43 39 32 22 9 5 1
Crizotinib 47 37 16 9 2 0 0
2033
Table 2. Rates of Systemic and Intracranial Objective Response, Assessed by Blinded Independent Review.*
Variable Brigatinib Crizotinib Odds Ratio (95% CI)

Intention-to-treat population
No. of patients 137 138
Confirmed objective response
% (95% CI) 71 (62–78) 60 (51–68) 1.59 (0.96–2.62)
Complete response — no. (%) 5 (4) 7 (5)
Partial response — no. (%) 92 (67) 76 (55)
Rate of objective response at ≥1 assessment — % (95% CI) 76 (68–83) 73 (65–80) 1.13 (0.66–1.97)
Complete response — no. (%) 9 (7) 11 (8)
Partial response — no. (%) 95 (69) 90 (65)
Median duration (95% CI) of confirmed response — mo NR (NR–NR) 11.1 (9.2–NR)
12-mo rate (95% CI) of maintaining response 75 (63–83) 41 (26–54)
Patients with measurable brain metastases at baseline, as assessed
by blinded independent review†
Confirmed intracranial objective response
% (95% CI) 78 (52–94) 29 (11–52) 10.42 (1.90–57.05)
Intracranial complete response — no. (%) 2 (11) 0
Intracranial partial response — no. (%) 12 (67) 6 (29)
Rate of intracranial objective response at ≥1 assessment — % 83 (59–96) 33 (15–57) 9.29 (1.88–45.85)
(95% CI)
Intracranial complete response — no. (%) 2 (11) 0
Patients with any brain metastases at baseline, as assessed by
blinded independent review
Confirmed intracranial objective response
% (95% CI) 67 (51–81) 17 (8–31) 13.00 (4.38–38.61)
Intracranial complete response — no. (%) 16 (37) 2 (4)
Rate of intracranial objective response at ≥1 assessment — % (95% CI) 79 (64–90) 23 (12–38) 16.30 (5.32–49.92)
Intracranial complete response — no. (%) 19 (44) 4 (9)
* NR denotes not reached.

