Role of Chemotherapy, Targeted therapy and Immunotherapy in NSCLC Part I

Role of Chemotherapy, Targeted therapy and
Immunotherapy in treatment of Non-Small Cell Lung
cancer (NSCLC) (Part I)
Mohammed Fathy Bayomy, MSc, MD
Lecturer
Clinical Oncology & Nuclear Medicine
Faculty of Medicine
Zagazig University

Lung Cancer: US Incidence and 5-Yr Relative Survival
(2008-2014)
SEER 18 Cancer Statistics Review, 2008-2014. https://seer.cancer.gov/statfacts/html/lungb.html
Percent of Cases by Stage
5%
16%
22%
57%
Localized:
confined to
primary site
Regional:
spread to
regional
lymph nodes
Distant:
cancer has
metastasized
Unknown:
unstaged
5-Yr Relative Survival by Stage
100
90
80
70
60
50
40
30
20
10
0
Stage
56.3
29.7
4.7 7.8
Aliveat5Yrs(%)

The Evolution of NSCLC
 Increasing incidence of adenocarcinoma
 Higher percentage of lung cancer in nonsmokers
 Increasing number of identified targetable mutational drivers
‒ Adenocarcinoma
‒ Nonsmokers
 More treatment options for NSCLC with EGFR, ALK, BRAF, and/or ROS1
aberrations
Meza. PLoS One. 2015;10:e0121323.

Progress in treatment for Lung Cancer Over 40 Yrs
Mid-70s
Chemo agents show low
response in NSCLC but higher
in SCLC
1976
Histologic types of
lung cancer
determined
1979
Cisplatin approved by FDA
1983
Etoposide, methotrexate,
and carboplatin approved
by FDA
1989
Carboplatin approved
by FDA for NSCLC
1990s
Adjuvant chemo improves
survival in NSCLC

1991
QoL identified as independent
prognostic factor
1992
 EGFR identified as target in
NSCLC
 Chemotherapy plus RT
increases survival vs RT alone
1995
Vinorelbine approved for first-
line treatment of advanced
NSCLC
1996
Gemcitabine + cisplatin
approved for first-line
treatment of advanced
NSCLC
1998
Paclitaxel in combination with
cisplatin approved for first-line
treatment of advanced NSCLC
1999
Docetaxel approved for advanced
NSCLC after failure of platinum
regimen
2000
Neoadjuvant chemo
demonstrates survival
results

2001
Chemo doublet
efficacy plateau
demonstrated
2002
Docetaxel + cisplatin
approved for chemo-
naive NSCLC
2003
Gefitinib approved for
EGFR+ NSCLC;
distribution later limited
2004
 Discovery that certain EGFR
mutations associated with
response to EGFR TKIs
 Erlotinib for
NSCLC approved for
second- and third-line
treatment
2005
Paradigm established
for histology
confirmation before
chemo and targeted
therapy
2006
Bevacizumab approved
for
first-line treatment
2007
 Transforming EMLA4-ALK
fusion gene identified in
NSCLC
 Genetic variants identified
that
confer risk for developing lung
cancer
2008
Pemetrexed approved in
combination for locally
advanced NSCLC
2009
Pemetrexed
approved as
maintenance
treatment for NSCLC

2010
2011
2012
Nab-paclitaxel +
carboplatin
approved
2013
 Erlotinib approved for
EGFR+ NSCLC
 Afatinib approved for
EGFR+ NSCLC
2014
 Ramucirumab + docetaxel
approved for metastatic NSCLC
 Ceritinib approved for ALK+
metastatic NSCLC
2015
 Alectinib approved for ALK+ metastatic NSCLC with
progression after crizotinib
 Necitumumab + gem/cis approved for
first-line treatment of metastatic squamous NSCLC
 Osimertinib approved for metastatic EGFR T790M+
NSCLC after failure of EGFR TKI therapy
 Gefitinib approved for metastatic NSCLC with EGFR
exon 19 deletion or exon 21 (L858R) substitution
mutations
2016
Crizotinib approved to treat
pts with ROS-1+ metastatic
NSCLC
 Targetable genetic abnormalities
identified in up to 60% of lung
adenocarcinomas
 Erlotinib maintenance treatment
approved
 Use of biomarkers established for
identifying optimal therapy, age of
personalized medicine
Crizotinib approved for ALK+
NSCLC

Targeted Therapy
Immunotherapy ± Chemotherapy
Supportive Care
Stage IV or
Recurrent Disease
Stage I Surgery (Radiation if Inoperable)
Stage II Surgery + Adjuvant Chemotherapy
Concurrent Chemoradiation ±
Consolidation Immunotherapy
Stage III
Duma. Mayo Clin Proc. 2019;94:1623.
Overview of Current NSCLC Treatment Paradigm

2003 2004 2005 2006
ALPI
HR =0.96
N=1207
ANITA
HR =0.76
N=840
JBR.10
HR =0.69
N=482
IALT
HR =0.86
N=467
CALGB 9633
HR = 0.83
N=344
RADIANT
MAGRIT
E1505
Closed to
Accrual
2008 2013 2014
ALPI–MVP vs OBS Stage I-IIIA Scagliotti GV et al. J Natl Cancer Inst 2003; 95: 1453-61
BLT-CPPP-based vs OBS Stage I-III Waller D et al. Eur J Cardiothorcic Surg 2004;26:173-182
IALT–CDDP-based vs OBS Stage I-IIIA Arriagada R et al. N Engl J Med 2004; 350: 350-61
JBR.10–CDDP-VNR vs OBS Stage IB-II Winton T et al. N Engl J Med 2005; 352:2589-97
ANITA–CDDP-VNR vs OBS Stage IB-IIIA Douilland JY et al. Lancet Oncol 2006; 7: 719-27
CALGB 9633–PAC-CARBO vs OBS Stage IB Strauss GM et al. J Clin Oncol 2008; 26: 5043-51
BLT
HR = 1.02
N=381
Adjuvant Therapy Timeline
ITACA
CALGB 30506
ALCHEMIST
CTONG1104
Afatinib Adjuv
EURECA

Meta-Analysis: Adjuvant Cisplatin-Based Chemo in NSCLC
Non-Small Cell Lung Cancer Collaborative Group. BMJ. 1995;311:899-909.
Mos
Surgery
100
90
80
70
60
50
40
30
20
10
0
0 6 12 18 24 30 36 42 48 54 60
Survival(%)
Surgery + Chemo
HR: 0.87
P = 0.08
8 trials, 1394 patients

IALT: Cisplatin-Based Adjuvant Treatment for NSCLC After
Resection
Arriagada R, et al. N Engl J Med. 2004;350:351-360.
Patients with stage
I-III NSCLC aged
18-75 yrs with no previous
malignancy after complete
surgical resection*
(N = 1867)
Cisplatin-based chemotherapy†
(n = 932)
No chemotherapy
(n = 935)
*Postoperative radiotherapy performed at discretion of institution.
†Chemotherapy regimens: etoposide: 56.5%; vinorelbine: 26.8%; vinblastine: 11.0%; vindesine: 5.8%.

IALT: Survival Benefit Observed With Chemotherapy vs
No Chemotherapy
Arriagada R, et al. N Engl J Med. 2004;350:351-360. Le Chevalier T, et al. ASCO 2003. Abstract 6.
Endpoint Chemo
(n = 932)
Control
(n = 935)
P Value
Median survival, mos 50.8 44.4 < .03
Median DFS, mos 40.2 30.5 < .003
5-yr survival, % 44.5 40.4 < .03
5-yr DFS, % 39.4 34.3 < .003

0
20
40
60
80
100
0 1 2 3 4 5
P <0.03
Surgery + CT
Surgery only
0 1 2 3 4 5
100
80
60
40
20
0
Yrs
%Survival
P =0.08
Yrs
%Survival
Surgery + CT
Surgery only
IALT Survival Meta-Analysis Survival
Arriagada R, et al. N Engl J Med. 2004;350:351-360. Non-Small Cell Lung Cancer Collaborative Group. BMJ. 1995;311:899-909.
IALT vs Meta-Analysis: Overall Survival

JBR.10: Adjuvant Vinorelbine + Cisplatin for Resected
NSCLC
Patients with completely
resected T2N0, T1N1,
or T2N1 NSCLC;
ECOG PS 0/1
(N = 482)
*Dose of 30 mg/m2 for first 18 patients; reduced due to hematologic toxicity.
Median
follow-up: 5.1 yrs
Median
follow-up: 5.3 yrs
Winton T, et al. N Engl J Med. 2005;352:2589-2597.
Stratified by nodal status (N0 vs N1) and
Ras status (neg/pos/unknown)
Vinorelbine 25 mg/m2*
wkly for 16 wks +
Cisplatin 50 mg/m2 on Days 1, 8
every 4 wks for 4 cycles
(n = 242)
Observation
(n = 240)

JBR.10: Adjuvant Vinorelbine + Cisplatin for Resected
NSCLC: Survival
Recurrence-Free Survival Overall Survival
100
80
60
40
20
0
0 2 4 6 8 10
Survival(%)
Yrs
P < .001
Observation
Vinorelbine
+ cisplatin
100
80
60
40
20
0
0 2 4 6 8 10
Survival(%)
Yrs
P = .009
Observation
Vinorelbine
+ cisplatin

Absolute improvement in 5-yr OS: 11%
Butts CA, et al. J Clin Oncol. 2010;28:29-34.
Median OS 5-Yr OS
94 mo 67%
72 mo 56%
HR: 0.78 (95% CI: 0.61-0.99)
P = 0.04
JBR.10: Vinorelbine vs Observation in Resected Stage
IB/II NSCLC: OS
At Risk, n
Observation
Vinorelbine
Percentage
0
20
40
60
80
100
0
240
242
3
155
182
6
117
135
9
Time (Yrs)
58
67
12
9
12
15
0
0
Vinorelbine
Observation

JBR.10: Adjuvant Vinorelbine + Cisplatin for NSCLC:
Survival by Stage
OS: Stage 1B OS: Stage II
100
80
60
40
20
0
0 2 4 6 8 10
Survival(%)
Yrs
P = .79
Observation
Vinorelbine + cisplatin
100
80
60
40
20
0
0 2 4 6 8 10
Survival(%)
Yrs
P = .004
Observation
Vinorelbine
+ cisplatin

Chemo
Tumor < 4 cm
Suggestion of benefit with chemo for larger tumors but nearly significant harm for smaller tumors
Interaction P = .022
Butts CA, et al. J Clin Oncol. 2010;28:29-34.
HR: 1.73 (95% CI: 0.98-3.04)
P = .056
JBR.10: Vinorelbine vs Observation in Resected Stage IB
NSCLC: OS by Tumor Size
At Risk, n
Observation
Vinorelbine
Observation
54
45
47
33
40
27
20
13
4
1
0
0
Percentage
0
20
40
60
80
100
0 3 6 9
Time (Yrs)
12 15
Tumor ≥ 4 cm
HR: 0.66 (95% CI: 0.39-1.14)
P = .133
54
66
36
54
29
43
20
23
1
5
0
0
Percentage
0
20
40
60
80
100
0 3 6 9
Time (Yrs)
12 15
Chemo
Observation
At Risk, n
Observation
Vinorelbine

ANITA: Adjuvant Vinorelbine + Cisplatin vs Observation
 Open, multicenter study
 Delta expected in the 2-yr survival rate: 10%
 Expected deaths: 466 events
Douillard JY, et al. Lancet Oncol. 2006;7:719-727.
Patients with stage
IB-IIIA NSCLC aged
18-75 yrs with no previous
malignancy after complete
surgical resection*
(N = 840)
Vinorelbine 30 mg/m2 IV wkly x 16 +
Cisplatin 100 mg/m2 IV on Days 1, 29, 57, 85
(n = 407)
Observation
(n = 433)
*Postoperative radiotherapy performed at discretion of institution.
Stratified by center, stage, and
histology

ANITA: Adjuvant Vinorelbine + Cisplatin vs Observation:
OS
Median OS 5-Yr OS*
43.7 mo 27%
65.7 mo 46%
HR: 0.80 (0.66-0.96);
P-value=0.017
*Without RT; with RT, 5-yr OS was 45% w/ chemo and 32% w/o chemo.
Observation
Chemotherapy
100
75
50
25
0
Survival(%)
At risk, n
Observation
Chemotherapy
433
407
211
228
293
288
119
144
65
63
17
18

Stage II
ANITA: Adjuvant Vinorelbine + Cisplatin vs Obs: OS by
Stage
N0 Patients N1 Patients
N2 Patients100
75
50
25
0
SurvivalDistribution
Function(%)
0 20 40 60 80 100
Time After Randomization (Mos)
Chemotherapy
Observation
100
75
50
25
0
Function(%)
0 20 40 60 80 100
100
75
50
25
0
Function(%)
0 20 40 60 80 100

CALGB 9633: Adjuvant Chemotherapy in Stage IB NSCLC
Patients with completely
resected T2N0M0
stage IB NSCLC
(N = 344)
Adjuvant Chemotherapy
Paclitaxel 200 mg/m2 IV +
Carboplatin AUC 6
4 cycles over 12 wks
(n = 173)
Observation
(n = 171)
Strauss GM, et al. J Clin Oncol. 2008;26:5043-5051.
Stratified by squamous vs other, poorly
differentiated vs other, and mediastinoscopy:
yes vs no

CALGB 9633: Adjuvant Chemo in Stage IB NSCLC: OS and
DFS
Overall Survival Disease-Free Survival
1.0
0.8
0.6
0.4
0.2
0
0
SurvivalProbability
20 40 60 80 100 120
Time (Mos)
HR = 0.83
90% CI: 0.64 to 1.08
P = .125
Chemotherapy (N = 173)
Control (N = 171)
1.0
0.8
0.6
0.4
0.2
0
0
SurvivalProbability
20 40 60 80 100 120
Time (Mos)
HR = 0.80
90% CI: 0.62 to 1.02
P = .065
Control (N = 171)

CALGB 9633: Adjuvant Chemo in Stage IB NSCLC: OS by
Tumor Size
Overall Survival: Tumor ≥ 4 cm Overall Survival: Tumor < 4 cm
1.0
0.8
0.6
0.4
0.2
0
0
SurvivalProbability
20 40 60 80 100 120
Time (Mos)
HR = 0.69
90% CI: 0.48 to 0.99
P = .043
Control (N = 97)
1.0
0.8
0.6
0.4
0.2
0
0
SurvivalProbability
20 40 60 80 100 120
Time (Mos)
HR = 1.12
90% CI: 0.75 to 1.07
P = .32
Control (N = 71)

LACE Meta-Analysis: OS Benefit From Postoperative
Cisplatin in Early-Stage NSCLC
 Pooled analysis of 5 trials
evaluating adjuvant cisplatin-based
chemotherapy (N = 4584)
‒ Cisplatin/vinorelbine most
commonly used agent (only
combination shown to prolong OS)
 Chemotherapy led to improved OS
‒ HR: 0.89
‒ Absolute benefit of 3.9% and 5.4%
at 3 and 5 yrs, respectively
Pignon. JCO. 2016;26:3552.
OS
100
80
60
40
20
0
OS(%)
0 1 2 3 4 5 ≥ 6
Yrs From Randomization
Chemotherapy
No chemotherapy

LACE Meta-Analysis: Adjuvant Chemotherapy Effect and
Stage
 Chemotherapy may be detrimental for stage IA, but these patients were generally not given the potentially best combination of
cisplatin + vinorelbine (13% of stage IA patients vs 43% for other stages)
Pignon JP, et al. J Clin Oncol. 2008;26:3552-3559.
Stage IA 104/347 1.40 (0.95-2.06)
Stage IB 515/1371 0.93 (0.78-1.10)
Stage II 893/1616 0.83 (0.73-0.95)
Stage III 878/1247 0.83 (0.72-0.94)
Category Deaths/Patients, n
HR for OS
(Chemo vs Control) HR (95% CI)
Chemotherapy Better Control Better
0.5 1.0 1.5 2.0 2.5
Test for trend: P = .04

ARM A: Cisplatin Doublet
(Investigator’s Choice)*
x 4, 21-day cycles
(n = 749)
ARM B: Cisplatin Doublet
(Investigator’s Choice)*
x 4, 21-day cycles +
Bevacizumab* Q3W, ≤ 1 yr
(n = 752)
Treatment-naive pts
with stage IB to IIIA,
resected NSCLC,
6-12 wks post-op, with
adequate nodal sampling,
no planned post-op RT,
acceptable organ function
(N = 1501)
Follow-up:
every 3 mos for 2 yrs,
every
6 mos through Yr 5,
annually through Yr 10
E1505: Study Design
 Primary endpoint: OS
 Secondary endpoint: DFS
 Study powered for primary endpoint only,[1] not for the subset analyses[2]
1. Wakelee HA, et al. WCLC 2015. Abstract 1608.
2. Wakelee HA, et al. ASCO 2016. Abstract 8507.
Stratified by cisplatin doublet, stage,
histology, sex
*See notes section for full regimens.

E1505: Median OS by Bevacizumab Use
Wakelee HA, et al. ASCO 2016. Abstract 8507.
Reproduced with permission.
Median follow-up: 50.3 mos
CT: not reached
CT + bev: 85.8 mos
HR: 0.99 (95% CI: 0.82-1.19;
P = .90)
1.0
0.8
0.6
0.4
0.2
0
0 12 24 36 48 60 72 84 96
Mos From Registration
ProbabilityofOS

8484
E1505 Chemotherapy Subset Analysis: DFS by Histology
 DFS not significantly different between chemotherapy groups
P = 0.58
1.0
0.8
0.6
0.4
0.2
0
0 12 24 36 48 60 72
ProbabilityofDFS
Nonsquamous
P =0.83
1.0
0.8
0.6
0.4
0.2
0
0 12 24 36 48 60 72
Squamous
Cis/docetaxel (119 events/201 cases)
Cis/gemcitabine (75 events/131 cases)
Cis/pemetrexed (228 events/497 cases)
Cis/vinorelbine (122 events/249 cases)
ProbabilityofDFS

E1505 Chemotherapy Subset Analysis: OS by Histology
 OS not significantly different between chemotherapy groups
P = 0.18 P = 0.99
1.0
0.8
0.6
0.4
0.2
0
0 12 24 36 48 60 72 84 96
1.0
0.8
0.6
0.4
0.2
0
0 12 24 36 48 60 72 84 96
ProbabilityofOS
Nonsquamous
ProbabilityofOS
Squamous
Cis/pemetrexed (126 events/497 cases)

Phase III Trials of Adjuvant EGFR TKI Therapy in NSCLC
 BR.19: adjuvant gefitinib vs placebo in molecularly unselected NSCLC post
chemotherapy
‒ No evidence to support adjuvant EGFR TKI; study closed early
 RADIANT: adjuvant erlotinib vs placebo in EGFR IHC+ or FISH+ NSCLC
‒ Early DFS benefit in EGFR mut+ patients without OS benefit . . .
no clear benefit to overtreating many if not curing
 ADJUVANT: adjuvant gefitinib vs cisplatin/vinorelbine in EGFR mut+ NSCLC
‒ Glaring problems: 2/3 of patients were stage IIIA, no rigorous imaging suggests patients
were understaged, omits CT with proven survival benefit on gefitinib arm; basically, a trial
of EGFR TKI vs CT in metastatic NSCLC
‒ DFS goes to 0% for all patients
Goss. J Clin Oncol. 2013;31:3320. Kelly. J Clin Oncol. 2013;31:3320. Zhong. Lancet Oncol. 2018;19:139.