† Measurable brain metastases were at least 10 mm in diameter.
all rate of intracranial objective response (objec- disease progression as the first site of disease
tive response at one or more assessments, includ- progression, alone or with concurrent systemic
ing confirmed and unconfirmed responses) in progression. In patients without brain metastases
this population was 83% (95% CI, 59 to 96) with at baseline, 1% (1 of 94 patients) in the brigatinib
brigatinib and 33% (95% CI, 15 to 57) with group and 5% (5 of 91 patients) in the crizotinib
crizotinib. Overall, 9% (12 of 137 patients) in group had intracranial disease progression as
the brigatinib group and 19% (26 of 138 pa- the first site of disease progression.
tients) in the crizotinib group had intracranial The estimated rate of 12-month survival with-
out intracranial disease progression among pa- 15%), peripheral edema (39% vs. 4%), vomiting
tients with baseline brain metastases was 67% (39% vs. 18%), an increased alanine aminotrans-
(95% CI, 47 to 80) in the brigatinib group and ferase level (32% vs. 19%), decreased appetite
21% (95% CI, 6 to 42) in the crizotinib group; (20% vs. 7%), photopsia (20% vs. 1%), dysgeusia
the estimated rate of 12-month survival without (19% vs. 4%), and visual impairment (16% vs.
intracranial disease progression in the intention- 0%). Grade 3 to 5 adverse events occurred in 61%
to-treat population was 78% (95% CI, 68 to 85) of patients in the brigatinib group and in 55% of
in the brigatinib group and 61% (95% CI, 50 to patients in the crizotinib group. No cases of
71) in the crizotinib group. The rate of survival pancreatitis were reported. Symptoms possibly
without intracranial disease progression among related to increased creatine kinase levels (myal-
patients with baseline brain metastases was gia and muscle pain) did not differ substantially
higher in the brigatinib group than in the crizo- between the groups, nor were they apparently re-
tinib group (hazard ratio for intracranial disease lated to the grade of increased creatine kinase
progression or death, 0.27; 95% CI, 0.13 to 0.54) levels (Table 3).
(Fig. 2D), and the rate of survival without intra- Fourteen patients had adverse events leading
cranial disease progression among patients in to death within 30 days after the last dose of
the intention-to-treat population was higher in the trial drug (7 [5%] in the brigatinib group and
the brigatinib group than in the crizotinib group 7 [5%] in the crizotinib group); none of the
(hazard ratio, 0.42; 95% CI, 0.24 to 0.70). An events were deemed by the investigators to be
exploratory competing-risks analysis of intracra- related to trial treatment. Interstitial lung dis-
nial disease progression, systemic progression, ease or pneumonitis at any time occurred in 4%
and death in the intention-to-treat population (5 of 136) of patients in the brigatinib group and
showed that the cause-specific hazard ratio for 2% (3 of 137) of patients in the crizotinib group.
time to progression of intracranial disease was Grade 3 or 4 interstitial lung disease or pneumo-
0.30 (95% CI, 0.15 to 0.60) (Fig. S3 in the Sup- nitis occurred in 3% (4 of 136) and 0.7% (1 of
plementary Appendix). 137), respectively. Interstitial lung disease or pneu-
At data cutoff, 34 patients in the intention-to- monitis of any grade with early onset (defined as
treat population had died (17 patients in the occurring within 14 days after the initiation of
brigatinib group [12%] and 17 patients in the treatment) was observed in 4 of 136 patients
crizotinib group [12%]). The 1-year rate of over- (3%) in the brigatinib group (onset on days 3 to
all survival was 85% (95% CI, 76 to 91) with 8) and was not observed in patients who received
brigatinib and 86% (95% CI, 77 to 91) with crizotinib. All 4 patients discontinued brigatinib
crizotinib (Fig. S4 in the Supplementary Appen- after the pulmonary event, according to the pro-
dix). The median overall survival was not reached tocol. Among patients who crossed over from
in either group. crizotinib to brigatinib, the rate of interstitial
lung disease or pneumonitis of any grade was
Safety 3% (1 of 35 patients), and it occurred on day 3.
The most common (>25% of patients overall) An investigator- or protocol-mandated dose
adverse events of any grade that occurred during reduction for any adverse events occurred in
treatment were gastrointestinal symptoms, in- 29% of treated patients in the brigatinib group
creased blood creatine kinase levels, and in- and 21% of treated patients in the crizotinib
creased alanine aminotransferase levels (Table 3, group. The most common adverse events leading
and Table S2 in the Supplementary Appendix). to dose reduction are shown in Table S3 in the
Adverse events that occurred at a higher incidence Supplementary Appendix. A total of 12% of pa-
by more than 5 percentage points with brigatinib tients who received brigatinib and 9% of patients
than with crizotinib included an increased cre- who received crizotinib discontinued treatment
atine kinase level (brigatinib [39%] vs. crizotinib owing to adverse events.
[15%]), cough (25% vs. 16%), hypertension (23%
vs. 7%), and an increased lipase level (19% vs. Discussion
12%). Adverse events that were more common
with crizotinib than with brigatinib included In ALTA-1L, brigatinib, as compared with crizo-
nausea (crizotinib [56%] vs. brigatinib [26%]), tinib, had superior efficacy against systemic and
diarrhea (55% vs. 49%), constipation (42% vs. intracranial disease. At the first interim analysis,

Table 3. Adverse Events of Any Grade That Differed by More Than 5 Percentage Points in Frequency between Groups.
Event Brigatinib (N = 136) Crizotinib (N = 137)
Any Grade Grade ≥3 Any Grade Grade ≥3
number of patients (percent)

Any adverse event 132 (97) 83 (61) 137 (100) 76 (55)
Diarrhea 67 (49) 2 (1) 75 (55) 3 (2)
Increased blood creatine kinase level* 53 (39) 22 (16) 21 (15) 2 (1)
Nausea 36 (26) 2 (1) 77 (56) 4 (3)
Cough 34 (25) 0 22 (16) 0
Hypertension 31 (23) 13 (10) 10 (7) 4 (3)
Increased alanine aminotransferase level 26 (19) 2 (1) 44 (32) 13 (9)
Increased lipase level† 26 (19) 18 (13) 16 (12) 7 (5)
Vomiting 25 (18) 1 (1) 54 (39) 3 (2)
Constipation 20 (15) 0 57 (42) 1 (1)
Increased amylase level† 19 (14) 7 (5) 9 (7) 1 (1)
Pruritus 18 (13) 1 (1) 6 (4) 1 (1)
Rash 14 (10) 0 3 (2) 0
Decreased appetite 10 (7) 1 (1) 27 (20) 4 (3)
Dermatitis acneiform 9 (7) 0 2 (1) 0
Dyspepsia 8 (6) 0 18 (13) 0
Epistaxis 8 (6) 0 0 0
Bradycardia 7 (5) 1 (1) 17 (12) 0
Peripheral edema 6 (4) 1 (1) 53 (39) 1 (1)
Dysgeusia 6 (4) 0 26 (19) 0
Upper abdominal pain 6 (4) 1 (1) 18 (13) 2 (1)
Pain in extremity 6 (4) 0 17 (12) 1 (1)
Increased blood creatinine level 3 (2) 0 19 (14) 1 (1)
Neutropenia 2 (1) 0 12 (9) 6 (4)
Pleural effusion 2 (1) 1 (1) 9 (7) 2 (1)
Photopsia 1 (1) 0 28 (20) 1 (1)
Gastroesophageal reflux disease 1 (1) 0 12 (9) 0
Visual impairment 0 0 22 (16) 0
Deep-vein thrombosis 0 0 8 (6) 0
* Myalgia was reported in 6% of patients in the brigatinib group and 4% of patients in the crizotinib group; musculoskeletal pain was report-
ed in 4% and 6% of the patients, respectively. No myalgia or musculoskeletal pain of grade 3 or greater was reported in either group.
† No clinical cases of pancreatitis were reported in either group.
the prespecified threshold for significance for vival at 12 months was 67% in the brigatinib
the primary end point of blinded independent group (median not reached) and 43% in the crizo-
review–assessed progression-free survival with tinib group (median, 9.8 months, consistent with
brigatinib was met. With a median follow-up results seen in other randomized trials).4,27
of 11.0 months in the brigatinib group and 9.3 The hazard ratio for disease progression or
months in the crizotinib group, brigatinib was death was 0.55 (95% CI, 0.34 to 0.88) among
associated with a 51% lower risk of disease pro- patients who had not received chemotherapy and
gression or death than crizotinib (hazard ratio, 0.35 (95% CI, 0.14 to 0.85) among those who
0.49; P<0.001). The rate of progression-free sur- had received previous chemotherapy. One of the
lowest hazard ratios for disease progression or sion-free survival of 34.8 months among pa-
death was noted among patients with baseline tients in the alectinib group (hazard ratio, 0.43;
brain metastases (0.20; 95% CI, 0.09 to 0.46). 95% CI, 0.32 to 0.58); this shows that efficacy in
Although the hazard ratio for disease progres- these patients can improve over time because of
sion or death did not reach significance among the greater emergence of plateaus in the Kaplan–
patients without baseline brain metastases (0.72; Meier curve in the experimental group than in
95% CI, 0.44 to 1.18), this interim analysis may the control group.29
emphasize differences in early progression. Be- The safety profiles of brigatinib and crizotinib
cause of the known poor efficacy of crizotinib in were consistent with those in previous stud-
the CNS, CNS progression events may tend to ies.4,16,27,31 Elevated creatine kinase levels were not
have an earlier onset than other events; there- associated with the frequency or severity of my-
fore, differences in early progression-free sur- algia or musculoskeletal pain, and there were no
vival will be most apparent among patients with cases of clinical pancreatitis. The rate of dose
baseline brain disease. reduction because of any adverse events was 29%
Randomized, phase 3 trial data show that in the brigatinib group and 21% in the crizotinib
treatment options for patients with advanced group, in part reflecting more stringent protocol-
NSCLC who have not previously received ALK mandated dose modifications for laboratory ab-
inhibitors include crizotinib, alectinib, and ceri- normalities with brigatinib as compared with
tinib.4,27,28 Despite shorter follow-up, these initial crizotinib modifications, which followed stan-
results for brigatinib as compared with crizo- dard labeling (Table S3 in the Supplementary
tinib in patients who had not previously received Appendix, and the trial protocol). Early-onset
ALK inhibitors appear similar to results from the pulmonary events (interstitial lung disease and
phase 3 BO28984 (ALEX) trial, which compared pneumonitis) were observed with brigatinib but
alectinib, a second-generation ALK inhibitor, with not crizotinib. The rate of these events with
crizotinib. Both the rates of progression-free brigatinib among patients who had not previ-
survival and overall response were similar in the ously received ALK inhibitors (3%) appears to be
crizotinib groups in both trials.27 Previous che- lower than among patients with disease that was
motherapy was not permitted in the ALEX trial refractory to crizotinib (6% in ALTA),16 despite
as it was in ALTA-1L, and ALK status was cen- similar drug exposures (Fig. S5 in the Supple-
trally confirmed by Ventana immunohistochem- mentary Appendix). Consistent with a multivari-
ical analysis.27 Crossover from crizotinib at pro- ate analysis indicating that a longer washout
gression was not permitted in the ALEX trial27 as period (≥7 days) from crizotinib reduced the risk
it was in ALTA-1L. of these events,16 the rate of early-onset pulmo-
The primary end point of the ALEX trial was nary events among patients who crossed over
investigator-assessed progression-free survival. from crizotinib to brigatinib (3%), which required
Analysis by an independent review committee a 10-day minimum washout period, appeared to
was conducted only at the primary analysis time be lower than the rate observed in ALTA.
point.27,29 Investigator-assessed progression-free A key strength of this open-label trial is that
survival with a median follow-up of 17.6 to 18.6 progression-free survival was assessed by a blind-
months was associated with 12-month event-free ed independent review committee, minimizing
survival rates of 68.4% (95% CI, 61.0 to 75.9) in the potential for bias associated with investigator
the alectinib group and 48.7% (95% CI, 40.4 to assessments. In addition, ALK positivity defined
56.9) in the crizotinib group (hazard ratio, 0.47; with the use of real-world assays was incorpo-
95% CI, 0.34 to 0.65; P<0.001), with no median rated, potentially increasing the applicability of
reached in the alectinib group. The analysis of these data, and patients could have received previ-
the independent review committee showed me- ous chemotherapy. A limitation of this analysis
dian progression-free survival of 25.7 months is that overall survival data will be confounded
(95% CI, 19.9 to could not be estimated) in the by crossover of patients in the crizotinib group
alectinib group and 10.4 months (95% CI, 7.7 to to brigatinib during the trial and subsequent use
14.6) in the crizotinib group (hazard ratio, 0.53; of other tyrosine kinase inhibitors after discon-
95% CI, 0.38 to 0.73; P<0.001).27,30 A post hoc tinuation of the trial by patients from either
analysis with an additional 10 months of follow- group. With further follow-up, data in both
up showed investigator-assessed median progres- groups will mature and help to better contextu-