RADIANT Trial: Adjuvant Erlotinib vs Placebo in Patients
With Resected EGFR+ NSCLC
 Radiology assessment: every 3 mos on treatment and yearly during long-term follow-up
 Primary endpoint: DFS
 Secondary endpoints: OS; DFS and OS in pts with del(19)/L858R (EGFR M+)
Phase III trial
Altorki NK, et al. Chicago Multidisciplinary Symposium in Thoracic Oncology 2014. Abstract 4.
No adjuvant
chemotherapy
Up to 4 cycles of
platinum-based
doublet
Tumor samples
EGFR IHC+ and/or EGFR FISH+
Randomization
stratified by: histology,
stage, prior adjuvant
chemo, EGFR FISH
status, smoking
status, country
(N = 973)
Erlotinib
150 mg/day
(n = 623)
≤ 180 d
≤ 90 d
2:1
Up to 4 cycles of
platinum-based
doublet
Placebo
(n = 350)
2-yr treatment
period
Stage
IB-IIIA NSCLC
Complete
surgical
resection

RADIANT Trial: DFS
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
ProbabilityofDFS
0 6 12 18 24 30 36 42 48 54 60 66
Mos
Placebo (156 events)
Median: 48.2 mos
Erlotinib (254 events)
Median: 50.5 mos
Log-rank P = .3235
HR: 0.90 (95% CI: 0.741-1.104)
Erlotinib
Placebo
Most common AEs with erlotinib (any grade): rash (86.4%), diarrhea (52.2%),
pruritus (26.4%); grade ≥ 3 over 2%: rash (22.3%), diarrhea (6.2%)

RADIANT Trial: Conclusions
 Adjuvant erlotinib did not prolong DFS in the overall population
 AEs primarily grade 2 or less
‒ Most common AEs with erlotinib (any grade): rash, diarrhea, pruritus;
grade ≥ 3 over 2%: rash, diarrhea

ADJUVANT: Study Design
 Randomized, phase III trial
Pts 18-74 yrs of age with
completely resected
pathologic stage II-IIIA (N1-
N2) NSCLC and centrally
confirmed EGFR activating
mutation (exon 19 del or exon
21 L858R);
ECOG PS 0-1
(N = 222)
Gefitinib 250 mg/day
for up to 2 yrs
(n = 111)
Vinorelbine 25 mg/m2 on Days 1, 8 +
Cisplatin 75 mg/m2 on Day 1
every 3 wks for up to 4 cycles
(n = 111)
Stratified by EGFR mutation, N stage
Wu YL, et al. ASCO 2017. Abstract 8500.
 Primary endpoint: DFS
 Secondary endpoints: 3-yr DFS, 5-yr DFS, OS, 5-yr OS, safety,
HRQoL, exploratory biomarker analyses
Follow-up every
12 wks until PD,
unacceptable
toxicity, death, or
study withdrawal

ADJUVANT: Baseline Characteristics
Characteristic
Gefitinib
(n = 111)
Vinorelbine/
Cisplatin
(n = 111)
Median age, yrs (range) 58 (32-74) 60 (26-76)
Female, % 58.6 58.6
Never smoker, % 73.9 76.6
ECOG PS 1, % 64.9 76.6
Pathology stage, %
 IIA
 IIB
 IIIA
 NA
29.7
3.6
64.9
1.8
29.7
3.6
64.0
2.7
Pathology, %
 Adenocarcinoma
 Squamous
 Adenosquamous
 NA
91.9
4.5
1.8
1.8
94.6
0.9
2.7
1.8
Characteristic
Gefitinib
(n = 111)
Vinorelbine/
Cisplatin
(n = 111)
EGFR mutation status, %
 Exon 19 deletion
 Exon 21 L858R
 False positive
52.3
47.7
0
51.4
47.7
0.9
Lymph node status, %
 N1
 N2
 NA
36.0
64.0
0
33.3
64.9
1.8
Type of resection, %
 Lobectomy
 Bilobectomy
 Pneumonectomy
 Wedge
 NA
83.8
11.7
2.7
0
1.8
82.0
12.6
2.7
1.8
0.9

ADJUVANT: DFS (Primary Endpoint)
*Secondary endpoint.
Gefitinib
(n = 111)
Vinorelbine/
Cisplatin
(n = 111)
HR for
Recurrence
(95% CI)
P Value
Median DFS, mos
 3-yr DFS rate,* %
28.7
34
18.0
27
0.60 (0.42-0.87) .005

ADJUVANT: DFS Subgroup Analysis
Subgroup Pts, n HR (95% CI) P Value P for Interaction
Overall (cox HR model) 222 0.58 (0.40-0.83) .003
Sex
 Male
 Female
89
130
0.60 (0.33-1.09)
0.58 (0.37-0.92)
.094
.020
.754
Smoker
 No
 Yes
167
52
0.61 (0.40-0.92)
0.56 (0.27-1.19)
.018
.132
.896
EGFR mutation status
 Exon 19 deletion
 Exon 21 L858R
115
108
0.55 (0.33-0.92)
0.62 (0.37-1.04)
.024
.071
.701
Lymph nodes
 N1
 N2
77
143
0.89 (0.45-1.76)
0.52 (0.34-0.80)
.743
.003
.232
Pathology
 Adenocarcinoma
 Nonadenocarcinoma
207
11
0.58 (0.40-0.84)
0.85 (0.16-4.46)
.004
.852
.506

ADJUVANT: AEs
AEs in ≥ 10% of
Pts, %
Gefitinib
(n = 106)
Vinorelbine/
Cisplatin
(n = 87)
All
Grades
Grade
≥ 3
All
Grades
Grade
≥ 3
Any 57.5 12.3 80.5 48.3
Neutropenia 2.8 0 52.9 34.5
Anemia 1.9 0.9 50.6 5.7
Leukopenia 3.8 0 47.1 16.1
Myelosuppression 0 0 13.8 3.4
Nausea 2.8 0 43.7 6.9
Vomiting 4.7 0 41.4 9.2
Anorexia 1.9 0 23.0 0
AEs in ≥ 10%
of Pts, %
Gefitinib
(n = 106)
Vinorelbine/
Cisplatin
(n = 87)
All
Grades
Grade
≥ 3
All
Grades
Grade
≥ 3
Rash 40.6 0.9 0 0
Elevated ALT 27.4 1.9 3.4 0
Elevated AST 11.3 1.9 1.1 0
Diarrhea 26.4 0.9 4.6 0
Cough 10.4 0 17.2 0
Fatigue 3.8 0 11.5 0
Fever 0.9 0 10.3 1.1

ADJUVANT: HRQoL
Pts With Clinically
Relevant
Improvement, %
Gefitinib
Vinorelbine +
Cisplatin
OR (95% CI) P Value
Total FACT-L 53.2 34.9 0.48 (0.25-0.91) .025
LCSS 71.3 46.0 0.34 (0.18-0.67) .002
TOI 40.2 24.2 0.47 (0.23-0.97) .041

ADJUVANT: Conclusions
 Gefitinib achieved superior DFS vs vinorelbine/cisplatin in pts with
completely resected stage II-IIIA (N1-N2) EGFRmut+ NSCLC
‒ Median DFS: 28.7 vs 18.0 mos (HR: 0.60; 95% CI: 0.42-0.87; P = .005)
‒ 3-yr DFS rate: 34% vs 27%
‒ OS data immature
 Gefitinib safety profile consistent with prior reports, with no cases of
interstitial lung disease
 Investigators conclude that adjuvant treatment with gefitinib for 2 yrs is
safe, feasible and could be preferred approach in resected N1/N2 EGFRmut+
NSCLC

Adjuvant EGFR TKI in NSCLC? Still an Open Question
 Adjuvant Lung Cancer Enrichment Marker Identification and Sequencing Trials
(post-op erlotinib vs placebo)
1. NCT02194738. 2. NCT02193282. 3. NCT02201992.
CONSENT
Preoperative,
during
adjuvant
therapy, or
after
Erlotinib x 2 yrs
Placebo x 2 yrs
Crizotinib x 2 yrs
Placebo x 2 yrs
ALCHEMIST-Screening[1]
Patients with resected or resectable
nonsquamous,
stage IB*-IIIA NSCLC
(Est. N = 8300)
SPECIMENS
Tumor EGFR and
ALK genotyping,
blood and tissue
sent to NCI
FOLLOW-UP
Patients not entering
trials followed Q6M
for 5 yrs
Phase III ALCHEMIST-EGFR (Est. N = 450)[2] Phase III ALCHEMIST-ALK (Est. N = 378)[3]
EGFR mut+
NSCLC
ALK-
rearranged
NSCLC
*≥ 4 cm.

ALCHEMIST Study: Genetic Testing in Resectable Stage
IB-IIIA NSCLC
ClinicalTrials.gov. NCT02194738.
Trials conducted at sites in the
NCI Clinical Trials Networks: NCTN & NCORP
Nonsquamous NSCLC (N = 6000-8000) Clinical/pathologic stage
IB (≥ 4 cm), II, IIIA
Post-op cohort with negative surgical margins
Pre-op
cohort
Post-op
cohort
Complete
resection +
standard adj
therapy per
treating
physician
Central
EGFR & ALK
genotyping
FFPE tissue & blood
specimen
EGFR mutation:
Phase III trial of erlotinib vs placebo
x 2 yrs
(n = 410) after any adj tx
ALK rearranged:
Phase III trial of crizotinib vs
placebo x 2 yrs
(n = 360) after any adj tx
Without molecular
alterations: Followed every
6 mos x 5 yrs after any adj
tx
FFPE tissue from biopsy done at
recurrence

Future Directions: Moving Immunotherapy Into
Resectable Early-Stage NSCLC
Data
Status
Resectable
Neoadjuvant Adjuvant
Reported
 LCMC3 (Phase II)[1]
Atezo → adj Atezo
MPR: 22%
 Pilot Study (Phase II)[3]
Nivo
MPR: 45%
 NADIM (Phase II)[4]
CT + Nivo → adj Nivo
MPR: 80%
Not yet
reported
 CheckMate 816 (Ph III)[2]
Nivo + Ipi or
Nivo + CT vs
CT
 TOP 1501 (Phase II)[5]
Pembro → adj Pembro
 KEYNOTE-671 (Phase III)[6]
CT + Pembro → adj Pembro
 ANVIL
(Phase III)[7]
Nivo vs Obs
 PEARLS
(Phase III)[8]
Pembro vs PBO
 IMpower010
(Phase III)[9]
Atezo vs BSC
 IFCT1401 (Phase III)[10]
Durva vs PBO
1. Rusch. WCLC 2018. Abstr MA04.09. 2. NCT02998528. 3. Forde. NEJM. 2018;378:1976. 4. Provencio. WCLC 2018.
Abstr OA01.05. 5. NCT02818920. 6. NCT03425643. 7. NCT02595944. 8. NCT02504372. 9. NCT02486718. 10.
NCT02273375.

Unresectable Stage III NSCLC: The Challenges
Brain as sanctuary site for metastases
Systemic
Control
Loco-regional
control

Combined Modality Therapy in Stage III NSCLC:
Meta-Analyses of Chemoradiotherapy Strategies
Strategy
No. of
Trials
N
Absolute Benefit at
Yr 3, %
HR for Survival
(95% CI)
P Value
Sequential CRT vs
RT alone[1] 22 3839 2.6
0.88
(0.82-0.94)
.0001
Concurrent CRT vs
RT alone[1] 16 2910 3.2
0.88
(0.81-0.95)
.0008
Concurrent CRT vs
Sequential CRT[2] 6 1205 5.7
0.84
(0.74-0.95)
.004
1. Rolland. J Thor Oncol. 2007;2:S309 (Abstract A1-04). 2. Aupérin. J Clin Oncol. 2010;28:2181.

Combined Modality Therapy in Stage III NSCLC:
Concurrent vs Sequential Chemoradiotherapy
 Meta-analysis of concurrent CRT vs sequential CRT in 6 trials of locally
advanced NSCLC (N = 1205)
‒ ↑ OS (HR: 0.84; P = .004)
‒ ↑ PFS (HR: 0.90; P = .07)
‒ ↓ risk of locoregional progression (HR: 0.77; P = .01)
‒ No difference in rates of distant progression between strategies
‒ Acute esophageal toxicity (grade 3/4) increased from 4% to 18% (RR: 4.9; 95%
CI: 3.1-7.8; P < .001)
‒ No difference in acute pulmonary toxicity
Aupérin. J Clin Oncol. 2010;28:2181.

Reasonable Treatment Options for Patients with Stage III
NSCLC
Concurrent CRT*:
Thoracic RT
Induction CT →
Concurrent CRT†:
Concurrent CRT† →
Consolidation CT:
CTCT
Induction CTInduction CT
Thoracic RT
CTCT
Thoracic RT
CTCT
Consol. CTConsol. CT
*Full-dose chemotherapy, typically cisplatin/etoposide. †Full- or low-dose chemotherapy, typically weekly carboplatin/paclitaxel.

Guideline-Recommended Chemotherapy Regimens
Used With Radiation Therapy for Stage III NSCLC
 Cisplatin + etoposide
 Cisplatin + vinblastine
 Carboplatin + pemetrexed
 Cisplatin + pemetrexed
 Carboplatin + paclitaxel

Long-Term Survival with Concurrent Chemoradiotherapy
in Stage III NSCLC
 RTOG 94-10: sequential vinblastine/cisplatin → RT
vs concurrent vinblastine/cisplatin + RT[1]
 START: tecemotide (MUC1-specific immunotherapy)
vs placebo after CRT[2]
1. Curran. J Natl Cancer Inst. 2011;103:1452. 2. Butts. Lancet Oncol. 2014;15:59.
Sequential
Concurrent
HR: 0.88 (0.75-1.03)
Concurrent CRT
Sequential CT → RT
100
75
50
25
0
OS(%)
Yrs Since Randomization
50 1 2 3 4
100
80
60
40
20
0
OS(%)
0 6 12 18 24 30 36 42 48 54 60 66
Mos Since Randomization
0.5 1.0 2.0
Favors tecemotide Favors placebo
Concurrent CRT (n = 806)
Sequential CT → RT (n = 433)
30.8
19.4
20.6
24.6
.0175
.419
0.78 (0.64-0.96)
1.11 (0.86-1.43)
P valueHR (95% CI)
Median OS, Mos
Tecemotide Placebo
Tecemotide
(n = 829)
Placebo
(n = 410)
Median OS, mos
(95% CI)
25.6
(22.5-29.2)
22.3
(19.6-25.5)
3-yr OS, n (%) 204 (40) 88 (37)
~20% to 25%
“cures”

Contemporary RT Techniques in Stage III NSCLC:
Long-Term OS Results from RTOG 0617
 Phase III RTOG 0617 evaluated
standard dose vs high dose CRT*
± cetuximab in stage IIIA or IIIB NSCLC
 5-yr OS of ~ 32% with high-quality
standard dose RT (before era of
immunotherapy)
‒ No role for RT dose escalation
‒ No benefit of cetuximab addition
 Now even better results when
immunotherapy added to CRT
backbone!
 Stage III NSCLC is a potentially
curable disease!
Bradley. Lancet Oncol. 2015;16:187. Bradley. ASTRO 2017. Abstract S105.
RT Dose Died
Median OS,
Yrs (95% CI)
HR (95% CI)
60 Gy
(n = 218)
150
2.4
(2.0-3.2)
RL
74 Gy
(n = 207)
163
1.7
(1.5-2.0)
1.35
(1.08-1.69)
OS
+ Censored
One-sided log-rank P = .004
Patients at Risk, n
*CT: paclitaxel/carboplatin.
100
80
60
40
20
0
0 1 2 3 4 5
Yrs
60 Gy
74 Gy
218
207
171
143
123
92
89
64
70
52
54
37
32.1%
23%
5-Yr OS

 Vast majority of patients with stage III NSCLC are eligible for concurrent CRT but
registry data show that 50% or fewer of these patients receive it
Ahmed. Clin Lung Cancer. 2017;18:706.
We Need to Ensure That All Eligible Patients With
Stage III NSCLC Receive Concurrent Chemoradiation
Selected Treatment Plans of Patients With NSCLC Who Were
Deemed Not to Be Candidates for CRT but in Fact Were Eligible
Clinical images courtesy of Kristin Higgins, MD. Emory University.

Concurrent Chemoradiotherapy in Stage III NSCLC: Lack
of Progress Before PACIFIC Study
 Multiple strategies tested without success
‒ Incorporation of newer chemotherapy drugs
‒ Induction/consolidation strategies
‒ Incorporation of targeted therapies (eg, cetuximab)
‒ Vaccine strategies
‒ Dose escalation of thoracic RT
Senan. J Clin Oncol. 2016;34:953. Santana-Davila. J Clin Oncol. 2015;33:567. Bradley. Lancet Oncol. 2015;16:187. Hanna. J Clin Oncol.
2008;26:5755. Gandara. Clin Lung Cancer. 2006;8:116. Ahn. J Clin Oncol. 2015;33:2660.

KN-024:
1st-line
pembro
better than CT
for PD-L1 ≥ 50%
Discovery of ICIs
Evolution of Care: Immunotherapy in NSCLC
2012 2015 2018
Drug
Approvals/
Other
Landmarks
Select
Trials
TMB associated
with response to
ICIs
20172016
Nivo:
2nd line post-CT
PACIFIC:
durva post-CRT
better than
placebo for
unresectable,
stage III
1990s
Pembro:
2nd line post-CT
PD-L1 ≥ 1%
Phase I: Nivolumab
efficacy demonstrated
Pembro:
1st line
PD-L1 ≥ 50%
KN-189 and -407:
1st-line pembro + CT
better than CT for nonsq
and sq, respectively
Durvalumab:
unresectable,
stage III
Atezo:
2nd line post-CT
IMpower150:
1st-line atezo + CT/bev better
than CT/bev for nonsq
2019
Pembro:
1st line PD-L1 ≥ 1%
Pembro + plt/pem
1st line nonsq
Pembro + carbo/pac or nab-pac:
1st line squamous
Atezo + CT/bev:
1st line nonsq
KN-042
1st-line pembro
better than CT for
PD-L1 ≥ 1%
Herbst. Nature. 2018;553:446. Ishida. EMBO J. 1992;11:3887. Nishimura. Immunity. 1999;11:141. Freeman. J Exp Med.
2000;192:1027. Antonia. NEJM. 2017;377:1919. Paz-Ares. NEJM. 2018;379:2040. Gandhi. NEJM. 2018;378:2078. Socinski.
NEJM. 2018;378:2288. Gettinger. J Clin Oncol. 2015;33:2004. Reck. NEJM. 2016;375:1823. Mok. Lancet. 2019;393:1819.

PACIFIC: Consolidation Durvalumab After Concurrent
CRT for Unresectable, Stage III NSCLC
 Randomized, double-blind, placebo-controlled phase III trial
Patients with locally advanced,
unresectable, stage III NSCLC without
PD after definitive platinum-based
concurrent CRT (≥ 2 cycles);
WHO PS 0/1 and
life expectancy ≥ 12 wks
(N = 713)
Durvalumab 10 mg/kg IV Q2W
for up to 12 mos
(n = 476)
Placebo IV Q2W
for up to 12 mos
(n = 237)
Until disease
progression or
unacceptable
toxicity
Stratified by age (< 65 vs ≥ 65 yrs), sex (male vs female), and
smoking history (current/former vs never)
Antonia. NEJM. 2018;379:2342. Antonia. WCLC 2018. Abstr PL02.01. Antonia. NEJM. 2017;377:1919. Paz-Arez. ESMO 2017. Abstr LBA1_PR.
Randomized
within 1-42 days
after cCRT
 Primary endpoints: PFS by BICR per RECIST v1.1, OS
 Secondary endpoints including: ORR, DoR, TTDM, PFS2, safety/tolerability, PROs

PACIFIC: Efficacy
100
80
60
40
20
0
PFS(%)
0 3 6 9 12 15 18 21 24 27
Mos
HR: 0.52 (95% CI: 0.42-0.65; P < .001)
Durvalumab
Placebo
Median PFS, Mos
(95% CI)
16.8 (13.0-18.1)
5.6 (4.6-7.8)
12-Mo PFS, %
(95% CI)
55.9 (51.0-60.4)
35.3 (29.0-41.7)
18-Mo PFS, %
(95% CI)
44.2 (37.7-50.5)
27.0 (19.9-34.5)
Placebo
Patients
at Risk, n
Durvalumab
Placebo
476
237
377
163
301
106
264
87
159
52
86
28
44
15
21
4
4
3
1
0
Durvalumab
Antonia. NEJM. 2017;377:1919. Paz-Arez. ESMO 2017. Abstr LBA1_PR.
PFS (ITT)
Events/Patients,
n/N (%)
214/476
157/237
Patients,%(95%CI)
P < .001
Treatment effect (RR): 1.78
(95% CI: 1.27-2.51)
Durvalumab
(n = 443)
Placebo
(n = 213)
0
5
10
15
20
25
30
35
28.4
(24.28-
32.89)
16.0
(11.31-
21.59)
Objective Response

PACIFIC: PFS by Subgroup
*HR not calculated for subgroup with best
response of CR because had < 20 patients.
Subgroup Durvalumab Placebo Unstratified HR for
Disease Progression or Death (95% CI)Patients, n
All patients
Sex
Male
Female
Age at randomization
< 65 yrs
≥ 65 yrs
Smoking status
Smoker
Nonsmoker
Disease stage
IIIA
IIIB
Tumor histologic type
Squamous
Nonsquamous
Best response to CRT
CR*
PR
SD
PD-L1 status
≥ 25%
< 25%
Unknown
EGFR mutation
Positive
Negative
Unknown
476
334
142
261
215
433
43
252
212
224
252
9
232
222
115
187
174
29
315
132
237
166
71
130
107
216
21
125
107
102
135
7
111
114
44
105
88
14
165
58
0.55 (0.45-0.68)
0.56 (0.44-0.71)
0.54 (0.37-0.79)
0.43 (0.32-0.57)
0.74 (0.54-1.01)
0.59 (0.47-0.73)
0.29 (0.15-0.57)
0.53 (0.40-0.71)
0.59 (0.44-0.80)
0.68 (0.50-0.92)
0.45 (0.33-0.59)
‒
0.55 (0.41-0.75)
0.55 (0.41-0.74)
0.41 (0.26-0.65)
0.59 (0.43-0.82)
0.59 (0.42-0.83)
0.76 (0.35-1.64)
0.47 (0.36-0.60)
0.79 (0.52-1.20)
0.25 0.50 1.00 2
Durvalumab Better Placebo Better
Antonia. NEJM. 2017;377:1919. Paz-Ares. ESMO 2017. Abstr LBA1_PR.