alize the role of brigatinib as compared with a speakers bureau and advisory fees from AstraZeneca, Pfizer,
Eli Lilly, Bristol-Myers Squibb, Novartis, Roche, Merck Sharp &
other next-generation ALK inhibitors. Dohme, Boehringer Ingelheim, and Takeda; Dr. Griesinger, re-
In conclusion, among patients with ALK-posi- ceiving grant support, advisory board fees, and lecture fees from
tive NSCLC who had not previously received an Roche, Takeda, Pfizer, Novartis, Celgene, Boehringer Ingel-
heim, Eli Lilly, Bristol-Myers Squibb, Merck Sharp & Dohme,
ALK inhibitor, progression-free survival was sig- AstraZeneca, and Siemens, advisory board fees and lecture fees
nificantly longer among those who received briga- from Chugai Pharmaceutical and AbbVie, and lecture fees from
tinib than among those who received crizotinib. Bayer; Dr. Morabito, receiving honoraria from AstraZeneca,
Bristol-Myers Squibb, Boehringer Ingelheim, Merck Sharp &
A data sharing statement provided by the authors is available Dohme, Pfizer, and Roche; Dr. Felip, receiving consulting fees
with the full text of this article at NEJM.org. and advisory fees from AstraZeneca, Bristol-Myers Squibb,
Supported by Ariad Pharmaceuticals, a wholly owned subsid- Boehringer Ingelheim, Celgene, Eli Lilly, Guardant Health,
iary of Takeda Pharmaceutical. Merck Sharp & Dohme, Novartis, Pfizer, Roche, Takeda, and
Dr. Camidge reports receiving grant support and honoraria Merck; Dr. Califano, receiving consulting fees and advisory fees
from Takeda and honoraria from AstraZeneca, Arrys Therapeu- from AstraZeneca, Bristol-Myers Squibb, Boehringer Ingelheim,
tics and Kyn Therapeutics, Genoptix, G1 Therapeutics, Mersana Lilly Oncology, Merck Sharp & Dohme, Novartis, Pfizer, Roche,
Therapeutics, Roche and Genentech, Ignyta, Daiichi Sankyo, Ji- and Takeda; Dr. Ghosh, receiving honoraria and fees for serving
angsu Hansoh Pharmaceutical, Biothera, Lycera, Revolution on a speakers bureau from Pfizer; Dr. Spira, receiving grant sup-
Medicines, Orion, Clovis Oncology, Celgene, and Novartis; Dr. port, consulting fees, and advisory fees from Takeda, and con-
H.R. Kim, receiving honoraria from AstraZeneca, Roche, and sulting fees and advisory fees from AbbVie, AstraZeneca, Bris-
Boehringer Ingelheim; Dr. Ahn, receiving advisory fees from tol-Myers Squibb, and Eli Lilly; Dr. Gettinger, receiving grant
AstraZeneca, Bristol-Myers Squibb, Boehringer Ingelheim, Mer- support and consulting fees from Ariad Pharmaceuticals and
ck Sharp & Dohme, and Novartis, and consulting fees from Al- Takeda and Bristol-Myers Squibb, consulting fees from Janssen,
pha Biopharmaceuticals; Dr. Yang, receiving honoraria and advi- and grant support from Genentech, Roche, Incyte, and Iovance
sory fees from Eli Lilly, Boehringer Ingelheim, Roche/ Biotherapeutics; Dr. Tiseo, receiving advisory fees from Bristol-
Genentech/Chugai, Merck Sharp & Dohme, Pfizer, Novartis, Myers Squibb, Boehringer Ingelheim, Eli Lilly, Merck Sharp &
Bristol-Myers Squibb, Ono Pharmaceutical, Takeda, and Astra- Dohme, Otsuka Pharmaceutical, Pfizer, and Roche, and grant
Zeneca, and advisory fees from Bayer, Astellas, Merck Serono, support and advisory fees from AstraZeneca; Dr. Gupta, being
Celgene, Merrimack, Yuhan Pharmaceuticals, Daiichi Sankyo, employed by Millennium Pharmaceuticals; Drs. Haney and Ker-
and Jiangsu Hansoh Pharmaceutical; Dr. Han, receiving grant stein, being employed by and holding stock and ownership in-
support and honoraria from Roche, advisory fees from Astra- terests in Ariad Pharmaceuticals; and Dr. Popat, receiving grant
Zeneca, Novartis, Bristol-Myers Squibb, and Takeda, advisory support, honoraria, consulting fees, and travel support from
fees and honoraria from Merck Sharp & Dohme, and honoraria Boehringer Ingelheim, grant support from Epizyme, Clovis On-
from Ono Pharmaceutical; Dr. J.-S. Lee, receiving honoraria and cology and Eli Lilly, grant support, consulting fees, and travel
advisory fees from AstraZeneca and Ono Pharmaceutical; Dr. support from Bristol-Myers Squibb, grant support, honoraria,
Hochmair, receiving fees for serving on a speakers bureau from and consulting fees from Roche, grant support and honoraria
AstraZeneca, Roche, Novartis, and Takeda; Dr. Li, receiving fees from Takeda, honoraria and consulting fees from AstraZeneca,
for serving on a speakers bureau from AstraZeneca and grant honoraria from Chugai Pharma, consulting fees from Novartis,
support and fees for serving on a speakers bureau from Roche Guardant Health, and AbbVie, grant support and consulting fees
and Genentech; Dr. Chang, receiving honoraria from AstraZen- from Pfizer, and consulting fees and travel support from Merck
eca, F. Hoffmann–La Roche, Eli Lilly Oncology, Pfizer, Boeh- Sharp & Dohme. No other potential conflict of interest relevant
ringer Ingelheim, Bristol-Myers Squibb, and Merck Sharp & to this article was reported.
Dohme; Dr. K.H. Lee, receiving honoraria and advisory fees Disclosure forms provided by the authors are available with
from AstraZeneca, Eli Lilly, Boehringer Ingelheim, Bristol- the full text of this article at NEJM.org.
Myers Squibb, and Merck Sharp & Dohme; Dr. Gridelli, receiv- We thank the patients and their families and caregivers, as well
ing fees for serving on a speakers bureau and advisory fees from as the trial coordinators and the operations staff, for participation
Pfizer and Roche; Dr. Delmonte, receiving consulting fees and in the trial; Jessica Tyler, Ph.D., of Millennium Pharmaceuticals, a
advisory fees from AstraZeneca and Boehringer Ingelheim; Dr. wholly owned subsidiary of Takeda Pharmaceutical, for editorial
Garcia Campelo, receiving honoraria, fees for serving on a assistance with an earlier version of the manuscript; and Lauren
speakers bureau, and advisory fees from Ariad Pharmaceuticals, Gallagher, Ph.D., and Lela Creutz, Ph.D., of Peloton Advantage for
AstraZeneca, Roche, Pfizer, Bristol-Myers Squibb, Boehringer professional medical writing assistance funded by Millennium
Ingelheim, and Takeda; Dr. Bearz, receiving fees for serving on Pharmaceuticals.
Appendix
The authors’ full names and academic degrees are as follows: D. Ross Camidge, M.D., Ph.D., Hye Ryun Kim, M.D., Ph.D., Myung‑Ju
Ahn, M.D., Ph.D., James Chih‑Hsin Yang, M.D., Ph.D., Ji‑Youn Han, M.D., Ph.D., Jong‑Seok Lee, M.D., Maximilian J. Hochmair, M.D.,
Jacky Yu‑Chung Li, M.B., B.S., Gee‑Chen Chang, M.D., Ph.D., Ki Hyeong Lee, M.D., Ph.D., Cesare Gridelli, M.D., Angelo Delmonte,
M.D., Ph.D., Rosario Garcia Campelo, M.D., Dong‑Wan Kim, M.D., Ph.D., Alessandra Bearz, M.D., Frank Griesinger, M.D., Ph.D.,
Alessandro Morabito, M.D., Enriqueta Felip, M.D., Ph.D., Raffaele Califano, M.D., Sharmistha Ghosh, M.B., Ch.B., Alexander Spira,
M.D., Ph.D., Scott N. Gettinger, M.D., Marcello Tiseo, M.D., Ph.D., Neeraj Gupta, Ph.D., F.C.P., Jeff Haney, M.A., David Kerstein, M.D.,
and Sanjay Popat, M.B., B.S., Ph.D.
The authors’ affiliations are as follows: the University of Colorado Cancer Center, Aurora (D.R.C.); Division of Medical Oncology,
Department of Internal Medicine, Yonsei Cancer Center, Yonsei University College of Medicine (H.R.K.), Samsung Medical Center
(M.-J.A.), and Seoul National University Hospital (D.-W.K.), Seoul, National Cancer Center, Goyang (J.-Y.H.), Seoul National Univer-
sity Bundang Hospital, Seongnam (J.-S.L.), and Chungbuk National University Hospital, Chungbuk National University College of
Medicine, Cheongju (K.H.L.) — all in South Korea; National Taiwan University Hospital (J.C.-H.Y.) and the Faculty of Medicine, School
of Medicine, National Yang-Ming University (G.-C.C.), Taipei, and the Division of Chest Medicine, Department of Internal Medicine,