PACIFIC: Most Frequent AEs of Note
Any-Cause AEs in ≥ 10% of Patients in
Either Arm,* n (%)
Durvalumab† (n = 475) Placebo (n = 234)
Any Grade Grade 3/4 Any Grade Grade 3/4
Any 460 (96.8) 142 (29.9) 222 (94.9) 61 (26.1)
Cough 168 (35.4) 2 (0.4) 59 (25.2) 1 (0.4)
Pneumonitis or radiation pneumonitis‡ 161 (33.9) 16 (3.4) 58 (24.8) 6 (2.6)
Pyrexia 70 (14.7) 1 (0.2) 21 (9.0) 0
Pneumonia 62 (13.1) 21 (4.4) 18 (7.7) 9 (3.8)
Rash 58 (12.2) 1 (0.2) 17 (7.3) 0
Hypothyroidism 55 (11.6) 1 (0.2) 4 (1.7) 0
*Additional AEs in ≥ 10% of patients in either arm included: fatigue, dyspnea, diarrhea, decreased appetite, nausea, arthralgia, pruritus, upper
respiratory tract infection, constipation, headache, asthenia, back pain, musculoskeletal pain, anemia. †Included 2 patients randomized to placebo
but who received ≥ 1 dose of durvalumab. ‡Assessed by investigators with sponsor review and adjudication.
Antonia. NEJM. 2017;377:1919. Paz-Ares. ESMO 2017. Abstr LBA1_PR.
Grade 3/4 toxicity overall: 29.9% with durvalumab vs 26.1% with placebo

PACIFIC: OS (ITT)
Antonia. NEJM. 2018;379:2342. Antonia. WCLC 2018. Abstr PL02.01.
12-Mo OS, %
(95% CI)
83.1 (79.4-86.2)
75.3 (69.2-80.4)
24-Mo OS, %
(95% CI)
66.3 (61.7-70.4)
55.6 (48.9-61.8)
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
ProbabilityofOS
3 6 9 12 15 18 21 24 27 30 33 36 4539 42
Durvalumab 476 464 431 415 385 364 343 319 274 210 115 57 23 2 0 0
Placebo 237 220 198 178 170 155 141 130 117 78 42 21 9 3 1 0
0
HR: 0.68 (99.73% CI: 0.47-0.997; P = .0025)
Median follow-up: 25.2 mos (range: 0.2-43.1)
Durvalumab
Placebo
Events/Patients,
n/N (%)
183/476 (38.4)
116/237 (48.9)
Median OS, Mos
(95% CI)
NR (34.7-NR)
28.7 (22.9-NR)
Durvalumab
Placebo
Patients at Risk, n

PACIFIC: PFS and OS by Subgroup (ITT)
All patients
Sex Male
Female
Age at randomization < 65 yrs
≥ 65 yrs
Smoking status Smoker
Non-smoker
Disease stage IIIA
IIIB
Tumor histologic type Squamous
Nonsquamous
Prior definitive CT Cisplatin
Carboplatin
Best response to prior treatment CR
PR
SD
PFS HR (95% CI)* OS HR (95% CI)*
Antonia. WCLC 2018. Abstr. PL02.01. Antonia. NEJM. 2018;379:2342. Antonia. NEJM. 2017;377:1919.
*Not calculated if subgroup has < 20 events.
Data cut-off for PFS: February 13, 2017.
Data cut-off for OS: March 22, 2018.
Durvalumab better Placebo better
0.25 0.5 1.00 2.00
0.25 0.50 1.00 2.00
NA NA

Subgroup PFS, HR (95% CI) OS, HR (95% CI)
Overall
PD-L1 status (prespecified)
≥ 25%
< 25%
Unknown*
PD-L1 status (post hoc)†
≥ 1%
1% to 24%
< 1%
EGFR mutation
Positive‡
Negative
Unknown
PACIFIC: Specific Subsets?
Antonia. WCLC 2018. Abstr. PL02.01. Antonia. NEJM. 2018;379:2342. Antonia. NEJM. 2017;377:1919.
*Unknown PD-L1 status in 37% of patients; testing not required, obtained pre-CRT.
†1% cutoff used in unplanned post hoc analysis requested by a health authority.
‡HR (95% CI) not calculated because fewer than 20 events.
0.25 0.5 1.0 2.0
0.25 0.5 1.0 2.0
NA‡

Phase II LUN 14-179: Consolidation Pembrolizumab
Following Concurrent CRT in Unresectable Stage III NSCLC
 N = 93 patients with stage III NSCLC
(60% IIIA, 40% IIIB); n = 92 for efficacy
analysis
 Primary endpoint: time to metastatic
disease or death
 Histology: 55% adeno, 44% squamous,
1% mixed; 95% current/former
smokers
 Prior Tx: 72% carbo/pac, 26% cis/etop,
2% cis/pem, all with chest RT
 4-8 wks after CRT  pembro 200 mg
IV Q3W for up to 1 yr
 Pembro: 16% rec’d < 4 cycles, 84%
rec’d > 4 cycles, 37% completed 1 yr
 16 patients (17%) with grade > 2
pneumonitis, 1 pneumonitis-related
death
Durm. ASCO 018. Abstr 8500. Durm. WCLC 2018. Abstr OA01.07.
Outcome n = 92
Median PFS, mos
24-mo PFS, %
15.0
41.4
Median OS, mos
24-mo OS, %
NR
61.5

PACIFIC Is Just the Tip of the Iceberg:
A Subset of Trials in Unresectable NSCLC
Trial Phase Study Arms Outcomes
PACIFIC[1] III cCRT → Durva vs PBO  Median OS: NR vs 28.7 mos
 24-mo OS: 66.3% vs 55.6%
 Median PFS: 17.2 vs 5.6 mos
 18-mo PFS: 49.5% vs 26.7%
LUN 14-179[2] II cCRT → Pembro  Median OS: NR
 24-mo OS: 61.5%
 Median PFS: 15.0 mos
 24-mo PFS: 41.4%
DETERRED[3] II
(non-
randomized)
CRT vs CRT/Atezo
→ CT/Atezo
→ Atezo
 Median OS: 20.1 mos vs NR
 1-yr OS: 60% vs 77%
 Median PFS: 20.1 mos vs NR
 1-yr PFS: 60% vs 66%
NICOLAS[4] II Nivo + cCRT Not yet reported
AFT-16[5] II Atezo → cCRT → CT → Atezo Not yet reported
1. Antonia. NEJM. 2018;[Epub]. 2. Durm. WCLC 2018. Abstr OA01.07. 3. Lin. WCLC 2018.
Abstr OA01.06. 4. NCT02434081. 5. NCT03102242.

Conclusions: Unresectable, Stage III NSCLC
 Durvalumab consolidation following concurrent chemoradiotherapy is preferred
treatment in this setting
‒ Highly significant PFS and OS benefit in broad population post-CRT in PACIFIC trial
‒ OS benefit sustained out at least 2 yrs, no evidence that OS curves converge just after
treatment ends
‒ Should patients with PD-L1 < 1% and/or an EGFR mutation be excluded from receiving
durvalumab?
 Many trials ongoing in neoadjuvant, adjuvant, and stage III unresectable NSCLC,
with range of immunotherapy strategies
‒ We await trial results, but have reason to be optimistic

Lung Cancer Remains a Major Global Health Burden
 One of the most common cancers and leading cause of cancer deaths in US and
worldwide[1,2]
‒ New cases, 2016 (estimated): US, 224,390; global, 1.8 million
‒ Deaths, 2016 (estimated): US, 158,080; global, 1.5 million
 85% of cases are NSCLC (~ 188,000)[3]
‒ Stage IV at diagnosis: 40% (~ 75,000)[4]
 Standard of care for stage IV NSCLC: systemic therapy[5]
1. GLOBOSCAN Cancer Fact Sheets. 2012. 2. Siegel RL, et al. CA Cancer J Clin. 2016;66:7-30. 3.
American Cancer Society. Non-small-cell Lung Cancer. 4. SEER Cancer Statistics Review, 1975-2002.

Advanced NSCLC in an Era of Exciting New Targeted Agents
and Immunotherapies
 Whirlwind of scientific advances and FDA approvals for NSCLC in the past decade
 Tempting to believe we have “moved past” the era of chemotherapy
‒ In fact, only EGFR, ALK, and ROS1 mutations have FDA-approved targeted therapies
‒ Often monoclonal antibodies are used in combination with chemo
‒ PD-1–targeted immunotherapies approved in platinum-refractory NSCLC
‒ Pembrolizumab: 19% ORR; 45% ORR among pts with ≥ 50% of tumor cells expressing PD-L1[1]
‒ Nivolumab: 19% ORR[2]
 Most, if not all, advanced NSCLC pts will have chemotherapy at some point during the course of their
illness
1. Garon EB, et al. N Engl J Med. 2015;372:2018-2028.
2. Paz-Ares L, et al. ASCO 2015. Abstract LBA109.

Considerations for First-line Therapy of Advanced
NSCLC in 2019
 Clinical features
‒ Performance status
‒ Age, comorbidities, and smoking
history
‒ Nutritional status (eg, weight loss)
‒ Hemoptysis
‒ CNS metastases
‒ Previous chemotherapy in
adjuvant or locally advanced
setting
 Histologic subtyping
‒ Non-squamous vs squamous
 Molecular subtyping
‒ EGFR mutation, ALK or ROS1
translocation
or
‒ Next-generation sequencing

What Tools Can Facilitate Personalized Therapy in
Advanced-Stage NSCLC?
 How do we optimize therapy in individual pts (ie, first line, second line,
third line)?
Chemotherapy Checkpoint InhibitorsTargeted Therapy
Genomics-driven
TKIs:
 EGFR
 ALK
 ROS1
Histologic
subtyping for
chemotherapy
Anti–PD-1
Anti–PD-L1
Anti–CTLA-4

All recommendations are category 2A unless otherwise indicated.
*For PS 0–4; §Beware of flare phenomenon in subset of patients who discontinue EGFR TKI. If disease flare occurs, restart EGFR TKI; ¶Afatinib + cetuximab may be considered in patients with disease progression on
EGFR TKI therapy; **The data in the second-line setting suggest that PD-1/PD-L1 inhibitor monotherapy is less effective, irrespective of PD-L1 expression, in EGFR+/ALK+ NSCLC; §§Albumin-bound paclitaxel may be
substituted for either paclitaxel or docetaxel in patients who have experienced hypersensitivity reactions after receiving paclitaxel or docetaxel despite premedication, or for patients where the standard premedications
(i.e. dexamethasone, H2 blockers, H1 blockers) are contraindicated; ¶¶Carboplatin-based regimens are often used for patients with comorbidities or those who cannot tolerate cisplatin; ***Contraindications for
treatment with PD-1/PD-L1 inhibitors may include active or previously documented autoimmune disease and/or current use of immunosuppressive agents or presence of an oncogene, which would predict lack of
benefit; §§§If progression on PD-1/PD-L1 inhibitor, switching to another
PD-1/PD-L1 inhibitor is not routinely recommended; ¶¶¶Bevacizumab should be given until progression; ****Any regimen with a high risk of thrombocytopenia and the potential risk of bleeding should be used with
caution in combination with bevacizumab; §§§§Criteria for treatment with bevacizumab: non-squamous NSCLC, and no recent history of haemoptysis. Bevacizumab should not be given as a single agent, unless as
maintenance if initially used with chemotherapy; ¶¶¶¶Plasma-based testing should be considered at progression on EGFR TKIs for the T790M mutation. If plasma-based testing is negative, tissue-based testing with
rebiopsy material is strongly recommended. Practitioners may want to consider scheduling the biopsy concurrently with plasma testing referral; *****Consider osimertinib (regardless of T790M status) or pulse erlotinib
for progressive leptomeningeal disease; §§§§§In the randomised Phase III trial of dacomitinib, patients with brain metastases were not eligible for enrollment. In the setting of brain metastases, consider other options;
¶¶¶¶¶Beware of flare phenomenon in subset of patients who discontinue EGFR TKI. If disease flare occurs, restart EGFR TKI.
1. National Comprehensive Cancer Network. NCCN Guidelines: Non-small Cell Lung Cancer Version 3. 2019. https://www.nccn.org/ (Accessed: 18 December 2018).
National Comprehensive Cancer Network guidelines: EGFR mutation+
advanced or metastatic adenocarcinoma1

National Comprehensive Cancer Network guidelines: BRAF V600E mutation+,
ALK translocation+, ROS1 translocation+ advanced or metastatic adenocarcinoma1
*For PS 0–4; §Beware of flare phenomenon in subset of patients who discontinue ALK inhibitor. If disease flare occurs, restart ALK inhibitor; ¶Patients who are intolerant to crizotinib may be switched to ceritinib, alectinib, or brigatinib;
**If not previously given; §§Ceritinib, alectinib, or brigatinib are treatment options for patients with ALK-positive metastatic NSCLC that has progressed on crizotinib; ¶¶Lorlatinib is a treatment option after progression on crizotinib and
alectinib, brigatinib, or ceritinib; ***Single-agent vemurafenib or dabrafenib are treatment options if the combination of dabrafenib + trametinib is not tolerated; ¶¶¶Albumin-bound paclitaxel may be substituted for either paclitaxel or
docetaxel in patients who have experienced hypersensitivity reactions after receiving paclitaxel or docetaxel despite premedication, or for patients where the standard premedications (i.e. dexamethasone, H2 blockers, H1 blockers) are
contraindicated; §§§Carboplatin-based regimens are often used for patients with comorbidities or those who cannot tolerate cisplatin; ****Contraindications for treatment with
PD-1/PD-L1 inhibitors may include active or previously documented autoimmune disease and/or current use of immunosuppressive agents or presence of an oncogene, which would predict lack of benefit; §§§§If progression on PD-
1/PD-L1 inhibitor, switching to another
PD-1/PD-L1 inhibitor is not routinely recommended; ¶¶¶¶Bevacizumab should be given until progression; *****Any regimen with a high risk of thrombocytopenia and the potential risk of bleeding should be used with caution in
combination with bevacizumab; §§§§§Criteria for treatment with bevacizumab: non-squamous NSCLC, and no recent history of haemoptysis. Bevacizumab should not be given as a single agent, unless as maintenance if initially used
with chemotherapy

All recommendations are category 2A unless otherwise indicated.
*Contraindications for treatment with PD-1/PD-L1 inhibitors may include active or previously documented autoimmune disease and/or current use of immunosuppressive agents or presence of an oncogene, which would predict lack of
benefit; §If progression on PD-1/PD-L1 inhibitor, switching to another PD-1/PD-L1 inhibitor is not routinely recommended; ¶Bevacizumab should be given until progression; **Any regimen with a high risk of thrombocytopenia and the potential
risk of bleeding should be used with caution in combination with bevacizumab; §§Criteria for treatment with bevacizumab: non-squamous NSCLC, and no recent history of haemoptysis. Bevacizumab should not be given as a single agent,
unless as maintenance if initially used with chemotherapy; ¶¶The data in the second-line setting suggest that PD-1/PD-L1 inhibitor monotherapy is less effective, irrespective of PD-L1 expression, in EGFR+/ALK+ NSCLC; ***If progression on
PD-1/PD-LI inhibitor, switching to another PD-1/PD-L1 inhibitor is not routinely recommended; §§§Pembrolizumab is approved for patients with NSCLC tumours with PD-L1 expression levels ≥1%, as determined by an FDA-approved test; ¶¶¶If
not previously given; ****If not already given, options for PS 0–2 include (nivolumab, pembrolizumab, or atezolizumab), docetaxel (Category 2B), pemetrexed (Category 2B), gemcitabine (Category 2B), or ramucirumab + docetaxel
(Category 2B); options for PS 3–4 include BSC. Options for further progression are BSC or clinical trial;. Clinical Trials: NCCN believes that the best management of any patient with cancer is in a clinical trial. Participation in clinical trials is
especially encouraged; ¶¶¶¶If patient has not received platinum-doublet chemotherapy, refer to systemic therapy. If patient received platinum chemotherapy and anti-PD-1/PD-L1, refer to subsequent therapy; *****If pembrolizumab
monotherapy given; §§§§§If pembrolizumab/carboplatin/pemetrexed or pembrolizumab/cisplatin/pemetrexed given; ¶¶¶¶¶ If atezolizumab/carboplatin/paclitaxel/bevacizumab given; ******If bevacizumab was used with a first-line
pemetrexed/platinum chemotherapy regimen.
National Comprehensive Cancer Network guidelines:
non-targetable advanced or metastatic adenocarcinoma1

*Contraindications for treatment with PD-1/PD-L1 inhibitors may include active or previously documented autoimmune disease and/or current use of immunosuppressive agents or presence of an oncogene, which would
predict lack of benefit; §If progression on PD-1/PD-L1 inhibitor, switching to another PD-1/PD-L1 inhibitor is not routinely recommended; ¶The data in the second-line setting suggest that PD-1/PD-L1 inhibitor monotherapy
is less effective, irrespective of PD-L1 expression, in EGFR+/ALK+ NSCLC; **Pembrolizumab is approved for patients with NSCLC tumours with PD-L1 expression levels ≥1%, as determined by an FDA-approved test; §§If
not previously given; ¶¶If not already given, options for PS 0–2 include (nivolumab, pembrolizumab, or atezolizumab), docetaxel (Category 2B), pemetrexed (Category 2B), gemcitabine (Category 2B), or
ramucirumab + docetaxel (Category 2B); options for PS 3–4 include BSC. Options for further progression are BSC or clinical trial; ***If pembrolizumab monotherapy given; §§§If patient has not received platinum-
doublet chemotherapy, refer to systemic therapy. If patient received platinum chemotherapy and anti-PD-1/PD-L1, refer to subsequent therapy; ¶¶¶ If pembrolizumab/(cisplatin or carboplatin)/(paclitaxel or albumin-bound
paclitaxel) given.
National Comprehensive Cancer Network guidelines:
advanced or metastatic squamous cell carcinoma1

1. Planchard D, et al. Ann Oncol 2018;29(Suppl. 4):iv192–iv237 [Published online 3 October 2018; updated 26 January 2019].
*Not EMA-approved.
§MCBS score for the combination of bevacizumab with gefitinib or erlotinib.
¶PS 0–1: 4–6 cycles cisplatin or carboplatin based doublets (gemcitabine, docetaxel, paclitaxel, vinorelbine), cisplatin/pemetrexed, carboplatin/pemetrexed; carboplatin/nab-PC, ± bevacizumab; PS 2, <70 years and PS 0–2, selected ≥70 years:
4–6 cycles carboplatin-based chemotherapy, single-agent chemotherapy (gemcitabine, vinorelbine, docetaxel, pemetrexed).
European Society for Medical Oncology guidelines:
targetable advanced or metastatic NSCC1
Treatment strategy should take into account histology, molecular pathology, age, PS, comorbidities and patient preference

non-targetable advanced or metastatic NSCC1
*In absence of contraindications and conditioned by the registration and accessibility of anti-PD-(L)1 combinations with platinum-based chemotherapy, this strategy will be preferred to platinum-based chemotherapy in patients with PS 0–1 and PD-L1 <50%.
Alternatively, if TMB can accurately be evaluated, and conditioned by the registration and accessibility, nivolumab plus ipilimumab should be preferred to platinum-based standard chemotherapy in patients with NSCLC with a high TMB; §Not EMA-approved.

advanced or metastatic SCC1
*In absence of contraindications and conditioned by the registration and accessibility of anti-PD-(L)1 combinations with platinum-based chemotherapy, this strategy will be preferred to platinum-based chemotherapy in patients with PS 0–1 and PD-L1 <50%.
Alternatively, if TMB can accurately be evaluated, and conditioned by the registration and accessibility, nivolumab plus ipilimumab should be preferred to platinum-based standard chemotherapy in patients with NSCLC with a high TMB; §Molecular testing is not
recommended in SCC, except in those rare circumstances when SCC is found in a never-, long-time ex- or light-smoker (<15 pack-years); ¶Not EMA-approved.

Moving the Bar: First-line Decision Tree for Advanced
NSCLC
NSCLC
PD-L1 high
(≥ 50%)
PD-1i
Targetable genetic
mutation
No mutation/
PD-L1 low (1%-49%)
or negative (< 1%)
ALK ROS1
Squamous
histology
Nonsquamous
histology
EGFR TKI ALK TKI ROS1 TKI
BRAF +
MEK TKI
Chemotherapy
± PD-1i
Chemotherapy
± PD-(L)1i
EGFR BRAF

Melosky. Front Oncol. 2017;7:38. Masters. J Clin Oncol. 2015;33:3488. Necitumumab PI. Pembrolizumab PI. Atezolizumab PI. Nivolumab PI. Afatinib PI. Ramucirumab PI. ESMO Guidelines
Committee. eUpdate – Metastatic non-small-cell lung cancer algorithms. June 28, 2017.
Poor PS
Nonsquamous Squamous Single-agent or
combination CT,
consider hospice
Carboplatin/paclitaxel
(or nab-paclitaxel) +
pembrolizumab,
plt doublet,
cisplatin/gemcitabine/
necitumumab
Progression
Based on prior therapy: atezolizumab, nivolumab, pembrolizumab (if ≥ 1% PD-L1+) after platinum-based doublet chemotherapy alone, or other
systemic agents including docetaxel ± ramucirumab, pemetrexed, gemcitabine, or afatinib
Clinical (PS)
Clinical
Histologic
Bevacizumab eligible Bevacizumab ineligible
Carboplatin/pemetrexed + pembrolizumab,
carboplatin/paclitaxel + bevacizumab +
atezolizumab, plt doublet ± bevacizumab,
Carboplatin/pemetrexed +
pembrolizumab, plt doublet
Advanced NSCLC:
No Actionable Biomarker
Good PS
Based on prior therapy for patients with stable disease or better: bevacizumab, pemetrexed, bevacizumab + pemetrexed,
pembrolizumab, gemcitabine, or docetaxel; observation is also a reasonable option
FirstlineMaintenance
Second
lineand
beyond
Treatment Algorithm for Advanced NSCLC in 2019:
No Actionable Biomarker

Treatment Algorithm for Advanced NSCLC in 2019:
Molecular Biomarker Positive
ALK positive
Progression
EGFR mutation positive
Advanced NSCLC
(molecular biomarker positive)
ROS1 positive PD-L1 ≥ 50%
positive
Crizotinib
Follow treatment options for adenocarcinoma or squamous cell carcinoma without actionable biomarker
Pembrolizumab
Osimertinib
EGFR T790M
mutation negative or
previous osimertinib
Alectinib, brigatinib,
ceritinib, or
lorlatinib dependent
on previous therapy
Alectinib (preferred),
ceritinib, or crizotinib
Osimertinib (preferred)
erlotinib, afatinib, gefitinib, or
dacomitinib
EGFR T790M
mutation positive
BRAF V600E
positive
Dabrafenib/
trametinib*
Firstline
Second
lineand
beyond
*Or as second-line after chemotherapy

Immunotherapy Treatment Algorithm for NSCLC in 2019
 Where does TMB fit in, if anywhere?
NonsquamousSquamous
Pembrolizumab or
Pembrolizumab + CT
Pembrolizumab +
Carboplatin/Pemetrexed -or-
Chemotherapy Alone
Pembrolizumab +
Carboplatin/Pemetrexed
Pembrolizumab or
Pembrolizumab + CT
Pembrolizumab +
Carboplatin/Paclitaxel or nab-Paclitaxel
Pembrolizumab +
Carboplatin/Paclitaxel or nab-Paclitaxel
PD-L1 ≥ 50%
PD-L1 ≥ 1-49%
PD-L1 < 1%
Atezolizumab+
Carboplatin/Pemetrexed+
Bevacizumab

Chemotherapy for treatment
of Metastatic (Advanced) NSCLC

First-line Platinum Doublet Chemotherapy
 Benefit of cisplatin vs carboplatin for treatment of advanced NSCLC is
controversial
 Treatment-related adverse events should be considered during treatment
selection
‒ Cisplatin associated with increased nausea, vomiting, neurotoxicity, and
renal toxicity
‒ Carboplatin associated with thrombocytopenia
 Carboplatin-based regimens most often used in the US
1. Fossella F, et al. J Clin Oncol. 2003;21:3016-3024.
2. Hotta K, et al. J Clin Oncol. 2004;22:3852-3859.
3. Artizzoni A, et al. J Natl Cancer Inst. 2007;99:847-857.