Taichung Veterans General Hospital, Taichung (G.-C.C.) — all in Taiwan; the Ludwig Boltzmann Institute for COPD and Respiratory
Epidemiology, Department of Respiratory and Critical Care Medicine, Otto Wagner Hospital, Vienna (M.J.H.); Queen Elizabeth Hospi-
tal, Kowloon, Hong Kong (J.Y.-C.L.); Azienda Ospedaliera S. Giuseppe Moscati, Avellino (C.G.), the Scientific Institute of Romagna for
the Study and Treatment of Cancer, Meldola (A.D.), Centro di Riferimento Oncologico, Istituto Nazionale Tumori, IRCCS Struttura
Operativa Complessa Oncologia Medica A, Aviano (A.B.), Thoracic Medical Oncology, Istituto Nazionale Tumori, IRCCS Fondazione G.
Pascale, Naples (A.M.), and the Medical Oncology Unit, University Hospital of Parma, Parma (M.T.) — all in Italy; Complejo Hospital-
ario Universitario de A Coruña, Coruña (R.G.C.), and Vall d’Hebron University Hospital, Barcelona (E.F.) — both in Spain; the Depart-
ment of Hematology and Oncology, University Department of Internal Medicine–Oncology, Pius-Hospital Medical Campus, University
of Oldenburg, Oldenburg, Germany (F.G.); the Department of Medical Oncology, Christie NHS Foundation Trust, and Division of
Cancer Sciences, University of Manchester, Manchester (R.C.), and Guy’s and St. Thomas’ NHS Foundation Trust (S.G.) and Royal
Marsden Hospital and the National Heart and Lung Institute, Imperial College London (S.P.), London — all in the United Kingdom;
Virginia Cancer Specialists Research Institute and US Oncology Research, The Woodlands, TX (A.S.); Yale Cancer Center, New Haven,
CT (S.N.G.); and Millennium Pharmaceuticals, Cambridge, MA (N.G., J.H., D.K.).
References
1. Gainor JF, Varghese AM, Ou SH, et al. ery of brigatinib (AP26113), a phosphine cell lung cancer. N Engl J Med 2014;370:
ALK rearrangements are mutually exclu- oxide-containing, potent, orally active in- 1189-97.
sive with mutations in EGFR or KRAS: an hibitor of anaplastic lymphoma kinase. 22. Shaw AT, Felip E, Bauer TM, et al. Lor-
analysis of 1,683 patients with non-small J Med Chem 2016;59:4948-64. latinib in non-small-cell lung cancer with
cell lung cancer. Clin Cancer Res 2013;19: 13. Gettinger SN, Bazhenova LA, Langer ALK or ROS1 rearrangement: an interna-
4273-81. CJ, et al. Activity and safety of brigatinib tional, multicentre, open-label, single-arm
2. Wong DW, Leung EL, So KK, et al. in ALK-rearranged non-small-cell lung first-in-man phase 1 trial. Lancet Oncol
The EML4-ALK fusion gene is involved in cancer and other malignancies: a single- 2017;18:1590-9.
various histologic types of lung cancers arm, open-label, phase 1/2 trial. Lancet 23. Besse B, Solomon BJ, Felip E, et al.
from nonsmokers with wild-type EGFR Oncol 2016;17:1683-96. Lorlatinib in patients with previously
and KRAS. Cancer 2009;115:1723-33. 14. Uchibori K, Inase N, Araki M, et al. treated ALK+ advanced non-small cell
3. Koivunen JP, Mermel C, Zejnullahu K, Brigatinib combined with anti-EGFR anti- lung cancer (NSCLC): updated efficacy
et al. EML4-ALK fusion gene and efficacy body overcomes osimertinib resistance in and safety. Presented at the Annual Meet-
of an ALK kinase inhibitor in lung cancer. EGFR-mutated non-small-cell lung cancer. ing of the American Society of Clinical On-
Clin Cancer Res 2008;14:4275-83. Nat Commun 2017;8:14768. cology, Chicago, June 1–5, 2018 (poster).
4. Solomon BJ, Mok T, Kim D-W, et al. 15. Bazhenova L, Hodgson JG, Langer CJ, 24. Horn L, Infante JR, Reckamp KL, et al.
First-line crizotinib versus chemotherapy et al. Activity of brigatinib (BRG) in crizo- Ensartinib (X-396) in ALK-positive non-
in ALK-positive lung cancer. N Engl J Med tinib (CRZ)-resistant ALK+ NSCLC patients small cell lung cancer: results from a first-
2014;371:2167-77. (pts) according to ALK plasma mutation in-human phase I/II, multicenter study.