First-line Platinum Doublet Chemotherapy in NSCLC:
Safety
 Platinum doublet arm: pemetrexed/carboplatin
‒ Lower rate of discontinuing treatment
‒ Due to early death, early progression, or clinical deterioration
 Severe anemia, neutropenia: most significant increased toxicity in doublet
arm vs pemetrexed monotherapy
‒ Can be anticipated, often prevented
Zukin M, et al. J Clin Oncol. 2013;x:2849-2853.

Can You Safely Treat PS 2 Pts With Doublet
Chemotherapy?
 Randomized phase III trial of single-agent pemetrexed vs carboplatin/pemetrexed in
pts with advanced nonsquamous NSCLC and ECOG PS 2
‒ One third of pts were older than 70 yrs of age
Zukin M, et al. J Clin Oncol. 2013;31:2849-2853.
Survival Outcome, Mos Pemetrexed
x 4 Cycles
Carboplatin/Pemetrexed
x 4 Cycles
Median PFS 2.8 5.8
Median OS 5.3 9.3

1.0
0.8
Schiller JH, et al. N Engl J Med. 2002;346:92-98.
ECOG 1594
0.6
0.4
0.2
0
0 5 10 15 20 25 30
Mos
OS(%)
Cisplatin/paclitaxel
Cisplatin/gemcitabine
Cisplatin/docetaxel

Scagliotti GV, et al. J Clin Oncol. 2008;26:3543-3551.

Cisplatin + Pemetrexed (C/P) vs Cisplatin + Gemcitabine
(C/G) in Advanced NSCLC: OS by Histology
Mos
SurvivalProbability
SquamousNonsquamous
Mos
SurvivalProbability
Scagliotti GV, et al. J Clin Oncol. 2008;26:3543-3551.
C/P
C/G
C/P vs C/G
Median Survival
11.8 mos
10.4 mos
Adjusted HR: 0.81
(95% CI: 0.70-0.94)
C/P
C/G
C/P vs C/G
Median Survival
9.4 mos
10.8 mos
Adjusted HR: 1.23
(95% CI: 1.00-1.51)
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
300 6 12 18 24
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
300 6 12 18 24

Select Antiangiogenic Agents for NSCLC
Drug Target Route Frequency Clinical Status
Bevacizumab VEGF ligand IV q3w Approved
Ramucirumab VGFR-2 IV q3w Approved
Sorafenib Raf, Kit, Flt-3, VEGFR-2, VEGFR-3, PDGFR- PO Twice daily Not approved
Vandetanib VEGFR-2, VEGFR-3, RET, EGFR PO Daily Not approved
Sunitinib VEGFR-1, VEGFR-2, VEGFR-3, PDGFR-α, PDGFR-,
Flt-3, c-kit
PO Twice daily Not approved
Cediranib VEGFR-2, VEGFR-1, VEGFR-3, c-kit, Flt-3 PO Daily Not approved
Motesanib VEGFR-1, VEGFR-2, VEGFR-3, PDGFR, RET, kit PO Daily Not approved
Axitinib VEGFR-1, VEGFR-2, VEGFR-3, PDGFR-, kit PO Twice daily Not approved
Pazopanib VEGFR-2, VEGFR-2, VEGFR-3, PDGFR-α, PDGFR-,
c-kit
PO Daily Not approved
Nintedanib VEGFR1-3; FGFR1,3; PDGFα,β PO Twice daily Phase III

Phase III Studies of Anti-VEGF Antibody Therapy in NSCLC
1. Sandler A, et al. N Engl J Med. 2006;355:2542-2550. 2. Reck M, et al. J Clin Oncol. 2009;27:1227-1234.
3. Herbst RS, et al. Lancet. 2011;377:1846-1854. 4. Johnson BE, et al. J Clin Oncol. 2013;31:3926-3934.
*Significantly different from comparator.
Trial Treatment line Treatment Arms Median Survival, Mos Trial Outcome
ECOG 4599[1] First line Paclitaxel/carboplatin
+ bevacizumab
OS: 12.3* Positive
Paclitaxel/carboplatin OS: 10.3
AVAiL[2] First line Cisplatin/gemcitabine
+ bevacizumab
PFS
High dose: 6.5*
Low dose: 6.7*
Positive for
primary endpoint
+ placebo
PFS: 6.1
BETA[3] Second line Bevacizumab + erlotinib OS: 9.3 Negative
Erlotinib + placebo OS: 9.2
ATLAS[4] First line maintenance Bevacizumab + erlotinib PFS: 4.8* Positive for
primary endpoint
Bevacizumab + placebo PFS: 3.7

Phase III Studies of Bevacizumab in NSCLC
*Significantly different from comparator.

Sandler A, et al. N Engl J Med. 2006;355:2542-2550.
ECOG E4599: Paclitaxel/Carboplatin ± Bevacizumab in
Advanced Non-squamous NSCLC
Pts with recurrent or advanced
nonsquamous NSCLC, no prior
chemotherapy
(N = 878)
Paclitaxel 200 mg/m2 on Day 1 +
Carboplatin AUC 6.0 mg/mL/min on Day 1
for six 3-wk cycles; no crossover to bevacizumab permitted
(n = 444)
Carboplatin AUC 6.0 mg/mL/min on Day 1
for six 3-wk cycles +
Bevacizumab 15 mg/kg on Day 1
every 3 wks until PD or unacceptable toxicity
(n = 434)
Endpoint, % PC BPC Significance
ORR (CR + PR) 15.0 35.0 P < .001
Median OS, mos 10.3 12.3
HR: 0.79;
P = .003
Median PFS, mos 4.5 6.2
HR: 0.66;
P < .001

ECOG E4599: Bev + Carboplatin and Paclitaxel in
Metastatic Non-squamous NSCLC
Sandler A, et al. N Engl J Med. 2006;355:2542-2550. Sandler A, et al. J Thorac Oncol. 2010;5:1416-1423.
Overall Population Adenocarcinoma
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
ProbabilityofOS
Mos
10.3
0 426 12 18 24 30 36
12.3
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
ProbabilityofOS
Mos
10.3
0 426 12 18 24 30 36
14.2
CP + bev (n = 434)
CP (n = 444)
HR: 0.79 (95% CI: 0.67-0.92;
P = .003)
CP + bev (n = 300)
CP (n = 302)
HR: 0.69 (95% CI: 0.58-0.83;
P = .009)

Biomarkers for Antiangiogenic Therapy
 Hypertension appears to be an early predictor of benefit with VEGF
inhibitors
 ICAM and VEGF-A levels seem to be more prognostic than predictive
 VEGF polymorphisms hold promise and may be predictive
 Plasma cytokine/angiogenic factors are intriguing
 IL-6, IL-12 hold promise
 Still in need of truly predictive biomarkers to help us move forward in
antiangiogenic therapy

Primary endpoint: ORR
Secondary endpoints: PFS, OS, safety
Patients with stage IIIb/IV
NSCLC, ECOG PS 0-1, no
previous chemotherapy for
metastatic disease
(N = 1050)
Nab-Paclitaxel 100 mg/m2 on Days 1, 8, 15 +
Carboplatin AUC 6 on Day 1
No premedication
Carboplatin AUC 6 on Day 1
Premedication: dexamethasone, antihistamines
Stratified by stage (IIIb vs IV),
age (< 70 yrs vs > 70 yrs), sex,
histology (squamous vs non-squamous), geographic region
21-day cycles
Socinski MA, et al. J Clin Oncol. 2012;30:2055-2062.
Carboplatin/Nab-Paclitaxel vs Carboplatin/
Paclitaxel in Advanced NSCLC: Responses

P = .005
RRR: 1.31
33%
25%
ITT
Socinski MA, et al. J Clin Oncol. 2012;30:2055-2062.
Carboplatin/Nab-Paclitaxel vs Carboplatin/
Paclitaxel in Advanced NSCLC: Responses
ResponseRate(%)
P < .001
RRR: 1.680
P = .808
RRR: 1.034
41%
26%
24% 25%
0
10
20
30
40
50
Squamous Nonsquamous
229 221 292 310
 Less neuropathy reported with nab-paclitaxel vs paclitaxel
521 531n =
Carboplatin/nab-paclitaxel

Squire study: Gem/Cis + Necitumumab vs Gem/Cis in
Stage IV Squamous NSCLC: OS
Thatcher N, et al. Lancet Oncol. 2015;16:763-774.
100
80
60
40
20
0
403632280 4 8 12 16
Gemcitabine/cisplatin + necitumumab
Censored pts
Gemcitabine/cisplatin
Censored pts
24
OS(%)
Pts censored, n (%)
Median OS, mos (95% CI)
Stratified P value (log-rank)
Stratified HR (95% CI)
127 (23)
11.5 (10.4-12.6)
.01
0.84 (0.74-0.96)
Gemcitabine/
Cisplatin +
Necitumumab
(n = 545)
Gemcitabine/ Cisplatin
(n = 548)
106 (19)
9.9 (8.9-11.1)
20
Mos
ECOG PS 0-2

Carboplatin and Paclitaxel ± Veliparib in Previously
Untreated Advanced NSCLC
Veliparib: A poly(ADP-ribose) polymerase inhibitor that interferes with DNA damage repair and sensitizes
tumors to radiation and chemotherapy treatments
Primary endpoint: PFS
Secondary endpoints: OS, ORR, peripheral neuropathy
Ramalingam S, et al. Chicago Multidisciplinary Symposium in Thoracic Oncology 2014. Abstract 8.
Phase II trial
Metastatic or advanced NSCLC (N = 158; ~
50 sites, 8 countries)
Carboplatin/Paclitaxel* +
Veliparib 120 mg BID† (n = 105)
Carboplatin/Paclitaxel* +
Placebo BID† (n = 53)
*Carboplatin AUC 6 mg/mL/min and paclitaxel 200 mg/m2 on Day 3 of 21.
†Veliparib/placebo on Days 1-7 of 21-day cycle.
2:1 randomization

Carboplatin and Paclitaxel ± Veliparib:
PFS, OS, ORR, DOR
*From Cox proportional hazard model, adjusting for baseline ECOG performance status and sex.
Outcome Placebo and
Carboplatin +
Paclitaxel
(n = 53)
Veliparib and
Carboplatin +
Paclitaxel
(n = 105)
HR Adjusted HR*
PFS, median mos (95% Cl) 4.2 (3.1-5.6) 5.8 (4.2-6.1) 0.74 (0.46-1.17) 0.57 (0.35-0.92)
Squamous 4.1 (2.8-NA) 6.1 (5.8-8.3) 0.50 (0.24-1.04) 0.32 (0.14-0.73)
Nonsquamous 5.0 (2.8-5.6) 4.3 (2.8-6.0) 0.94 (0.52-1.71) 0.76 (0.41-1.42)
OS, median mos (95% Cl) 9.1 (5.4-12.3) 11.7 (8.8-13.7) 0.77 (0.52-1.15) 0.71 (0.48-1.07)
Squamous 8.4 (5.0-12.9) 10.3 (8.3-13.2) 0.71 (0.41-1.23) 0.70 (0.39-1.24)
Nonsquamous 11.1 (4.8-14.6) 12.8 (8.0-17.5) 0.85 (0.48-1.51) 0.72 (0.40-1.29)
ORR, % (95% Cl) 28 (17-42) 31 (22-40) -- --
DOR, median mos (95% Cl) 3.3 (2.7-4.3) 6.9 (4.4-7.0) 0.11 (0.03-0.50) --

Veliparib Plus Carboplatin and Paclitaxel: Conclusions
 Improvements in PFS and OS were observed with the addition of
veliparib to carboplatin and paclitaxel, particularly in the squamous
histology subgroup
 Veliparib plus carboplatin and paclitaxel were well tolerated
 Based on results in the squamous histology subgroup,
a phase III pivotal trial of the addition of veliparib to carboplatin and
paclitaxel has been initiated for pts with previously untreated
advanced or metastatic squamous NSCLC (M11-089; Clinical Trial
NCT02106546)

Recommended First-line Chemotherapy Regimens for
Nonsquamous NSCLC
NCCN Guidelines. Non-Small-Cell Lung Cancer. v4.2019.
Performance Status 0/1 Performance Status 2
 Bevacizumab/carboplatin/paclitaxel
 Bevacizumab/carbo/pemetrexed
 Bevacizumab/cisplatin/pemetrexed
 Carboplatin/nab-paclitaxel
 Carboplatin/docetaxel
 Carboplatin/etoposide
 Carboplatin/gemcitabine
 Carboplatin/paclitaxel
 Carboplatin/pemetrexed
 Carboplatin/vinorelbine
 Cisplatin/docetaxel
 Cisplatin/etoposide
 Cisplatin/gemcitabine
 Cisplatin/paclitaxel
 Cisplatin/pemetrexed
 Cisplatin/vinorelbine
 Gemcitabine/docetaxel
 Gemcitabine/vinorelbine
 Nab-paclitaxel
 Carboplatin/pemetrexed
 Docetaxel
 Etoposide
 Gemcitabine
 Irinotecan
 Paclitaxel
 Pemetrexed
 Vinorelbine

Recommended First-line Chemotherapy Regimens for
Squamous NSCLC
NCCN Guidelines. Non-Small-Cell Lung Cancer. v4.2019.
Performance Status 0-1 Performance Status 2
 Cisplatin/docetaxel
 Cisplatin/etoposide
 Cisplatin/gemcitabine/necitumumab
 Cisplatin/paclitaxel
 Cisplatin/vinorelbine
 Nab-paclitaxel
 Cisplatin/gemcitabine/necitumumab
 Docetaxel
 Etoposide
 Gemcitabine
 Irinotecan
 Paclitaxel

Maintenance Therapy: Strategies
 Continuation of a doublet beyond 4 cycles
 Initiating a new agent (“switch”)
‒ Carboplatin and paclitaxel followed by pemetrexed as maintenance.
‒ Carboplatin and gemcitabine followed by docetaxel as maintenance.
‒ Platinum-based doublets followed by erlotinib as maintenance.
 Continuation of a targeted agent
‒ Carboplatin, paclitaxel and bevacizumab followed by bevacizumab as maintenance.
 Continuing 1 (or 2) of the same agents from the original combination (continuation)
‒ Cisplatin and pemetrexed followed by pemetrexed as maintenance.
‒ Carboplatin and gemcitabine followed by gemcitabine as maintenance.
‒ Cisplatin/pemetrexed and bevacizumab followed by bevacizumab and pemetrexed as maintenance.

Continuation Maintenance
Cai H, et al. Clin Lung Cancer. 2013;14:333-341.
HR: 0.54
(95% CI: 0.46-0.63;
P < .00001)
HR: 0.61
(95% CI: 0.51-0.74;
P < .00001)
HR: 0.65
(95% CI: 0.59-0.72;
P < .00001)
Meta-analysis of NSCLC Maintenance Therapy: PFS
Switch Maintenance
0.2 0.5 1 2 5
Favors Experimental Favors
Control
Brodowicz 2006
Perol 2010
Barlesi 2011
Paz-Ares 2012
Study or
Subgroup
Peto Odds Ratio Exp[O-
E/V], Fixed, 95% CI
0.2 0.5 1 2 5
Favors
Experimental
Favors
Control
Study or Subgroup
Peto Odds Ratio
Exp[0-E)/V], Fixed, 95% CI
1.1 Cytotoxic Agents
Fidias 2009
Ciuleanu 2009
1.2 Molecularly Targeted Agents
Cappuzzo 2010
Gaatar 2010
Perol 2010
Zhang 2012

HR: 0.80
(95% CI: 0.63-1.01);
P = .06)
HR: 0.81
(95% CI: 0.71-0.92);
P = .001)Total
HR: 0.80 (95% CI: 0.72-0.92);
P =.0002)
7 trials report no detrimental effect on QOL
Meta-analysis of NSCLC Maintenance Therapy: OS
Cai H, et al. Clin Lung Cancer. 2013;14:333-341.
Continuation Maintenance
0.2 0.5 1 2 5
Favors Experimental Favors
Control
Brodowicz 2006
Perol 2010
Belani 2010
Barlesi 2011
Paz-Ares 2012
Study or
Subgroup
Peto Odds
Ratio Exp[O-E/V],
Fixed, 95% CI
HR: 0.82
(95% CI: 0.66-1.01;
P = .06)
Switch Maintenance
0.2 0.5 1 2 5
Favors Switch
Maintenance
Favors
Control
Study or Subgroup
Peto Odds Ratio
Exp[0-E)/V], Fixed, 95% CI
1.1 Cytotoxic Agents
Ciuleanu 2009
Fidias 2009
1.2 Molecularly Targeted Agents
Cappuzzo 2010
Gaatar 2010
Perol 2010
Zhang 2012

Do you think maintenance therapy is worthwhile?
Do you think maintenance therapy is worthwhile depending on the
magnitude of the survival benefit?
Do you think maintenance therapy is worthwhile if there was no survival
benefit but symptom control benefit?
Do you think maintenance therapy is worthwhile depending on
mode of administration?
Peeters L, et al. J Thorac Oncol. 2012;7:1291-1295.
75% of patients would take maintenance therapy
for mild to moderate toxicity
Patient Perception of NSCLC Maintenance
No
Unsure
Yes
30
25
20
15
10
5
0
0 2 4
30
25
20
15
10
5
0
0 2 4 0 2 4
Symptom relief Tumor control
No
Unsure
Yes
Patients(n)
Cycles of First-line Therapy
0 2 4 0 2 4
No
Unsure
Yes
30
25
20
15
10
5
0
0 2 4 0 2 4 0 2 4 0 2 4
30
25
20
15
10
5
0
No
Unsure
Yes
1-yr OS
benefit
6-mo OS
benefit
3-mo OS
benefit
1-mo OS
benefit
Oral ODIV q3w
Patients(n)
Patients(n)Patients(n)

NSCLC Maintenance Therapy:
Advantages and Disadvantages
Advantages
 Maintains disease control
 Improves PFS
 Improves OS
 Maintains quality of life
 Opportunity to treat more patients
 Patients support maintenance therapy
Disadvantages
 Induction regimens of 4 vs 6 cycles may
achieve the same improvement in PFS
 Careful follow-up reveals more patients
available for second-line therapy than
initially estimated by early reports
 Grade 3/4 AEs in 30% to 40% of patients
 Cost prohibitive
 Lack of reliable biomarkers

Maintenance Pemetrexed vs Placebo in Stage IIIB/IV
NSCLC: OS by Histology
Median OS pemetrexed:
15.5 mos
Median OS pemetrexed:
9.9 mos
Median OS placebo:
10.3 mos
Median OS placebo:
10.8 mos
Nonsquamous (n = 481) Squamous (n = 182)
HR: .70 (95% CI: 0.56-0.88; P = .002) HR: 1.07 (95% CI: 0.49-1.73; P = .678)
SurvivalProbability
Mos
Belani CP, et al. ASCO 2009. Abstract 8000.
Pts with stage IIIB or IV NSCLC and ECOG PS 0-1; All treated with initial therapy (gemcitabine, docetaxel, or paclitaxel +
cisplatin or carboplatin) for four 21-day cycles (N = 663)
0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
SurvivalProbability
Mos
0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0

OS by Histology in Phase III Pemetrexed Studies
*Adenocarcinoma, large cell carcinoma, or other/indeterminate NSCLC histology
1. Scagliotti GV, et al. Oncologist. 2009;14:253-263. 2. Pemetrexed [package insert]. 2009.
NSCLC Histology Second-line
Pem vs Docetaxel[1]
First-line Pem/Cis
vs Gem/Cis[1]
Maintenance
Pem vs Placebo[2]
Pem Doc Cis/Pem Cis/Gem Pem Placebo
Nonsquamous,* n 205 194 618 634 325 156
Median OS, mos 9.3 8.0 11.0 10.1 15.5 10.3
Adjusted HR
(95% CI; P value)
0.78 (0.61-1.00;
.048)
0.84 (0.74-0.96;
.011)
0.70 (0.56-0.88;
.002)
Squamous, n 78 94 244 229 116 66
Median OS, mos 6.2 7.4 9.4 10.8 9.9 10.8
Adjusted HR
(95% CI; P value)
1.56 (1.08-2.26;
.018)
1.23 (1.00-1.51;
.050)
1.07 (0.77-1.50;
.678)

Heterogeneity of ERCC1, RRM1, and TS mRNA Expression in NSCLC
Maus MKH, et al. J Thorac Oncol. 2013;8:582-586.
ERCC1 RRM1 TS
*Gene expression level cutoff for drug sensitivity.
GeneExpressionLevel
Relativetoβ-actin
10
8
6
4
2
0
AC SCCA
10
8
6
4
2
0
AC SCCA
10
8
6
4
2
0
AC SCCA
ERCC1
(Reference < 1.7 for
platinum)*
% Below
Reference Level
NSCLC-total 43.4
NSCLC-AC 46.0
NSCLC-SCCA 30.7
RRM1
gemcitabine)*
% Below
Reference Level
NSCLC-total 39.6
NSCLC-AC 42.2
NSCLC-SCCA 13.0
TS
pemetrexed)*
% Below
Reference Level
NSCLC-total 41.3
NSCLC-AC 45.7
NSCLC-SCCA 25.9

Maintenance Pemetrexed vs Placebo in Stage IIIB/IV
NSCLC: AEs
*NCI CTC version 3.0.
†P < .05 for grade 3/4 neutropenia and fatigue.
Ciuleanu T, et al. Lancet. 2009;374:1432-1440.
Grade 3/4 AEs,* % Pemetrexed
(n = 441)
Placebo
(n = 222)
Grade 3/4 Grade 3/4
Neutropenia† 3 0
Anemia 3 < 1
Leukopenia 2 < 1
Fatigue† 5 < 1
Anorexia 2 0
Infection 2 0
Diarrhea < 1 0
Nausea < 1 < 1
Vomiting < 1 0
Sensory neuropathy < 1 0
Mucositis/stomatitis < 1 0

PRONOUNCE: Phase III Trial of Pem/Carbo → Pem vs
Pac/Carbo/Bev → Bev
 No significant difference in survival
 Carbo/pem → pem: more anemia and thrombocytopenia
 Carbo/pac/bev → bev: more neutropenia and peripheral neuropathy, plus total alopecia
Induction (Q3W, 4 cycles) Maintenance (Q3W until PD)
180 pts each
Bev-Eligible Population
Inclusion:
 Chemo-naive pts
 PS 0/1
 Stage IV, nonsquamous
 Stable treated CNS mets
Exclusion:
 Uncontrolled effusions
Zinner RG, et al. J Thorac Oncol. 2015;10:134-142.
Pemetrexed
(folic acid and vitamin B12)
+ Carboplatin
Paclitaxel
+ Carboplatin
+ Bevacizumab
R
1:1
Pemetrexed
(folic acid and vitamin B12)
Bevacizumab

CALGB 30607: Sunitinib as Switch Maintenance in
Advanced NSCLC
Stratification factors:
 ECOG PS (0 vs 1)
 Stage (IIIB vs IV)
 Prior use of bevacizumab (yes or no)
 Sex (male vs female)
Primary endpoint:
 PFS
Secondary endpoints:
 Toxicity
 Additional responses
 Impact on OS
 QoL
1:1
Chemo-naive
advanced NSCLC
ECOG PS 0-1
Non-PD
4 cycles of
1st-line platinum-
based doublet*
Placebo
(n = 104)
PD
Sunitinib
37.5 mg/day
(n = 106)
PD
Socinski MA, et al. Chicago Multidisciplinary Symposium in Thoracic Oncology 2014. Abstract 1.
*Bevacizumab allowed but was discontinued with the fourth
cycle.