5. Costa DB, Kobayashi S, Pandya SS, status. J Clin Oncol 2017;15:Suppl:9065. Clin Cancer Res 2018;24:2771-9.
et al. CSF concentration of the anaplastic abstract. 25. Eisenhauer EA, Therasse P, Bogaerts J,
lymphoma kinase inhibitor crizotinib. 16. Kim DW, Tiseo M, Ahn MJ, et al. et al. New response evaluation criteria in
J Clin Oncol 2011;29(15):e443-e445. Brigatinib in patients with crizotinib- solid tumours: revised RECIST guideline
6. Costa DB, Shaw AT, Ou SHI, et al. refractory anaplastic lymphoma kinase- (version 1.1). Eur J Cancer 2009;45:228-47.
Clinical experience with crizotinib in pa- positive non-small-cell lung cancer: a ran- 26. DeMets DL, Lan KK. Interim analysis:
tients with advanced ALK-rearranged non- domized, multicenter phase II trial. J Clin the alpha spending function approach.
small-cell lung cancer and brain metasta- Oncol 2017;35:2490-8. Stat Med 1994;13:1341-52.
ses. J Clin Oncol 2015;33:1881-8. 17. Huber RM, Kim DW, Ahn MJ, et al. 27. Peters S, Camidge DR, Shaw AT, et al.
7. Zhang I, Zaorsky NG, Palmer JD, Brigatinib (BRG) in crizotinib (CRZ)- Alectinib versus crizotinib in untreated
Mehra R, Lu B. Targeting brain metasta- refractory ALK+ non–small cell lung can- ALK-positive non–small-cell lung cancer.
ses in ALK-rearranged non-small-cell lung cer (NSCLC): efficacy updates and explor- N Engl J Med 2017;377:829-38.
cancer. Lancet Oncol 2015; 16(13):
e510- atory analysis of CNS ORR and overall 28. Soria JC, Tan DSW, Chiari R, et al.
e521. ORR by baseline (BL) brain lesion status. First-line ceritinib versus platinum-based
8. Doebele RC, Pilling AB, Aisner DL, et al. J Clin Oncol 2018;36:Suppl:9061. abstract. chemotherapy in advanced ALK-rearranged
Mechanisms of resistance to crizotinib in 18. Camidge DR, Kim DW, Tiseo M, et al. non-small-cell lung cancer (ASCEND-4):
patients with ALK gene rearranged non- Exploratory analysis of brigatinib activity a randomised, open-label, phase 3 study.
small cell lung cancer. Clin Cancer Res in patients with anaplastic lymphoma Lancet 2017;389:917-29.
2012;18:1472-82. kinase-positive non-small-cell lung can- 29. Camidge DR, Peters S, Mok T, et al.
9. Katayama R, Shaw AT, Khan TM, et al. cer and brain metastases in two clinical Updated efficacy and safety data from the
Mechanisms of acquired crizotinib resis- trials. J Clin Oncol 2018;36:2693-701. global phase III ALEX study of alectinib
tance in ALK-rearranged lung cancers. Sci 19. Bazhenova LA, Gettinger SN, Langer (ALC) vs crizotinib (CZ) in untreated ad-
Transl Med 2012;4:120ra17. CJ, et al. Brigatinib (BRG) in anaplastic vanced ALK+ NSCLC. J Clin Oncol 2018;
10. McCoach CE, Le AT, Gowan K, et al. lymphoma kinase (ALK)-positive non- 36:Suppl:9043. abstract.
Resistance mechanisms to targeted thera- small cell lung cancer (NSCLC): long- 30. Alecensa (alectinib). South San Fran-
pies in ROS1(+) and ALK(+) non-small cell term efficacy and safety results from a cisco, CA:Genentech, 2018 (package in-
lung cancer. Clin Cancer Res 2018; 24: phase 1/2 trial. Ann Oncol 2017;28:Suppl 5: sert).
3334-47. 479-80. abstract. 31. Ahn M, Camidge DR, Tiseo M, et al.
11. Katayama R, Khan TM, Benes C, et al. 20. Ou S-HI, Ahn JS, De Petris L, et al. Brigatinib in crizotinib-refractory ALK+
Therapeutic strategies to overcome crizo- Alectinib in crizotinib-refractory ALK- NSCLC: updated efficacy and safety re-
tinib resistance in non-small cell lung rearranged non–small-cell lung cancer: sults from ALTA, a randomized phase 2
cancers harboring the fusion oncogene a phase II global study. J Clin Oncol 2016; trial. J Thorac Oncol 2017; 12:
Suppl 2:
EML4-ALK. Proc Natl Acad Sci U S A 34:661-8. S1755-S1756.
2011;108:7535-40. 21. Shaw AT, Kim D-W, Mehra R, et al. Copyright © 2018 Massachusetts Medical Society.
12. Huang WS, Liu S, Zou D, et al. Discov- Ceritinib in ALK-rearranged non–small-