CALGB 30607 Sunitinib Switch Maintenance: PFS and OS
Median OS, mos (95% CI): sunitinib 11.7 (9.4-15.0); placebo 11.7 (9.9-14.0)
1.0
0.8
0.6
0.4
0.2
0
SurvivalProbability
Mos From Randomization
0 306 12 18 24
2-sided log-rank P = .0005
Pts at Risk, n
Sunitinib
Placebo
106
104
32
17
9
3
8
1
2
0
1
Median PFS (95% CI)
4.3 (3.2-4.9)
2.6 (1.8-3.0)
HR (95% CI)
0.61 (0.46-0.81)
Ref
Sunitinib
Placebo

CALGB 30607 Sunitinib Switch Maintenance:
Conclusions
 First trial evaluating an anti-angiogenic agent as maintenance therapy
in advanced NSCLC
 Sunitinib as switch maintenance in advanced NSCLC associated with
significant PFS improvement (HR: 0.61;
P = .0005)
‒ Effect independent of histology
 No significant improvement in OS (secondary endpoint)
 No new toxicity signals identified; toxicity profile similar to previous
studies of sunitinib

LCT in Oligometastatic NSCLC: Randomized Phase II
Study Design
 Primary outcome: PFS
‒ Time from randomization to PD or death if endpoint met, time from randomization to last imaging if no progression, and censored at time of
crossover due to toxicity
‒ Follow-up every 6-10 wks in Yr 1, then at physician’s discretion
Histologically confirmed
stage IV NSCLC, no
RECIST progression after
FLST, ≤ 3 metastases after
FLST
(n = 49)
No Local Consolidative Therapy (n = 24)
Local Consolidative Therapy (n = 25)
Slide credit: clinicaloptions.comGomez DR, et al. ASCO 2016. Abstract 9004.
Stratified by nodal status (NO/N1 vs N2/N3), EGFR/EML4-ALK status (yes/no), CNS metastases (yes/no), number of
metastases (1 vs 2/3), and response to FLST (SD vs PR/CR)
FLST*
(N = 74)
Standard
maintenance or
observation† PD/Toxicity
Consider LCT
(surgery ± radiation)
Crossover allowed
at PD
LCT
(surgery ±
radiation)
Standard
maintenance or
observation†
PD
*≥ 4 cycles of platinum-doublet chemotherapy, or ≥ 3 mos of afatinib, erlotinib, or gefitinib in setting of EGFR mutation,
or ≥ 3 mos of crizotinib if EML4-ALK fusion; pts could enroll before or during FLST, but randomization occurred after completion of FLST without progression.
†Physician choice.

LCT in Oligometastatic NSCLC: PFS
 Study closed by DSMC due to observed efficacy in the LCT arm after 49 pts randomized, at median follow-up of
18.7 mos
 Other factors associated with
improved PFS included number of
metastases after FLST
(1 vs 2 or 3; P = .043) and EGFR/ALK
status (positive vs negative;
P = 0.035)
 PFS benefit remained after censoring
pts with EGFR/ALK alterations
(HR: 0.41; 95% CI: 0.19-0.90; P = 0.022)
Gomez DR, et al. ASCO 2016. Abstract 9004. Reproduced with permission.
0
25
50
75
100
0 1 2 3
24
24
8
2
2
0
0
0
Pts at Risk, n
LCT
No LCT
Yrs
PFS(%)
LCT No LCT
Median
PFS, Mos
11.9 3.9
P value .005

LCT in Oligometastatic NSCLC: Patterns of Failure
 Progression in 30 pts (61%), most often in new lesions (n = 14) or distant metastases (n = 16);
progression at combination of new and known sites also common (n = 9)
 LCT arm had primarily distant and new lesion failures; no-LCT arm had more locoregional failures and
combined new and known site failures
 Trend toward significance in patterns of failure (P = .09)
‒ Locoregional-only failures higher in no-LCT arm vs LCT arm (17% vs 4%)
‒ Metastatic-only failures higher in LCT arm vs no-LCT arm (40% vs 25%)
‒ Both locoregional/metastatic failures higher in no-LCT vs LCT arm (29% vs 8%)
 Median TNSF: 11.9 mos in LCT arm vs 5.7 mos in no-LCT arm
(P = .0497)
Gomez DR, et al. ASCO 2016. Abstract 9004.

Eribulin vs Physician’s Choice in Pts With Progressive
NSCLC
 Open-label, parallel-group phase III study in pts with advanced NSCLC
progressed on ≥ 2 previous regimens
 Randomized 1:1 to 21-day cycle eribulin mesylate IV
1.4 mg/m2 on Days 1, 8 (n = 270) or the physician’s choice (21-day
cycles of vinorelbine, gemcitabine, pemetrexed [nonsquamous only], or
docetaxel; n = 270)
 Primary endpoint: OS
 Secondary endpoints: PFS, ORR, safety, tolerability
Spigel DR, et al. Chicago Multidisciplinary Symposium in Thoracic Oncology 2014. Abstract LB1.

Eribulin vs Physician’s Choice: Results
 Median OS 9.5 mos in both eribulin and TPC groups
(HR: 1.16; 95% CI: 0.95-1.41; P = .134)
 Median PFS 3.0 mos with eribulin vs 2.8 mos with TPC (HR: 1.09; 95%
CI: 0.90-1.32; P = .395)
 ORR: 12.2% and 15.2%, respectively
 Most frequent grade 3/4 AEs with eribulin: neutropenia (28.6%),
decreased neutrophil count (21.2%), decreased WBC count (13.4%),
and asthenia (8.2%)
 33.9% of pts had a serious AE (35.7% eribulin, 32.1% TPC)

Eribulin vs Physician’s Choice: Conclusions
 Eribulin did not significantly improve OS or PFS compared with TPC in
pts with advanced NSCLC
 AEs were manageable and consistent with previous eribulin studies

100
80
60
40
20
0
REVEL: Docetaxel ± Ramucirumab in Second-line NSCLC:
Response
 First study in the second-line setting to show survival advantage by adding biotherapy to chemotherapy
in NSCLC
 First and only angiogenesis inhibitor in advanced NSCLC to show benefit in squamous
Phase III Study
Perol M, et al. ASCO 2014. Abstract LBA8006.
0 3 6 9 12 15 18 21 24 27 30 33 36
Survival Time (Mos)
OS(%)
Ram + Doc
Pl + Doc
Ram + Doc Pl + Doc HR
P Value
ORR, %
(95% CI)
22.9
(19.7-26.4)
13.6
(11.0-16.5)
< .001
Median PFS, mos
(95% CI)
4.5
(4.2-5.4)
3.0
(2.8-3.9)
0.72
< .0001
Median OS, mos
(95% CI)
10.5
(9.5-11.2)
9.1
(8.4-10.0)
0.86
.0235

REVEL: Efficacy
 ORR: 23% vs 14% (P < .0001)
 DCR: 64% vs 53% (P < .0001)
 PFS HR: 0.76 (P < .0001)
Garon EB, et al. Lancet. 2014;384:665-673.
0 3 6 9 12 15 18 21 24 27 30 33 36
0
20
40
60
80
100
OS(%)
Mos
Ram + doc
Pbo + doc
Ram + doc
Pbo + doc
HR: 0.857 (95% CI: 0.759-0.979; P = .0235)
10.5 (9.5-11.2)
9.1 (8.4-10.0)
Median OS, Mos (95% CI)

REVEL: Adverse Events
AEs, %
Docetaxel + Ramucirumab
Overall
(n = 627)
Docetaxel Overall
(n = 618)
Any AEs
 Grade ≥ 3 AEs
• Fatigue
• Stomatitis
• Neutropenia
• Febrile neutropenia
98
79
14
4
49
16
95
71
10
2
40
10
Serious AEs 43 42
AEs of special interest
 Bleeding/hemorrhage
• Epistaxis
 Hypertension
Any Grade
29
19
11
Grade ≥ 3
2
< 1
6
Any Grade
15
7
5
Grade ≥ 3
2
< 1
2
Garon EB, et al. Lancet. 2014;384:665-673.

Targeted Therapy for Treatment
of Metastatic (Advanced) NSCLC

Li T, et al. J Clin Oncol. 2013;31:1039-1049.
Evolution of NSCLC Subtyping to a Multitude of
Molecular-Defined Subsets
NSCLC
as 1 disease
Histology-Based Subtyping
Squamous
34%
Other
11%
Adenoca
55%
Adenocarcinoma
Squamous Cell Cancer
ALK
HER2
BRAF
PIK3CA
AKT1
MAP2K1
NRAS
ROS1
RET
EGFR
KRAS
Unknown
EGFRvIII
PI3KCA
EGFR
DDR2
FGFR1 Amp
Unknown
First-targeted tx
ALK
EGFR

NSCLC Adenocarcinoma: Beyond EGFR Mutations and
ALK Translocation
EGFR KRAS
MET amp
(2.2%)
ERBB2 amp
(0.9%)
Govindan R. ISLAC 2013. Abstract PL05.1.
EGFR
BRAF
KRAS
None
ERBB2 (1.7%)
HRAS (0.4%)
NRAS (0.4%)
RET fusion (0.9%)
MAP2K1 (0.9%)
ALK fusion (1.3%)
ROS1 fusion (1.7%)
NF1
MET amp
(2.2%)
ERBB2 amp
(0.9%) RIT1
(2.2%)
8.3%
4.3%
7.0%
MET ex14
11.3%
32.2%
24.4%

47%
23%
12%
13%
Targetable Mutations in Lung Adenocarcinomas
Sholl. J Thorac Oncol. 2015;10:768. Meza. PLoS One. 2015;10:e0121323. Howlander. SEER Cancer Statistics Review, 1975-2014.
Histology-Based Subtyping
of Lung Cancer
Adenocarcinoma
Squamous and transitional cell carcinoma
Other non-small-cell carcinomas
Small-cell carcinoma
Other
Large-cell carcinoma
KRAS
(25%)
EGFR
(23%)
ALK
(7.9%)
ERBB2
(2.7%)
PIK3CA (0.8%)
No oncogenic
driver identified
(36%)
BRAF
(2.6%)
3%
2%
NRAS (0.7%)
MET (0.7%)
MEK1 (0.3%)
Adenocarcinoma:
The Lung Cancer
Mutation
Consortium
Experience

Targeted Therapy Focuses on Driver Gene Alterations:
“Oncogenic Addiction”
EGFR mutants ALK ROS/RET
1. Maemondo M, et al. N Engl J Med. 2010;362:2380-2388. 2. Mitsudomi T, et al. Lancet Oncol. 2010;11:121-128. 3. Rosell R, et al. Lancet
Oncol. 2012;13:239-246. 4. Zhou C, et al. Lancet Oncol. 2011;12:735-742. 5. Sequist LV, et al. J Clin Oncol. 2013;31:3327-3334. 6. Wu YL, et
al. Lancel Oncol. 2014;15:213-222. 7. Camidge DR, et al. Lancet Oncol 2012;10:1011-1019. 8. Kim DW, et al. ASCO 2012. Abstract 7533. 9.
Shaw AT, et al. N Engl J Med 2013;368:2385-2394.
Gefitinib[1,2] Erlotinib[3,4] Afatinib[5,6] Crizotinib[7-9]
Activity EGFR EGFR
EGFR
(ErbB family)
ALK, ROS1, MET
Target EGFR ALK
RR, % 60-80 50-80 ~ 60 ~ 60
PFS, mos 10-11 10-14 ~ 11 ~ 10
TRD, % 1~2 1~2 1.7 <1

Johnson B, et al. ASCO 2013. Abstract 8019.
Lung Cancer Mutation Consortium:
OS by Mutation and Treatment
Driver mutation + targeted therapy (n = 313)
Driver mutation + no targeted therapy (n = 265)
No driver mutation (n = 361)
100
80
60
40
20
0
OS(%)
0 1 2 3 4 5
Yrs

Alterations in Targetable Oncogenic Pathways in
Squamous Cell NSCLC
 Somatic mutations
 Homozygous deletions
 High-level focal
amplification
 Upregulation/
downregulation of gene
expression
Cancer Genome Atlas Network. Nature. 2012;489:519-525. Reprinted by permission from Macmillan
Publishers Ltd: Copyright 2012.

Treatment of EGFR Mutation–Positive NSCLC

EGFR Structure and Mutations
 EGFR: a tyrosine kinase
receptor
 Overexpression in NSCLC
 Increased gene copy number
 EGFR-activating mutations
have been identified
 Most common known cause
of acquired resistance to first-
and second-line EGFR TKIs is a
T790M secondary mutation
Lynch. NEJM. 2004;350:2129. Paez. Science. 2004;304:1497. Herbst. NEJM. 2008;359:136.

EGFR Mutational Epidemiology
 Found in ~ 10% of NSCLC
patients in the US
 More common in never-
smokers, adenocarcinomas,
females, Asians
 Predominantly located in
EGFR exons 18-21
 The specific EGFR mutation
identified is important:
sensitive mutations, primary
resistance mutations, and
de novo and acquired
resistance mutations
Irmer. Oncogene. 2007;26:5693. Pao. J Clin Oncol. 2005;23:2556. Wu. J Thorac Oncol. 2007;2:430.
Fang. Drug Des Devel Ther. 2014;8:1595. Shea. Ther Adv Respir Dis. 2016;10:113. Wang. Onco Targets Ther. 2016;9:3711.
EGFR Kinase Domain Mutations
Ligand Binding Transmembrane Tyrosine Kinase Autophosphorylation
N
N
K754R S768I*
L861Q*
A871G
L833V/
H835L/
L838V
E884K
L858R
~ 41%
Ins761 (EAFQ)/
Ins770 (ASV)/
Ins771 (G)/
Ins774 (NPH)
~ 3%
G719S*
~ 5%
E709A/
E709G
C
C
Y891
Y920
Y992
Y1045
Y1068
Y1086
Y1148
Y1173
T790M
~ 3%
EXON 18 19 20 21 22 23 24
del 747-752
and others
~ 48%
*Noncanonical EGFR mutations.

EGFR Mutations: Context
 Found in ~ 10% of NSCLC patients in the US[1]
 More common in never-smokers, adenocarcinomas, females, Asians[2]
 Predominantly located in EGFR exons 18-21[3]
‒ 85% to 90% of EGFR mutations are either deletions in exon 19 or a single
point mutation in exon 21 (L858R)
 The specific EGFR mutation identified is important
‒ There are sensitive mutations, primary resistance mutations (often exon
20), and de novo and acquired resistance mutations (T790M)[2-4]
1. Graham. Arch Pathol Lab Med. 2018;142:163. 2. Wang. Onco Targets Ther. 2016;9:3711.
3. Fang. Drug Des Devel Ther. 2014;8:1595. 4. Morgillo. ESMO Open. 2016;1:e000060.

ARCHER 1050:
1st-line dacomitinib
better than gefitinib
EGFR mutations sensitive to
erlotinib and gefitinib discovered
in lung adenocarcinoma
Evolution of Care: EGFR Mutation–Positive Advanced
NSCLC
EGFR inhibitors
enter clinical
development
IPASS:
1st-line gefitinib
better than CT
OPTIMAL:
1st-line erlotinib
better than CT
Mutation testing
included in
diagnostic workup
1997 2004 2009 2010 2013 2015
AURA: 2nd-line
osimertinib better than CT
against EGFR T790M
2018
Herbst. Nature. 2018;553:446. Mok. NEJM. 2009;361:947. Zhou. Lancet Oncol. 2011;12:735. Sequist. J Clin Oncol. 2013;31:3327. Yang. Lancet
Oncol. 2015;16:141. Wu. Lancet Oncol. 2014;15:213.Soria. NEJM. 2018;378:113. Wu. Lancet Oncol. 2017;18:1454. Mok. NEJM. 2017;376:1993.
Drug
Approvals/
Other
Landmarks
Select
Trials
Erlotinib:
1st line
Gefitinib:
1st line
Osimertinib:
2nd line
T790M+
with plasma
ctDNA test
Osimertinib:
1st line
FLAURA:
1st-line
osimertinib
better than
1st-gen TKI
2017
LUX-Lung 3 & 6:
1st-line afatinib
better than CT
Dacomitinib:
1st line
2016
Afatinib:
1st line

EGFR Tyrosine Kinase Inhibitors
Agent Indication Inhibition
First-generation TKI
 Gefitinib
 Erlotinib
First-line therapy Reversible
Second-generation TKI
 Afatinib
 Dacomitinib
First-line therapy Irreversible
Third-generation TKI
 Osimertinib
First-line therapy;
Second-line therapy for T790M+ NSCLC
Irreversible
Hirsh. Ther Adv Med Onc. 2018;10:1758834017753338.

EGFR TKIs: Structures
Erlotinib
 First generation
 FDA approved: 2011
Afatinib PI. Dacomitinib PI. Erlotinib PI. Gefitinib PI. Osimertinib PI.
Gefitinib
 First generation
Afatinib
 Second generation
Osimertinib
 Third generation
O
N
HN
N
N
F
O
O
CI
F
CI NH
N
N
H
N
COOH
COOH
COOH
COOH
CH3
CH3N
OO
O
N
N
N
N N
N
H O
O NH
O
O
OH
S
HN
N
N
O
O
O
O
HCI
Dacomitinib
 Second generation
O
H3CO
N
HN
HN
N
N
F
CI
H2O

EGFR TKIs: Targeted Therapies for EGFR Mutations
Afatinib PI. Erlotinib PI. Gefitinib PI. Osimertinib PI. Lin. Clin Lung Cancer. 2017;18:324. Morgillo. ESMO Open. 2016;1:e000060.
FDA-Approved EGFR TKI
Location of EGFR Mutation
Exon 18 Exon 19 Exon 20 Exon 21
Erlotinib
Gefitinib
Afatinib
-- Deletion -- L858R
Dacomitinib -- Deletion -- L858R
Osimertinib -- Deletion T790M L858R
Justified use (on/off label) of erlotinib,
gefitinib, afatinib, osimertinib
G719X* Insertion
A763_Y764insFQEA
S768I*
L861Q*
Mutations causing EGFR TKI
insensitivity
-- --
Insertion
C797S
T790M
--
*Afatinib approved with these mutations alone or in combination.

Parameter Erlotinib Gefitinib Afatinib Osimertinib Dacomitinib
Receptor
binding
EGFR/HER1,*
SRC, ABL?
EGFR/HER1,*
IGF, PDGF
EGFR/HER1,*
HER2, HER4
EGFR/HER1,*
HER2, HER3,
HER4, BLK,
ACK1
EGFR/HER1,*
HER2, HER4
EGFR binding Reversible Reversible Irreversible Irreversible Irreversible
Half-life, hrs 36 48 37 48 59-85
Food effect
(take on empty
stomach)
Increase F from
~ 60% to ~ 100%
No change Decrease
AUC by 39%
No change No change
CNS
penetration,
AUC ratio
0.03X
CSF/plasma
0.01X
CSF/serum
0.02X
CSF/plasma
2X
Brain/plasma
Data not
available
EGFR TKIs: Properties
*All inhibit exon 19 deletion and L858R.
Afatinib PI. Erlotinib PI. Gefitinib PI. Osimertinib PI. Boehrer. Cell Cycle. 2011;10:3168. Togashi. Cancer Chemother Pharmacol. 2012;70:399.
Tamiya. ESMO 2016. Abstract 1241P. Engelman. Cancer Res. 2007;67:11924. Gonzales. Mol Cancer Ther. 2008;7:1880. Jänne. Clin Cancer Res.
2011;17:1131. Ou. Drug Des Devel Ther. 2015;9:5641. Hochmair. Target Oncol. 2018;13:269.