BrigaVSCrizo ALK+

Uploaded by

Copyright:

Available Formats

BrigaVSCrizo ALK+

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

BrigaVSCrizo ALK+

Uploaded by

Copyright:

Available Formats

The n e w e ng l a n d j o u r na l of m e dic i n e

Brigatinib versus Crizotinib in ALK-Positive

n engl j med 379;21 nejm.org November 22, 2018 2027

2028 n engl j med 379;21 nejm.org November 22, 2018

n engl j med 379;21 nejm.org November 22, 2018 2029

311 Patients were assessed for eligibility

36 Were not eligible

275 Underwent randomization

41 Discontinued treatment 78 Discontinued treatment

35 Crossed over to brigatinib

Figure 1. Screening, Enrollment, Randomization, and Follow-up.

2030 n engl j med 379;21 nejm.org November 22, 2018

Table 1. Baseline Patient Characteristics in the Intention-to-Treat Population.*

Characteristic Brigatinib (N = 137) Crizotinib (N = 138) Total (N = 275)

n engl j med 379;21 nejm.org November 22, 2018 2031

2032 n engl j med 379;21 nejm.org November 22, 2018

n engl j med 379;21

November 22, 2018

Brigatinib NR (11–NR) 67 (47–80)

−40 −40 10 0.27 (95% CI, 0.13–0.54)

Best Change from Baseline in Target Lesions (%)

Variable Brigatinib Crizotinib Odds Ratio (95% CI)

* NR denotes not reached.

2034 n engl j med 379;21 nejm.org November 22, 2018

n engl j med 379;21 nejm.org November 22, 2018 2035

Event Brigatinib (N = 136) Crizotinib (N = 137)

Any Grade Grade ≥3 Any Grade Grade ≥3

number of patients (percent)

2036 n engl j med 379;21 nejm.org November 22, 2018

n engl j med 379;21 nejm.org November 22, 2018 2037

2038 n engl j med 379;21 nejm.org November 22, 2018

n engl j med 379;21 nejm.org November 22, 2018 2039

You might also like

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

* NR denotes not reached.