First-line EGFR TKIs vs Chemotherapy in
EGFR Mutation–Positive NSCLC: A Clear Pattern
1. Maemondo. NEJM. 2010;362:2380. 2. Mitsudomi. Lancet Oncol. 2010;11:121. 3. Yoshioka. ASCO 2014. Abstr 8117. 4. Zhou. Lancet Oncol.
2011;12:735-. 5. Zhou. Ann Oncol. 2015;26:1877. 6. Rosell. Lancet Oncol. 2012;13:239. 7. Khozin. Oncologist. 2014;19:774. 8. Sequist. J Clin
Oncol. 2013;31:3327. 9. Yang. Lancet Oncol. 2015;16:141. 10. Wu. Lancet Oncol. 2014;15:213.
Study N Treatment ORR, % Median PFS, Mos Median OS, Mos
NEJ002[1] 230
Gefitinib vs
carboplatin/paclitaxel
74 vs 31
10.8 vs 5.4
(P < .001)
30.5 vs 23.6
(HR: 0.89)
WJTOG
3405[2,3] 172
Gefitinib vs
cisplatin/docetaxel
62 vs 32
9.6 vs 6.6
(P < .001)
34.8 vs 37.3
(HR: 1.25)
OPTIMAL[4,5] 165
Erlotinib vs
carboplatin/gemcitabine
83 vs 36
13.1 vs 4.6
(P < .0001)
22.8 vs 27.2
(HR: 1.19)
EURTAC[6,7] 174
Erlotinib vs platinum-
based chemotherapy
58 vs 15
9.7 vs 5.2
(P < .0001)
22.9 vs 19.5
(HR: 0.93)
LUX-Lung 3[8,9] 345
Afatanib vs
cisplatin/pemetrexed
56 vs 23
11.1 vs 6.9
(P = .001)
28.2 vs 28.2
(HR: 0.88)
LUX-Lung 6[9,10] 364
Afatinib vs
cisplatin/gemcitabine
67 vs 23
11.0 vs 5.6
(P < .0001)
23.1 vs 23.5
(HR: 0.93)

First-line Treatment With EGFR TKIs in EGFR-Mutated
NSCLC
1. Maemondo. NEJM. 2010;362:2380. 2. Mitsudomi. Lancet Oncol. 2010;11:121. 3. Yoshioka. ASCO 2014. Abstr 8117.
4. Zhou C. Lancet Oncol. 2011;12:735. 5. Rosell. Lancet Oncol. 2012;13:239. 6. Sequist. J Clin Oncol. 2013;31:3327.
7. Wu. Lancet Oncol. 2014;15:213. 8. Park. Lancet Oncol. 2016;17:577. 9. Soria. NEJM. 2018;378:113. 10. Wu. Lancet Oncol. 2017;18:1454.
Agent/Study N Control Arm ORR, % Median PFS, Mos
Gefitinib
 NEJ002[1]
 WJTOG 3405[2,3]
230
172
Cisplatin/docetaxel
74 vs 31
62 vs 32
10.8 vs 5.4
9.6 vs 6.6
Erlotinib
 OPTIMAL[4]
 EURTAC[5]
165
174
Carboplatin/gemcitabine
Plt-based chemotherapy
83 vs 36
58 vs 15
13.1 vs 4.6
9.4 vs 5.2
Afatinib
 LUX-Lung 3[6]
 LUX-Lung 6[7]
 LUX-Lung 7[8]
345
364
319
Cisplatin/pemetrexed
Gefitinib
56 vs 23
67 vs 23
70 vs 56
11.1 vs 6.9
11.0 vs 5.6
11.0 vs 10.9
Osimertinib
 FLAURA[9] 556 Erlotinib or gefitinib 80 vs 76 18.9 vs 10.2
Dacomitinib
 ARCHER 1050[10] 452 Gefitinib 75 vs 72 14.7 vs 9.2

Comparison of First-line Options for EGFR-Positive
NSCLC in 2019
*Not approved.
Agent Comparator Agent PFS QoL OS Toxicity CNS
Osimertinib[1] Gefitinib/erlotinib +++ +++ Pending Less +++
Dacomitinib[2-4] Gefitinib +++ Pending + More ++
Gefitinib +
plt-based CT[5]*
Gefitinib +++ NR ++ More NR
Erlotinib +
bevacizumab[6,7]*
Erlotinib +++ NR = More NR
Afatinib[8,9] Gefitinib +/- = = More ++
Gefitinib[10,11] Plt-based CT +++ +++ = Less +
Erlotinib[12-15] Plt-based CT +++ NR = Less +
1. Soira. NEJM. 2018 378:113. 2. Wu. Lancet Onc. 2017;18:1454. 3. Mok. ASCO 2018. Abstr 9004. 4. Mok. J Clin Oncol. 2018;36:2244. 5.
Nakamura. ASCO 2018. Abstr 9005. 6. Seto. Lancet Oncol. 2014;15:1236. 7. Saito. Lancet Oncol. 2019;20:625. 8. Paz-Ares. Ann Oncol.
2017;28:270. 9. Park. Lancet Oncol. 2016;17:577. 10. Maemondo. NEJM. 2010;362:2380. 11. Mitsudomi. Lancet Oncol. 2010;11:121. 12. Zhou.
Lancet Oncol. 2011;12:735. 13. Zhou. Ann Oncol. 2015;26:1877. 14. Rosell. Lancet Oncol. 2012;13:239. 15. Khozin. Oncologist. 2014;19:774.

EGFR-Mutated NSCLC: First-line Treatment
Considerations
 Head-to-head randomized FLAURA trial (N = 556): superior PFS with first-line
osimertinib vs erlotinib/gefitinib (18.9 vs 10.2 mos, respectively)[1]
 Cost savings could be significant with sequential treatment
‒ Osimertinib: $584 per dose[2]
‒ Erlotinib: $338 per dose[2]
 Using erlotinib first followed by osimertinib seems just as good BUT
‒ Not everyone will have a T790M-acquired mutation in second line
‒ In FLAURA, only 46% of the standard-treatment group received a second-line
EGFR TKI–containing regimen[1]
‒ In FLAURA, 12% and 17% died before second-line therapy[1]
1. Soria. NEJM. 2018;378:113. 2. Medi-Span Price Rx. February 2019. Available at: https://www.wolterskluwercdi.com/price-rx.

 The landscape of NSCLC continues to evolve
‒ 47% of all lung cancer is adenocarcinoma, with 23% having EGFR
sensitizing mutations; represents ~ 10% of all advanced NSCLC in US
 Liquid biopsy is approved for detection of EGFR mutations
‒ Easier than standard tissue biopsy, with the potential to better capture
tumor heterogeneity with further optimization
 Third-generation EGFR TKIs (eg, osimertinib) are more specific for
mutated EGFR, associated with less toxicity
‒ Better CNS penetration
 Consider treatment beyond progression

What About Sequencing?
Erlotinib/gefitinib 1st line
10.2 mos
Erlotinib/gefitinib 1st line
10.2 mos
Osimertinib 1st line
18.9 mos
Osimertinib 2nd line
10.1 mos
T790M-
(~ 40%)
T790M+
(~ 60%)
All
. . . resistance inevitably develops
Yu. Clin Cancer Res. 2013;19:2240. Soria. NEJM. 2018;378:113. Mok. NEJM 2017; 376:629.

EGFR Mutation–Positive NSCLC: Current Treatment
Paradigm and First-line Approvals
 Approved for EGFR exon 19
deletions and exon 21 L858R
point mutation
‒ Afatinib, dacomitinib, erlotinib,
gefitinib, and osimertinib
 Approved for EGFR point
mutations G719X, S768I, and
L861Q
‒ Afatinib
Progression
Melosky. Clin Lung Cancer. 2018;19:42. Erlotinib PI. Gefitinib PI. Afatinib PI. Osimertinib PI. Dacomitinib PI.
EGFR mutation positive
Follow treatment options for
adenocarcinoma or squamous cell
carcinoma without actionable biomarker
Osimertinib
EGFR T790M
mutation negative or
previous osimertinib
Osimertinib (preferred),
afatinib, dacomitinib, erlotinib,
or gefitinib
EGFR T790M
mutation positive

Acquired Resistance to EGFR TKI: Proposed Approach to
EGFR T790M Mutation Genotyping
Oxnard. JCO. 2016;34:3375.
Conventional Approach for T790M Genotyping
Proposed Approach for T790M Genotyping
All patients undergo
biopsy, FDA-approved
FFPE assay for T790M
T790M+
T790M-
Third-generation EGFR TKI
Chemotherapy
FDA-approved plasma
assay for T790M and
sensitizing mutations
T790M+
T790M-
Skip biopsy, start third-
generation EGFR TKI
Biopsy, FDA-approved
FFPE assay for T790M
T790M+
T790M-
Third-generation
EGFR TKI
Chemotherapy

Lee CK, et al. J Natl Cancer Inst. 2013;105:595-605
Favors EGFR TKI Favors Chemo
Meta-analysis of Randomized First-line EGFR TKI Studies:
Improved PFS
Study
HR
(95% CI)
HR
(95% CI)
EGFRmut (first-line therapy)
EURTAC
First-SIGNAL
GTOWG
INTACT1-2
IPASS
LUX LUNG3
NEJ002
OPTIMAL
TALENT
TOPICAL
TRIBUTE
WJTOG3405
Subtotal
0.37 (0.25-0.54)
0.54 (0.27-1.10)
1.08 (0.24-4.90)
0.55 (0.19-1.60)
0.48 (0.36-0.64)
0.58 (0.43-0.78)
0.32 (0.24-0.44)
0.16 (0.11-0.26)
0.59 (0.21-1.67)
0.90 (0.39-2.06)
0.49 (0.20-1.20)
0.52 (0.38-0.72)
0.43 (0.38-0.49)

Improved QoL With First-line EGFR TKI for EGFR Mutation–
Positive NSCLC
 IPASS[1]: Gefitinib vs plt-based doublet chemotherapy showed improvement with FACT-L
 NEJ002[2]: Gefitinib vs plt-based doublet chemotherapy showed improvement assessed with Care
Notebook
 First Signal: Gefitinib vs plt-based doublet chemotherapy showed improvement assessed with EORTC
QoL C30 and Lung Cancer-13 questionnaires
 OPTIMAL[4]: Erlotinib vs plt-based doublet chemotherapy showed improvement in FACT-L and LCS
scores
 LUX-Lung-3[5] : Afatinib vs plt-based doublet chemotherapy showed statistically significant delay in
time to deterioration of cough, dyspnea; improvement in dyspnea scores, cognitive and physical role
functions assessed by EORTC QoL C30 and Lung Cancer-13 questionnaires
1. Thongprasert S, et al. J Thorac Oncol. 2011;6:1872-1180.
2. Oizumi S, et al. Oncologist. 2012;17:863-870.
3. Han JY, et al. J Clin Oncol. 2012;30:1122-1128.
4. Chen G, et al. Ann Oncol. 2013;24:1615-1622.
5. Yang JC, J Clin Oncol. 2013;31:3342-3350.

EGFR mutation-positive EGFR mutation-negative

Mok. NEJM. 2009;361:947.
IPASS: First-line Gefitinib vs Carboplatin/Paclitaxel in
Advanced NSCLC
 Open-label phase III trial conducted in Asian countries
 Primary endpoint: PFS
 Secondary endpoints: OS, ORR, QoL, symptom reduction, safety
 Biomarker analysis
Previously untreated
patients with stage IIIB/IV
NSCLC, adenocarcinoma,
never or ex-light smokers,
WHO PS 0-2
(N = 1217)
Gefitinib 250 mg/day PO
(n = 609)
Paclitaxel 200 mg/m2 IV on Day 1 +
Carboplatin AUC 5-6 mg/mL/min IV on Day 1
Up to six 3-wk cycles
(n = 608)

IPASS: PFS
 PFS with gefitinib superior to carboplatin/paclitaxel in ITT population
 EGFR mutations strongly predicted PFS (and tumor response) to first-line
gefitinib vs carboplatin/paclitaxel
Mok. NEJM. 2009;361:947.
EGFR Mutation Positive
HR: 0.48
(95% CI: 0.36-0.64;
P < .001)
1.0
0.8
0.6
0.4
0.2
0
0 4 8 12 16 20 24
EGFR Mutation Negative
HR: 2.85
(95% CI: 2.05-3.98;
P < .001)
1.0
0.8
0.6
0.4
0.2
0
0 4 8 12 16 20 24
Gefitinib
Carbo/pac
Gefitinib
Carbo/pac
Overall
ProbabilityofPFS
HR: 0.74
(95% CI: 0.65-0.85;
P < .001)
1.0
0.8
0.6
0.4
0.2
0
0 4 8 12 16 20 24
Gefitinib
Carbo/pac

 Primary endpoint: PFS
 Secondary endpoints: OS, ORR, DCR, safety/tolerability, QoL
 Randomization did not include stratification factors; analyses adjusted for age
(< vs ≥ 65 yrs) and prior gefitinib response (SD vs PR/CR)
IMPRESS: Cisplatin/Pemetrexed ± Gefitinib in EGFR-
Mutant NSCLC After PD
Chemo-naive patients 18 yrs of age
or older* with stage IIIB/IV NSCLC,
an activating EGFR mutation, CR/PR
> 4 mos or SD ≥ 6 mos with first-line
gefitinib and subsequent PD† within
4 wks before randomization
(N = 265)
Cisplatin 75 mg/m2 + Pemetrexed 500 mg/m2 (≤ 6 cycles) +
Gefitinib 250 mg
(n = 133)
Cisplatin 75 mg/m2 + Pemetrexed 500 mg/m2 (≤ 6 cycles) +
Placebo
(n = 132)
*20 yrs of age or older in Japan. †Based on radiologic evaluation by Jackman criteria and RECIST v1.1. Tumor assessments
performed ≤ 4 wks prior to treatment and every 6 wks following randomization.
Slide credit: clinicaloptions.comSoria. Lancet Oncol. 2015;16:990.

Gefitinib + CT (n = 133)
Placebo + CT (n = 132)
100
80
60
40
20
0
0 2 4 6 8 10 12 14
PFS(%)
IMPRESS: PFS in ITT Population
*Primary Cox analysis with covariates. HR < 1 implies lower risk of
progression with gefitinib.
Soria. Lancet Oncol. 2015;16:990.
HR: 0.86* (95% CI: 0.65-1.13; P = .27)
Median PFS, Mos
(95% CI)
Events, n (%)
Gefitinib + CT (n = 133) 5.4 98 (74)
Placebo + CT (n = 132) 5.4 107 (81)

60
WJOG 5108L Study: Erlotinib vs Gefitinib in Previously
Treated NSCLC
 Eligible pts had stage IIIB/IV or recurrent adenocarcinoma and previous chemotherapy; EGFR TKI naive
Urata Y, et al. J Clin Oncol. 2016;[Epub ahead of print].
EGFR Mutation–Positive
100
80
40
20
0
0 484236302418126
Mos
PFS(%)
198
203
0
0
3
0
5
1
11
4
17
15
31
38
74
72
143
136
No. at risk
Erlotinib
Gefitinib
Erlotinib
Gefitinib
10.0 (95% CI; 8.5-11.2)
8.3 (95% CI;7.2-9.7)
HR 1.093 (95% CI; 0.879-1.358) P = .424
Median (mos)

Phase III RELAY: First-line Erlotinib + Ramucirumab for
EGFR-Mutated Advanced NSCLC
Ramucirumab + Erlotinib
(n = 224)
Placebo + Erlotinib
(n = 225)
HR (95% CI) P Value
ORR, % 76.3 74.7 -- .7413
Median PFS, mos (95% CI) 19.4 (15.4-21.6) 12.4 (11.0-13.5) 0.591 (0.461-0.760) < .0001
Median OS, mos (95% CI)* NR NR 0.832 (0.532-1.303) .4209
Grade ≥ 3 toxicity, % 72 54 -- --
Nakagawa. ASCO 2019. Abstract 9000.
Median follow-up: 20.7 mos. *Interim analysis.
Patients with CNS metastases were excluded

ASPIRATION: Erlotinib (Before and After PD) in EGFR-
Mutated NSCLC
 Primary endpoint: PFS1 (time to PD or death by RECIST v1.1)
 Secondary endpoints: PFS2 (time to off-erlotinib PD if erlotinib was
extended beyond initial PD), ORR, DCR, OS, safety
Treatment-naive Asian adults
with stage IV or recurrent
NSCLC and activating EGFR
mutations (exon 18-21
except T790), ECOG PS 0-2
(N = 207)
Park. JAMA Oncol. 2016;2:305.
PFS1
PFS2
Erlotinib
150 mg/day
PD by
RECIST v1.1
Erlotinib*
150 mg/day
Off-erlotinib
PD
*Continued after initial PD at patient and/or investigator discretion, n = 93.

ASPIRATION: PFS
 In patients receiving post-PD
erlotinib
‒ Median PFS1: 11.0 mos
‒ Median PFS2: 14.1 mos
Park. JAMA Oncol. 2016;2:305.
Mos
ProbabilityofPFS
Subset Increase in mPFS2 vs
mPFS1, Mos
All patients receiving
post-PD erlotinib
3.1
Exon 19 deletion 3.9
L858R 4.8
PFS in Patients Receiving Post-PD Erlotinib
Patients
at Risk, n
PFS1
PFS2
93
93
52
70
10
27
0
0
0
0
1.0
0.8
0.6
0.4
0.2
0
PFS1
PFS2
0 4010 20 30

Ablative Therapy in Patients With Limited Progression
 Surgery or RT in patients with initial response to crizotinib or erlotinib
and progression to ≤ 4 extra-CNS sites (n = 25)
‒ PFS1: 9.8 mos
‒ PFS2: 6.2 mos
‒ CNS-PFS2: 7.1 mos
‒ Extra-CNS-PFS2: 4.0 mos
Weickhardt. J Thorac Oncol. 2012;7:1807.

Noncanonical EGFR Mutations: L861Q, G719X, S768I
 Afatinib: FDA approval in 2018 based on pooled analysis of LUX-Lung 2, 3, 6
(N = 32)[1]
‒ ORR: 66% (95% CI: 47% to 81%)
‒ DoR at ≥ 12 mos: 52%
‒ DoR at ≥ 18 mos: 33%
 Osimertinib?
‒ Phase II study in NSCLC with uncommon EGFR mutations (N = 36)[2]
‒ ORR: 50%,
‒ Median PFS: 9.5 mos
1. Afatinib PI. 2. Ahn. ASCO 2018. Abstr 9050.

* *
*
*
*
Noncanonical EGFR Mutations: EGFR Exon 20 Insertion
 Typically refractory to current EGFR TKIs
 Poziotinib: novel inhibitor of EGFR and HER2 exon 20 insertions
 Phase II trial: 44 evaluable patients with NSCLC and EGFR and HER2 exon 20 mutations
Heymach. WCLC 2018. Abstr OA02.06.
PD
SD
PR
Response not confirmed/follow-up pending
Remains on treatment
MaximumResponse
FromBaseline
40
20
0
-20
-40
-60
-80
Germline T790M
+ exon20ins
*
**
* * * *
*
*
* *
* *
*
T790M
*
ORR (best): 55%
ORR (confirmed): 43%
Median PFS: 5.5 mos

Afatinib Expanded Access Program: Study Design
 Open-label, single-arm study enrolled pts (treated, n = 322) with EGFR-
positive locally advanced or metastatic NSCLC ineligible for other
afatinib clinical trials prior to drug approval
 87.3% pts previously treated with TKI
 Pt EGFR mutation: del(19), 57.5%; L858R, 31.1%
 Stage IV at diagnosis: 67.7%
 Primary endpoint: safety
Kim ES, et al. Chicago Multidisciplinary Symposium in Thoracic Oncology 2014. Abstract 182.

Afatinib Expanded Access: Treatment-Related AEs ≥
10%
 Serious treatment-related AEs observed in 7.8% of pts
 Discontinuation due to drug-related AEs: 5.3%
100
80
60
40
20
0
77.0
36.0
16.8 13.4 12.7 12.4 11.2 10.6
01.20.60.3
1.21.2
1.9
9.9
Diarrhea Rash Mucosal
Inflammation
Stomatitis Nausea Dry skin Fatigue Paronychia
Grade 3/4* All grades
Pts(%)
*No grade 5 AEs were reported.

Afatinib Expanded Access: Best Tumor Response
*Tumor response is based on clinical, radiological, or other assessment.
†Includes 23 pts treated with first-line afatinib.
Response,* n (%;
95% Cl)
Overall
(N = 322)
TKI Naive†
(n = 41)
TKI Exposed
(n = 281)
Disease control 225 (69.9; 64.5-74.8) 34 (82.9; 67.9-92.8) 191 (68.0; 62.2-73.40)
Objective response 55 (17.1; 13.1-21.6) 14 (34.1; 20.1-50.6) 41(14.6; 10.7-19.3)
CR 5 (1.6; 0.5-3.6) 3 (7.3; 1.5-19.9) 2 (0.7; 0.1-2.5)
PR 50 (15.5; 11.8-20.0) 11 (26.8; 14.2-42.9) 39 (13.9; 10.1-18.5)
SD 170 (52.8; 47.2-58.4) 20 (48.8; 32.9-64.9) 150 (53.4; 47.4-59.3)

Afatinib Expanded Access: Conclusions
 Afatinib exhibited similar safety profile compared with earlier trials of
afatinib-treated pts
 Afatinib demonstrated good response in heavily pretreated population
‒ Best tumor response (17.1%); SD (52.8%); disease control (69.9%)
‒ Median PFS (3.6 mos) similar to results in other heavily pretreated
populations
 Afatinib has tolerable safety profile and should be considered for
treatment of pts with EGFR-positive NSCLC

LL3 and LL6: First-line Afatinib vs CT in Pts With
Advanced EGFR+ NSCLC
 Phase III trials in stage IIIB/IV EGFR mutation–positive NSCLC (LL3, N = 307; LL6, N = 364)
‒ Randomized 2:1 to oral afatinib 40 mg/day or up to 6 cycles of standard CT (LL3,
pemetrexed/cisplatin; LL6, gemcitabine/ cisplatin)
 Stratified by mutation type (del(19)/L858R/other, both trials) and by race (Asian/non-
Asian; LL3 only)
 Primary endpoint of median PFS previously reported
1. Sequist L, et al. J Clin Oncol. 2013;31:3327-3334. 2. Wu YL, et al. Lancet Oncol. 2014;15:213-222.
Median PFS LUX-Lung 3[1] LUX-Lung 6[2]
Afatinib, mos 11.1 11.0
Chemotherapy, mos 6.9 5.6
HR 0.58 (P = .001) 0.28 (P < .001)

LL3 and LL6 First-line Afatinib vs CT:
OS in NSCLC With Common Mutations
Sequist L, et al. Chicago Multidisciplinary Symposium in Thoracic Oncology 2014. Abstract 9.
Median follow-up: 41 mos Median follow-up: 33 mos
EstimatedOSProbability
LUX-Lung 3
1.0
0.8
0.6
0.4
0.2
0
0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51
Mos
Afatinib
(n = 203)
31.6
Cis/Pem
(n = 104)
28.2Median, mos
HR: 0.78 (95% CI: 0.58-1.06; P = .1090)
LUX-Lung 6
1.0
0.8
0.6
0.4
0.2
0
0 3 6 9 12 15
EstimatedOSProbability
18 21 24 27 30 33 36 39 42 45
Mos
Afatinib
(n = 216)
23.6
Cis/Pem
(n = 108)
23.5Median, mos
HR: 0.83 (95% CI: 0.62-1.09; P = .1756)

LL3 and LL6 First-line Afatinib vs CT:
OS Conclusions
 First-line afatinib significantly improved OS in all pts with EGFR del(19)
 No significant difference vs chemotherapy in OS of pts with EGFR L858R mutation
 First-line afatinib should be the standard of care for pts with EGFR del(19)
‒ Remains a treatment option for pts with EGFR L858R mutation
Sequist L, et al. Chicago Multidisciplinary Symposium in Thoracic Oncology 2014. Abstract 9.
Regimen
Median OS With EGFR del(19)
LUX-Lung 3
(Global pop.)
LUX-Lung 3
(Non-Asian pop.)
LUX-Lung 6
(Asian pop.)
Afatinib 33.3 mos 33.6 mos 31.4 mos
Chemotherapy 21.1 mos 20.0 mos 18.4 mos
HR (95% CI) 0.54 (0.36-0.79)
P = .002
0.45 (0.21-0.95)
P = .031
0.64 (0.44-0.94)
P = .023

Phase IIb LUX-Lung 7: Afatinib vs Gefitinib in
 Coprimary endpoints: PFS, TTF, OS
 Secondary endpoints: ORR, time to response, DoR, DCR, duration of
disease control, tumor shrinkage, QoL
Park. Lancet Oncol. 2016;17:577.
Afatinib*† 40 mg PO QD
(n = 160)
Gefitinib†‡ 250 mg PO QD
(n = 159)
Treatment continued
until PD or
unacceptable toxicity
Stratified by EGFR mutation (exon 19 deletion vs
L858R) and brain metastases at baseline (yes vs no)
Treatment-naive patients with
stage IIIB or IV lung adenocarcinoma,
exon 19 deletion or L858R
EGFR mutations, ECOG PS 0/1,
adequate organ function
(N = 319)
*Dose escalation to 50 mg allowed in absence of TEAEs.
†Treatment interruptions ≤ 14 days allowed. ‡Dose modifications allowed.

LUX-Lung 7: Survival
 Median follow-up: 42.6 mos
 Median treatment duration: afatinib, 13.7 mos; gefitinib, 11.5 mos
Paz-Ares. Ann Oncol. 2017;28:270.
Afatinib
Gefitinib
Median PFS, Mos
11.0
10.9 HR: 0.86 (95% CI: 0.66-1.12)
P = .2580
Afatinib
Gefitinib
Median OS, Mos
27.9
24.5
PFS
HR: 0.74 (95% CI: 0.57-0.95)
P = .0178
OS
0
Mos
PFS(%)
246 12 18 30 36 42
100
80
60
40
20
0Patients at
Risk, n
Afatinib
Gefitinib
3 9 2715 21 33 39 45 48 51
160
159
142
132
113
105
94
82
67
51
47
21
34
15
26
10
20
7
13
5
10
5
8
5
4
3
3
3
0
0
0
0
0
0
0
0
0
Mos
OS(%)
246 12 18 30 36 42
100
80
60
40
20
0Patients at
Risk, n
Afatinib
Gefitinib
3 9 2715 21 33 39 45 48 51
160
159
156
153
153
148
148
142
139
133
125
119
111
105
104
90
94
80
81
71
74
62
61
56
50
48
36
44
30
27
12
7
2
0
0
0

Dacomitinib
 Second-generation TKI
 ARCHER 1050: dacomitinib vs gefitinib in advanced NSCLC with EGFR-
activating mutations[1,2]
‒ N = 452, no CNS mets
‒ Median PFS: 14.7 vs 9.2 mos (HR: 0.59; P < .0001)
‒ Median OS: 34.1 vs 26.8 mos (HR: 0.76; P = .0438)
‒ Toxicity: 66% vs 8% had dose reduction
 FDA approved, but unclear role in EGFR landscape
‒ More toxic, not CNS active, with shorter PFS than osimertinib (and in a more
favorable population)
1. Wu. Lancet Oncol. 2017;18:1454. 2. Mok. ASCO 2018. Abstr 9004.

ARCHER 1050: Dacomitinib vs Gefitinib in EGFR-
Mutated Advanced NSCLC
 Primary endpoint: PFS by blinded independent review
 Secondary endpoints: PFS by investigator assessment, ORR, DoR, TTF,
OS, safety, patient-reported outcomes
Treatment-naive patients with
stage IIIB/IV or recurrent NSCLC,
EGFR-activating mutation(s);
ECOG PS 0/1; no prior systemic
therapy for advanced NSCLC;
no CNS metastases
(N = 452)
Stratified by race (Japanese vs Chinese vs other east Asian vs non-Asian),
EGFR mutation (exon 19 deletion vs L858R)
Dacomitinib 45 mg PO QD
(n = 227)
Gefitinib 250 mg PO QD
(n = 225)
Wu. Lancet Oncol. 2017;18:1454. Mok. JCO. 2018;36:2244.
Treatment
continued in 28-day
cycles until PD or
unacceptable
toxicity

0
ARCHER 1050: PFS by Blinded Independent Review
Dacomitinib (n = 227)
Gefitinib (n = 225)
Median PFS,
Mos (95% CI)
14.7 (11.1-16.6)
9.2 (9.1-11.0)
Events, n
136
179
Mos
PFS(%)
HR: 0.59 (95% CI: 0.47-0.74; P < .0001)
246 12 18 30 36 42
100
80
60
40
20
0
Censored
Wu. Lancet Oncol. 2017;18:1454.
Patients at Risk, n
(no. censored)
Dacomitinib
Gefitinib
227 (0)
225 (0)
154 (23)
155 (15)
106 (31)
69 (23)
73 (36)
34 (27)
20 (74)
7 (40)
6 (88)
1 (45)
0 (91)
0 (46)
0 (91)
0 (46)

ARCHER 1050: Overall Survival
Mok. JCO. 2018;36:2244.
HR: 0.760 (95% CI: 0.582-0.993; P = .0438)
Dacomitinib Gefitinib
Patients, n
Deaths, n
Median OS, mos
(95% CI)
30-mo OS rate, %
227
103
34.1
(29.5-37.7)
56.2
225
117
26.8
(23.7-32.1)
46.3
Median follow-up: 31.3 mos
Mos
OS(%)
100
80
60
40
20
0
0 6 12 18 24 30 36 42 48
Patients at Risk, n
Dacomitinib
Gefitinib
227
225
206
213
188
186
167
144
138
113
77
63
14
12
3
3
0
0
Censored++++
+++ +
+
+
+
+ ++ ++
+
+++++ ++ ++
++++ ++++++ +++++ ++++++++++++ ++++++++
+ ++ + ++
+
+
++
+++++++ ++++++ + +++++++++
++
+
+
++++++
+ + ++ ++++ ++
+
+++ +++
++
+++
+ +
++
+
+
++
++

ARCHER 1050: Safety
 Most frequent grade ≥ 3 AEs with dacomitinib: dermatitis acneiform (13.7%),
diarrhea (8.8%), paronychia (7.5%), rash (4.4%), stomatitis (3.5%)
 Most frequent grade ≥ 3 AE with gefitinib: ALT increase (8.5%), AST increase (4.0%)
Parameter Dacomitinib
(n = 227)
Gefitinib
(n = 224)
Median time to dose reduction, mos (range) 2.8 (0.3-20.3) 3.3 (1.2-25.7)
Median duration of dose reduction, mos (range) 11.3 (0.1-33.6) 5.2 (0.3-17.8)
Dacomitinib reduction to 30 mg/day,* n (%) 88 (38.8) NA
Dacomitinib reduction 15 mg/day,† n (%) 63 (27.8) NA
Patients with dose reduction, n (%) 151 (66.5) 18 (8.0)
Mok. ASCO 2018. Abstr 9004.
*First dose reduction. †Second dose reduction.

ARCHER 1050: First-line Dacomitinib vs Gefitinib for
0
Dacomitinib (n = 227)
Gefitinib (n = 225)
Median PFS,
Mos (95% CI)
14.7 (11.1-16.6)
9.2 (9.1-11.0)
Events, n
136
179
Mos
PFS(%)
HR: 0.59
(95% CI: 0.47-0.74; P < .0001)
246 12 18 30 36 42
100
80
60
40
20
0
Censored
Patients at Risk, n
(no. censored)
Dacomitinib
Gefitinib
227 (0)
225 (0)
154 (23)
155 (15)
106 (31)
69 (23)
73 (36)
34 (27)
20 (74)
7 (40)
6 (88)
1 (45)
0 (91)
0 (46)
0 (91)
0 (46)
Wu. Lancet Oncol. 2017;18:1454.
Patients, % Dacomitinib Gefitinib
Skin rash
(G3/4)
4 0
Diarrhea
(G3/4)
8 1
Dose
reduction
66 8
PFS
Patients with CNS metastases were excluded.
Patients with CNS metastases
were excluded

AURA: AZD9291 in Previously Untreated Advanced
NSCLC
Predefined
expansion cohorts
Sequential cohorts of pt with
previously untreated LA/
metastatic NSCLC with
confirmed EGFR mutation,
WHO
PS 0-1
Cohort 5 (240 mg) T790M+
Cohort 4 (160 mg)
(n = 30)
T790M+/-
Cohort 3 (80 mg)
(n = 30)
T790M+/-
Cohort 2 (40 mg) T790M+/-
Cohort 1 (20 mg) T790M+
Ramalingam SS, et al. ASCO 2015. Abstract 8000.
AZD9291 Dosing

-70
-50
-30
-10
AURA: Tumor Response and PFS
Ramalingam SS, et al. ASCO 2015. Abstract 8000. Reprinted with permission.
*Ongoing.
50
40
30
20
10
0
-20
-40
-60
-80
-90
-100
80 mg
160 mg
BestPercentage
ChangeFromBaseline
inTargetLesion
Individual Patients
D
D
DD D D D D
D
D
D*
Outcome
80 mg
(n = 30)
160 mg
(n = 30)
Total
(N = 60)
Maximum DoR, mos 13.8* 9.7*
PFS, % (95% CI)
 3 mos
 6 mos
 9 mos
 12 mos
90 (72-97)
83 (64-93)
83 (64-93)
73 (51-87)
97 (79-100)
90 (72-97)
78 (57-89)
NC
93 (83-97)
87 (75-93)
81 (68-89)
72 (55-64)

AURA: Safety
 Most common toxicities: skin rash, diarrhea, dry skin, stomatitis; mostly grade 1
 No grade ≥ 3 hyperglycemia, QT prolongation, or ILD-like events
Ramalingam SS, et al. ASCO 2015. Abstract 8000.
AE, %
80 mg
(n = 30)
160 mg
(n = 30)
Total
(N = 60)
Any event grade ≥ 3 33 43 38
Treatment-related AE 97 100 98
Treatment-related AE grade ≥ 3 10 20 15
Treatment-related AE leading to
discontinuation
7 3 5
Treatment-related serious AE 10 3 7

Osimertinib in NSCLC With EGFR T790M Mutation–
Positive Acquired Resistance
ORR: 61%
Jänne. NEJM. 2015;372:1689.
40 mg
80 mg
160 mg
20 mg
-100
-90
-80
-70
-60
-50
-40
-30
-20
-10
0
10
20
30
40
50
Best % Change From Baseline in Target Lesion
240 mg
D
D*D*
D
D D
D
DDDD
DD
D D
DD DDD
D DD
D D DD DD
DDD
DD
D
Best%ChangeFrom
BaselineinTargetLesion
*Values imputed as 20%.

FLAURA: First-line Osimertinib vs SoC for EGFR-Mutant
Advanced NSCLC
 Double-blind phase III study
 Primary endpoint: investigator-assessed PFS (RECIST v1.1)
 Secondary endpoints including ORR, DoR, DCR, depth of response, OS, PRO, safety
Treatment-naive patients
with advanced NSCLC
adenocarcinoma with
an EGFR exon 19 or 21
mutation, WHO PS 0/1,
stable CNS mets permitted
(N = 556)
Osimertinib 80 mg PO daily
(n = 279)
Erlotinib 150 mg or Gefitinib 250 mg
PO daily
(n = 277)
EGFR mutation (del[19] vs L858R) and
race (Asian vs non-Asian)
Soria. NEJM. 2018;378:113. Ramalingam. ESMO 2017. Abstr LBA2_PR.
Until PD or
unacceptable
toxicity
Crossover to open-
label osimertinib
allowed upon
progression and
T790M+
confirmation

1.0
0.8
0.6
0.4
0.2
0
0 6 9 2112 18 24 27153
FLAURA: PFS and OS
Ramalingam. ESMO 2017. Abstr LBA2_PR. Soria. NEJM. 2018;378:113.
Patients at Risk, n
Osimertinib
SoC
279
277
262
239
233
197
210
152
139
78
71
37
26
10
0
0
178
107
4
2
HR: 0.46
(95% CI: 0.37-0.57; P < .001)
Median OS, Mos
(95% CI)
Osimertinib (n = 279) NC (NC-NC)
SoC (n = 277) NC (NC-NC)
HR: 0.63
(95% CI: 0.45-0.88; P = .007*)
*P < .0015 was required for statistical significance at current maturity.
279
277
276
263
269
252
253
237
243
218
232
200
154
126
87
64
4
1
0
0
0
0
29
24
ProbabilityofOS
Mos
ProbabilityofPFS
Mos
Median PFS,
Mos (95% CI)
Osimertinib (n = 279) 18.9 (15.2-21.4)
SoC (n = 277) 10.2 (9.6-11.1)
OSPFS
1.0
0.8
0.6
0.4
0.2
0
0 6 9 2112 18 24 27153 30 33

FLAURA: PFS by Subgroup
Soria. NEJM. 2018;378:113.
Overall
Log-rank test: primary analysis
Cox proportional-hazards model
Sex
Male
Female
Age
< 65 yrs
≥ 65 yrs
Race
Asian
Non-Asian
Smoking history
Yes
No
Known or treated CNS metastases at trial entry
Yes
No
WHO PS
0
1
EGFR mutation at randomization
Exon 19 deletion
L858R
EGFR mutation by ctDNA
Positive
Negative
Centrally confirmed EGFR mutation
Positive
Negative
556
206
305
298
258
347
209
199
357
116
440
228
327
349
207
359
124
500
6
0.46 (0.37-0.57)
0.46 (0.37-0.57)
0.58 (0.41-0.82)
0.40 (0.30-0.52)
0.44 (0.33-0.58)
0.49 (0.35-0.67)
0.55 (0.42-0.72)
0.34 (0.23-0.48)
0.48 (0.34-0.68)
0.45 (0.34-0.59)
0.47 (0.30-0.74)
0.46 (0.36-0.59)
0.39 (0.27-0.56)
0.50 (0.38-0.66)
0.43 (0.32-0.56)
0.51 (0.36-0.71)
0.44 (0.34-0.57)
0.48 (0.28-0.80)
0.43 (0.34-0.54)
NC (NC-NC)
Subgroup Patients, n HR for Disease Progression or Death (95% CI)
Osimertinib Better Standard EGFR TKI Better
0.1 0.2 0.3 0.4 0.6 1.0 2.0 10.0

FLAURA: PFS Benefit With First-line Osimertinib by CNS
Metastasis and EGFR Mutation Testing Method
Soria. NEJM. 2018;378:113.
Subgroup Patients, n HR of Disease Progression or Death (95% CI)
10.00.1 0.2 0.3 0.4 0.6 1.0 2.0
Osimertinib Better Standard EGFR TKI Better
0.46 (0.37-0.57)
0.46 (0.37-0.57)
0.47 (0.30-0.74)
0.46 (0.36-0.59)
0.43 (0.32-0.56)
0.51 (0.36-0.71)
0.44 (0.34-0.57)
0.48 (0.28-0.80)
0.43 (0.34-0.54)
NC (NC-NC)
556
116
440
349
207
359
123
500
6
Overall
Log-rank test: primary analysis
Cox proportional-hazards model
Known or treated CNS metastases at trial entry
Yes
No
EGFR mutation at randomization
Exon 19 deletion
L858R
EGFR mutation at by ctDNA
Positive
Negative
Centrally confirmed EGFR mutation
Positive
Negative

FLAURA: PFS by CNS Metastases at Baseline
 CNS progression events occurred in 17 patients (6%) with osimertinib vs 42 patients (15%) with SoC
EGFR TKI
Median PFS,
Mos (95% CI)
Osimertinib (n = 53) 15.2 (12.1-21.4)
SoC (n = 63) 9.6 (7.0-12.4)
HR: 0.47
(95% CI: 0.30-0.74; P < .001)
Ramalingam. ESMO 2017. Abstr LBA2_PR. Soria. NEJM. 2018;378:113.
Patients at
Risk, n
Osimertinib
SoC
53
63
51
57
40
40
37
33
22
13
9
6
4
2
0
0
32
24
1
1
Mos
Patients at
Risk, n
Osimertinib
SoC
226
214
211
182
193
157
173
119
117
65
62
31
22
8
0
0
146
83
3
1
Mos
Median PFS,
Mos (95% CI)
Osimertinib (n = 226) 19.1 (15.2-23.5)
SoC (n = 214) 10.9 (9.6-12.3)
HR: 0.46
(95% CI: 0.36-0.59; P < .001)
ProbabilityofPFS
ProbabilityofPFS
With CNS Metastases at BL (n = 116) Without CNS Metastases at BL (n = 440)
1.0
0.8
0.6
0.4
0.2
0
0 3 6 9 12 15 18 21 24 27
1.0
0.8
0.6
0.4
0.2
0
0 3 6 9 12 15 18 21 24 27

100
80
60
40
20
0
FLAURA: CNS Progression on First-line Osimertinib vs
Standard EGFR TKI
Osimertinib (n = 279)
Standard EGFR TKI
(gefitinib, n = 183; erlotinib, n = 94)
Soria. NEJM. 2018;378:113.
Patients(%)
19
43
3
7
Known/Treated CNS
Mets at Trial Entry
No Known/Treated CNS
Mets at Trial Entry
CNS Progression

AE, n (%)
Osimertinib (n = 279) SoC (n = 277)
Grade 1 Grade 2 Grade 3 Grade 1 Grade 2 Grade 3
Diarrhea 120 (43) 35 (13) 6 (2) 116 (42) 35 (13) 6 (2)
Dry skin 87 (31) 12 (4) 1 (< 1) 76 (27) 21 (8) 3 (1)
Paronychia 52 (19) 44 (16) 1 (< 1) 55 (20) 34 (12) 2 (1)
Stomatitis 65 (23) 13 (5) 1 (< 1) 47 (17) 8 (3) 1 (< 1)
Dermatitis acneiform 61 (22) 10 (4) 0 71 (26) 50 (18) 13 (5)
Decreased appetite 27 (10) 22 (8) 7 (3) 25 (9) 22 (8) 5 (2)
Pruritus 40 (14) 7 (3) 1 (< 1) 30 (11) 13 (5) 0
Cough 34 (12) 12 (4) 0 25 (9) 16 (6) 1 (< 1)
Constipation 33 (12) 9 (3) 0 28 (10) 7 (3) 0
AST increased 18 (6) 6 (2) 2 (1) 38 (14) 18 (6) 12 (4)
ALT increased 11 (4) 6 (2) 1 (< 1) 31 (11) 19 (7) 21 (8)
Grade 4 AEs: osimertinib, n = 1 stomatitis, n = 1 prolonged QT interval, n = 4 not specified; SoC, n = 4 ALT increased, n = 7 not specified.
Soria. NEJM. 2018;378:113. Ramalingam. ESMO 2017. Abstr LBA2_PR.
FLAURA: Adverse Events
 Median duration of exposure, mos: 16.2 (range: 0.1-27.4) with osimertinib; 11.5 (range: 0-26.2) with SoC

Disease Progression on EGFR TKI (Erlotinib, Gefitinib, or
Afatinib) in NSCLC With EGFR Sensitizing Mutations
 PD: Clinical characteristics
‒ Global progression
‒ Slow growth globally
‒ Growth in several areas, but not all
 PD: Molecular characteristics
‒ SCLC transformation
‒ EGFR T790M (exon 20)
‒ MET amplification
‒ PIK3CA
‒ Unknown (other pathways)
EMT ~ 1% to 2%
HER2 amplification
~ 8% to 13%
BRAF ~ 1%
MET amplification
~ 5%
PIK3CA ~ 1% to 2%
SCLC alone ~ 6%
SCLC with PI3K ~ 4%
BypassTracks~20%
No identification
AR mechanism
~ 15% to 20%
T790M
alone
~ 40% to
55%
T790M
with EGFR
amplification
~ 10%
Other EGFR
point mutations
1% to 2%
EGFRTarget
Alteration~60%
Camidge. Nat Rev Clin Oncol. 2014;11:473.

Variation in Mechanisms of Acquired Resistance to
EGFR TKIs in NSCLC
Changes in T790M:
 Not explained by assay sensitivity alone
 Not correlated with biopsy site
60 pts
> 1 post-AR biopsy
18/60 (30%)
remained
T790M negative
27/60 (45%)
remained
T790M positive
15/60 (25%)
changed T790M status
169 pts
 Advanced EGFR-
mutant NSCLC
 AR to 1st-/2nd-
generation EGFR TKI
109 pts
1 post-AR biopsy
Piotrowska Z, et al. Chicago Multidisciplinary Symposium in Thoracic Oncology 2014. Abstract 11.

Conclusions
 Heterogeneity of T790M is common among EGFR-mutant lung cancer
pts with acquired resistance to first-/second-generation EGFR inhibitors
 A single postresistance biopsy may not be reflective of tumor
heterogeneity and may not fully predict response to novel targeted
agents
 New methods are needed to better define heterogeneous resistance
mechanisms and identify pts who may benefit from tailored therapy
Piotrowska Z, et al. Chicago Multidisciplinary Symposium in Thoracic Oncology 2014. Abstract 11.

Relative Frequencies of Acquired Resistance Mechanisms
to EGFR TKIs: Importance of T790M
Yu. Clin Cancer Res. 2013;19:2240.
T790M
60%
HER2
8%
Unknown
18%
HER2 + T790M
4%
MET
amplification
3%
Small cell + MET
1%
Small cell
1%
Small cell + T790M
2%
MET + T790M
3%

Need for Repeat Testing: Mechanisms of Resistance to
First- and Second-Generation EGFR TKIs
241Yu. Clin Cancer Res. 2013;19:2240.
1
1
2
3
3
4
8
18
60
0 10 20 30 40 50 60 70
Small Cell
Small Cell + MET
Small Cell + T790M
MET + T790M
MET amplification
HER2 + T790M
HER2
Unknown
T790M
Relative Frequencies (%)

Resistance to First- and Second-Generation EGFR TKIs in
NSCLC
 First-line use of EGFR TKI established since 2009. . . but resistance inevitably
develops
Sequist. Sci Transl Med. 2011;3:75ra26.
Changes in EGFR Over Time With EGFR TKI Therapy
Histology Adeno Adeno Adeno
Genotype L858R
TP53
L858R
TP53
T790M
L858R
TP53
EGFR TKI
Status
Sensitive Resistant Sensitive
Tumor Burden ↑ ↓ ↑ ↑ ↑ ↓
Treatment Chemo Erlotinib Chemo Chemo Erlotinib*
Timeline 2007 2008 2009 2010
*In combination with an investigational agent that did not target T790M.

AURA3: Osimertinib vs Platinum/Pemetrexed in EGFR
T790M–Positive Advanced NSCLC
 Multicenter, randomized, open-label phase III trial
 Primary endpoint: PFS (investigator assessed)
 Secondary endpoints: ORR (investigator assessed), DoR, DCR, tumor shrinkage, OS,
patient-reported outcomes, safety
Locally advanced or metastatic NSCLC
with disease progression and EGFR
T790M mutation after first-line EGFR
TKI therapy; ≤ 1 line of therapy for
advanced NSCLC
(N = 419)
Osimertinib 80 mg QD
(n = 279)
Platinum/Pemetrexed Chemotherapy*
Q3W for up to 6 cycles
(n = 140)
Mok. NEJM. 2017;376:629.
Stratified by race (Asian vs non-Asian)
*Pemetrexed 500 mg/m2 plus either carboplatin AUC 5 or cisplatin 75 mg/m2.

AURA3: PFS by Investigator Assessment
 PFS benefit with osimertinib evident
across evaluated subgroups
 Osimertinib: FDA approved for
patients with metastatic EGFR
T790M–positive NSCLC, as detected
in tissue or plasma by an FDA-
approved test, that has progressed
on or after EGFR TKI therapy
 Liquid biopsy has a 30% false-
negative rate for T790M detection;
therefore, negative results still
require tumor biopsy
 EGFR T790M mutation testing by
rebiopsy or liquid biopsy is standard
of care for patients with progression
on an EGFR TKI other than
osimertinib
100
80
60
40
20
0
PFS(%)
Osimertinib
Platinum/pemetrexed
0 3 6 9 12 15 18
Mos
Outcome
Osimertinib
(n = 279)
Plt/Pem
(n = 140)
HR (95% CI)
Median PFS, mos 10.1 4.4 0.30
(0.23-0.41)
P < .001(95% CI) (8.3-12.3) (4.2-5.6)
Mok. NEJM. 2017;376:629. Osimertinib PI. Oxnard. J Clin Oncol. 2016;34:3375.

AURA3: Osimertinib vs CT in NSCLC With EGFR T790M
Mutation–Positive Acquired Resistance
PFS (ITT)[1] CNS PFS[2]
1. Mok. NEJM. 2017;376:1993. 2. Mok. ASCO 2017. Abstr 9005.
Median CNS PFS,
Mos
Osimertinib (n = 75) 11.7
Plt/Pem (n = 41) 5.6
HR: 0.32 (95% CI: 0.15-0.69; P = .004)
100
80
60
40
20
0
PFS(%)
Osimertinib
Platinum/pemetrexed
0 3 6 9 12 15 18
Mos
HR: 0.30 (95% CI: 0.23-0.41; P < .001)
Median PFS, Mos
(95% CI)
Osimertinib (n = 279) 10.1 (8.3-12.3)
Plt/Pem (n = 140) 4.4 (4.2-5.6)
100
80
60
40
20
0
PFS(%)
Osimertinib
Platinum/pemetrexed
0 3 6 9 12 15 18
Mos

AURA3: Overall CNS Response in Patients Evaluable for
Response*
Outcome† Osimertinib 80 mg
(n = 30)
CT
(n = 16)
CNS ORR, % (95% CI) 70 (51-85) 31 (11-59)
OR (95% CI) 5.13 (1.44-20.64); P = .015
Median TTR, wks 6.1 6.1
Median DoR, mos
(95% CI)
8.9 (4.3-NC) 5.7 (NC-NC)
*Evaluable for response set: patients with ≥ 1 measurable CNS metastases on BL scan per BICR. †Confirmation not required for response.
‡Full analysis set: patients with ≥ 1 measurable and/or nonmeasurable CNS metastases on BL scan per BICR.
Wu. JCO. 2018;36:2702.
CR
PR
SD
PD
NE n = 5 (31%)
n = 1 (6%)
n = 5 (31%)
n = 4 (25%)
n = 1 (6%)
n = 1 (3%)
n = 1 (3%)
n = 7 (23%)
n = 19 (63%)
n = 2 (7%)
0 10 20 30 40 50 60 70
Patients (%)
CNS Overall Response†
Osimertinib 80 mg (n = 30)
Chemotherapy (n = 16)

Acquired Resistance to Osimertinib:
EGFR Mutation–Mediated Resistance
 Subset analysis of 76 patients with metastatic T790M+ NSCLC and
progression on osimertinib
 C797S (10.5%) was most common EGFR mutation upon PD; EGFR
amplification (9.2%) and L781Q, V726M, I744T, C775Y, G796S/D, and
T854I mutations also found
 In 35 patients, aberrations observed in bypass tracks including
ERBB2/3, FGFR3, HRAS, JAK1/2, MET, MTOR, NTRK1, PIK3CA
Zhou. ASCO 2018. Abstr 9077.

Acquired Resistance to Osimertinib
 Repeat biopsy after osimertinib
not SoC at present time
 Potential options based on result
‒ MET inhibitor
‒ First-generation EGFR TKI for
C797S
‒ Chemotherapy (platinum/
etoposide) for SCLC
transformation
 Clinical trials in development
Piotrowska. ASCO 2017. Abstr 9020.
Mechanism of Acquired Resistance
in Osimertinib-Resistant Patients (N = 23)
MET amp (7)
30%
T790M/
C797S (5)
22%
T790M loss
(unknown AR) (6)
26%
Other (4)
17%
SCLC
transformation (1)
4%
T790M Loss
(EGFR amp) (1)
4%

Osimertinib Resistance: EGFR C797S Mutation
 30% of patients
develop C797S
mutation after
treatment with
third-generation
EGFR TKI[1]
 C797S acquired
with or without
EGFR T790M
mutation[2]
249
Clonal Evolution of NSCLC and Resistance to Third-Generation EGFR TKIs[2]
1. Fogli. Pharmacogenomics. 2018;19:727. 2. Wang. J Hematol Oncol. 2016;9:59.
EGFR
T790M+ C797S+
Third-Generation EGFR TKIs
T790M+
T790M- T790M-
EGFR-sensitive mut+
EGFR T790M mut+
Other resistance mechanisms
to first-generation EGFR TKI
T790M+ plus C797S+
T790M- plus C797S+
Other resistance mechanisms
to third-generation EGFR TKI
with C797S-
C797S+
C797S-
C A T C A C G C A G C T C A T G C C C T T C G G C T G C C T

Prospects for Treating Resistance to Osimertinib
Niederst. Clin Cancer Res. 2015;21:3924.

EGFR Mutation–Positive NSCLC: Treatment Algorithm
After Progression on First-line Osimertinib
Osimertinib
C797S without T790M First-generation EGFR-TKI
Platinum-based
chemotherapy
± bevacizumab and
atezolizumab
Alternative Pathway
Active
Targeted therapy +
EGFR inhibitor
Platinum-based
chemotherapy
± bevacizumab and
atezolizumab
PD-1/PD-L1
inhibitor
Platinum-based
chemotherapy
± bevacizumab and
atezolizumab
Slow progression- Close observation is an option
Oligo-metastatic disease- Ablative therapy
CNS only progression after 1st or 2nd generation EGFR-TKIs- Osimertinib

Basis for a “Liquid Biopsy”
Lowes. Int J Mol Sci. 2016:17:E1505. Figure 1 of given citation is used in its original form under the terms and conditions of the
Creative Commons Attribution 4.0 International license (CC BY 4.0: https://creativecommons.org/licenses/by/4.0/).

Blood-Based Testing: What Is Circulating DNA?
 In lung cancer patients with progressive disease, dying tumor cells
release small pieces of DNA into the bloodstream[1,2]
 This DNA is called cell-free circulating tumor DNA (ctDNA), which
moves throughout the bloodstream[2]
 Identifying and analyzing cancer DNA from a blood sample allows
physicians to detect genetic changes in the tumor that may help guide
treatment[2,3]
1. Kimura H, et al. Clin Cancer Res. 2006;12:3915-3921. 2. Diaz LA Jr, et al.
J Clin Oncol. 2014;32:579-586. 3. Thress KS, et al. Lung Cancer. 2015;90:509-515.

Liquid vs Tissue Biopsy
254
Consideration ctDNA Assay Tissue Assay
Logistics  Ease of draw
 Venipuncture risks (variable)
 Serial testing easy
 Invasive, more challenging to obtain
 Biopsy risks (variable)
 Serial testing more difficult
Biology  ctDNA results not directly correlatable with
histology or cell phenotype
 More likely representative of whole tumor, with
caveat of differential tumor cell turnover having
potential to bias representation
 Biopsy results correlatable with
histology and cell phenotype
 Represents 1 small region of tumor
Preanalytical  Easier standardization but requires special handling
(without cell-stabilization tubes)
 Data on confounding patient-related factors limited
 More difficult to standardize across
sites but uses existing approaches
for tissue processing and handling
Clinical utility  Data to support treatment selection in advanced
cancers limited
 No data for other potential indications
 Data supporting treatment selection
across multiple tumor types (early
and advanced stages) substantial
Merker. J Clin Oncol. 2018;36:1631.

Genetic Testing: Tissue Biopsy vs Plasma ctDNAPlasma
ctDNA
Sensitivity: true positive/(true positive + false negative)
Specificity: true negative/(true negative + false positive)
Mroz. Cancer. 2017;123:917.
Assay Sensitivity and Specificity
Whole tumor
Impact of Subclonal Tumor
Evolution on Sampling
Sampling
time →
Sample 1
Tissue Biopsy
Mut+ Mut-
Mut+ True positive False positive
Mut- False negative True negative
Sample 2

Detection of EGFR Mutations in Plasma ctDNA
 FDA-approved, CE-marked IVD test uses plasma to test for EGFR mutations
‒ Plasma accuracy based on clinical trial samples (tissue served as gold standard)
EGFR Mutation Test, % (n/N)
Exon 19 deletion
Sensitivity
Specificity
82 (23/28)
97 (30/31)
L858R
Sensitivity
Specificity
87 (20/23)
97 (35/36)
T790M
Sensitivity
Specificity
73 (30/41)
67 (16/24)
Thress. Lung Cancer. 2015;90:509. Cobas EGFR Mutation Test v2.

Association Between Outcomes With Osimertinib in
NSCLC ± T790M as Detected by Plasma ctDNA vs Tissue
Oxnard. JCO. 2016;34:3375.
PFS by Tumor T790M Status PFS by Plasma T790M Status
PFS in Plasma T790M–Negative
Patients, by Tumor T790M Status
PFS in Plasma T790M–Positive
Patients, by Tumor T790M Status
Mos From First Dose
100
80
60
40
20
0
ProbabilityofPFS
0 3 6 9 12 15 18 21 24
Tumor T790M+ (n = 179)
Median PFS: 9.7 mos
(95% CI: 8.3-12.5)
Tumor T790M- (n = 58)
Median PFS: 3.4 mos
(95% CI: 2.1-4.3)
Log-rank test P < .001
Mos From First Dose
100
80
60
40
20
0
ProbabilityofPFS
0 3 6 9 12 15 18 21 24
Tumor T790M+ (n = 47)
Median PFS: 16.5 mos
(95% CI: 10.94-NC)
Tumor T790M- (n = 40)
Median PFS: 2.8 mos
(95% CI: 138-4.17)
Log-rank test P < .001
Tumor T790M unknown
(n = 17)
Mos From First Dose
100
80
60
40
20
0
ProbabilityofPFS
0 3 6 9 12 15 18 21 24
Plasma T790M+ (n = 169)
Median PFS: 9.7 mos
(95% CI: 8.3-11.1)
Plasma T790M- (n = 104)
Median PFS: 8.2 mos
(95% CI: 5.3-10.9)
Log-rank test P = .188
Mos From First Dose
100
80
60
40
20
0
ProbabilityofPFS
0 3 6 9 12 15 18 21 24
Tumor T790M+ (n = 111)
Median PFS: 9.3 mos
(95% CI: 8.25-10.94)
Tumor T790M- (n = 18)
Median PFS: 4.2 mos
(95% CI: 1.25-5.55)
Log-rank test P = .0002
Tumor T790M unknown
(n = 38)

Response to Osimertinib by Baseline EGFR T790M Status:
Tissue Biopsy vs Plasma ctDNA
Thress. Lung Cancer. 2015;90:509.
T790M+ T790M-
ORR (CR + PR)
25/41 24/41 7/24 11/31 40/41 37/41 17/24 25/31
DCR (CR + PR + SD)
Tissue biopsy
Plasma ctDNA
T790M+ T790M-
0
20
40
60
80
100
ClinicalResponse(%)
61%
59%
29%
35%
98%
90%
71%
81%
n/N =

Concordance Between Plasma cfDNA and
Tissue Biopsy DNA Measured by NGS
 Prospective study evaluating
suitability of NGS for analysis
of plasma cfDNA in patients
with EGFR mutation–positive
advanced NSCLC treated with
afatinib (N = 32)
 Mutations reliably detected
in plasma cfDNA by NGS
 However, if plasma is
negative, tissue biopsy is
reasonable
Iwama. Ann Oncol. 2017;28:136.
Patients With a Mutation (n)
Somatic Mutation Analysis by NGS in
Plasma cfDNA and Tumor DNA at Baseline
0 10 20 30
CTNNB1
TP53
EGFP
Tumor+/plasma+
Tumor+/plasma-
Tumor-/plasma+
2
7 4 1
23 9

Expanded Liquid Biopsy: FoundationACT ctDNA Assay
 Assay validated to detect 4 classes of genomic alterations in 62 genes
‒ Validation based on 267 samples from patients with solid tumors, 117 cell line
mixtures, and 42 synthetic DNA samples
 Apparent decrease in false negatives and false positives
FoundationACT ctDNA Assay Analytic Validation White Paper. 2016.
Specification, % MAF/Tumor Fraction Sensitivity Positive Predictive
Value
Base substitutions ≥ 0.5 > 98.9 > 99.9
Insertions/deletions (1-40 bp) ≥ 1 > 99.0 98.8
Rearrangements/fusions ≥ 1 > 99.0 98.0
Copy number variations* ≥ 20
< 20
95.3
Will vary depending on CNV level
and tumor fraction
97.6
Per-base specificity > 99.999
*In genes with at least 4 targets, copy number ≥ 8.

Analysis of Plasma cfDNA by NGS: A Potential Method to
Assess Response in EGFR-Mutated NSCLC
Patients With Progression
Between 4-24 Wks on Afatinib
Iwama. Ann Oncol. 2017;28:136.
Patients With Early
Progression < 4 Wk on Afatinib
6.0
5.0
4.0
3.0
2.0
1.0
0
1000
100
10
ND
Baseline PD
MutantAllele
FrequencybyNGS(%)
TP53 p.C277F
EGFR L858R
EGFR L858R (dPCR)
No.ofL858RMutant
Alleles(copies/mL)
80.0
60.0
40.0
20.0
0
25,000
20,000
15,000
ND
Baseline PD
MutantAllele
FrequencybyNGS(%)
No.ofExon19DelMutant
Alleles(copies/mL)
5000
10,000
4 Wks
TP53 p.S241Y
EGFR exon 19 del
EGFR exon 19 del (dPCR)

BLOOM Trial: Osimertinib in Advanced NSCLC With
Leptomeningeal Metastases
 Phase I study: planned N = 108; leptomeningeal cohort, n = 32
 Osimertinib dosed at 160 mg/day
‒ 23 patients had brain imaging assessment
‒ 10 with radiographic improvement
‒ 13 with stable disease
 At 12 wks of osimertinib
‒ 7/8 symptomatic patients improved
‒ 13/15 asymptomatic patients remained asymptomatic
Heymach. WCLC 2018. Abstr OA02.06.

BLOOM: Study Design—LM Cohort 1
Yang JC, et al. ASCO 2016. Abstract 9002.
 Efficacy assessments: OS, brain MRI and extracranial MRI or CT scan,*† CSF cytology,
neurological exam,* CNS symptoms*
 CT/MRI, CSF cytology and neurological exam every 6 wks
 1 cycle = 21 days of continuous dosing
Advanced or metastatic
NSCLC with confirmed
LM, EGFR L858R or
exon 19 deletion in
primary tumor, prior
EGFR-TKI treatment,
ECOG PS 0-2, stable
extracranial disease,
≥ 1 LM lesion by MRI
(N = 21)
Assessments
 AEs*
 Efficacy
 PK in CSF
 Quantification of
EGFRm DNA in CSF
*As assessed by study investigator. †RECIST for CNS disease; RECIST 1.1 for extracranial disease.
Osimertinib
160 mg PO QD

BLOOM: Efficacy
Yang JC, et al. ASCO 2016. Abstract 9002. Reproduced with permission.
Best Confirmed Neurological Status†
*Responses confirmed ≥ 4 wks after initial response. †Response assessed by neurological exam.
Pts(n)
Neurological Status at Baseline
21
18
15
12
9
6
3
0
Normal (n = 11) Abnormal (n = 10)
Improved
No change
Worsened
Early withdrawal
Unconfirmed
10
1
5
1 1
3
Response, n
N = 21
Confirmed* Unconfirmed
Best MRI imaging
intracranial response
 Responding 7 1
 Stable disease 9 2
CSF cytology clearance
 Responding 2
 Responding in
2 consecutive
samples
2
Improved neurological
function
5

BLOOM: Overall LM Response to Osimertinib in the
Evaluable Analysis Set (T790M-Unselected Cohort)
 Median DoR: 18.9 mos (range: 5.6-19.3 mos; 95% CI: 11.1-NC)
Investigator-Assessed Outcome T790M-Unselected Cohort
(n = 21)
LM response,* % (95% CI) 43 (22-66)
Best LM response, n (%)
 CR*
 Responding*
 SD ≥ 6 wks
 PD
 NE
1 (5)
8 (38)
9 (43)
1 (5)
2 (10)
*LM response defined as ≥ 1 confirmed CR or responding by investigator assessment, where confirmation done after 4 wks.
Slide credit: clinicaloptions.comYang. ASCO 2017. Abstr 2020.

BLOOM: Time on Treatment
 15 pts with treatment ongoing at time of data cutoff (March 10, 2016)
Yang JC, et al. ASCO 2016. Abstract 9002. Reproduced with permission.
*Pt death due to aspiration pneumonia.
Arrows represent observations at time of data cutoff
0 1 2 3 4 5 6 7 8 9 10 11 12 13
80 mg QD (dose reduction)
160 mg QD
Dose interrupted
T790M positive in the CSF
Mos Since Treatment Initiation
Pts
*
Pts with a confirmed
intracranial LM response (n
= 7)
Confirmed CSF clearance
Discontinued

Beyond Progression on TKIs: Preventing Disease Flare
267
 14/61 (23%)
experienced
disease flare
(hospitalization
or death due to
disease)
 Median TTP:
8 days
 Factors: short
time to PD,
pleural disease,
CNS disease
Age, Yrs Sex, Smoking Hx EGFR Mutation T790M Disease Site(s) Days Off TKI Description of Flare
47 Female, never del(19) No Pleura, brain, liver 21
Progressive liver metastases
with liver failure, death
64 Female, never del(19) No Pleura 21 Dyspnea
53 Female, never del(19) Unk Pleura, brain, bone 14 CNS progression
60 F, never Exon 21 L858R Unk Pleura, brain, bone 7 Hypoxia, bone pain
34 Male, never del(19) No Pleura, brain 11 Dyspnea
27 Female, never del(19) Unk Brain, liver, bone, pericardium 11 Bone pain
47 Female, never Exon 21 L858R Yes Pleura 7 New leptomeningeal disease
49 Male, never del(19) Yes Bone, liver 7 New brain metastases, seizure
61 Female, never Exon 21 L858R No Pleura, brain, liver, peritoneum 3 Abdominal pain
45 Female, never del(19) Yes Pleura, bone 8
New leptomeningeal
carcinomatosis, seizure, death
46 Female, former del(19) Yes Pleura, liver, bone 12 Epidural progression
62 Female, never del(19) Yes Pleura, bone, pericardium 8 Pericardial tamponade, death
42 Female, never
Exon 18 E709A
and G719A
Yes Pleura, bone 8
Acute pleural effusion
requiring drainage
67 Female, former Exon 21 L858R No Pleura, brain 8 Dyspnea
Chaft. Clin Cancer Res. 2011;17:6298.

EGFR TKI: Common Grade 3/4 AEs
Common
Grade 3/4 AEs, %
Erlotinib Gefitinib Afatinib Osimertinib Dacomitinib
Rash 13 3 16 1 21
Diarrhea 5 4 15 1 11
Fatigue 6 < 1 2 2 2
Afatinib PI. Dacomitinib PI. Erlotinib PI. Gefitinib PI. Osimertinib PI.

TIGER-X: Rociletinib in Previously Treated EGFR
Mutation-Positive NSCLC
Rociletinib 500 mg BID
(n = 119)
EGFR mutation– positive
previously treated advanced
or recurrent NSCLC with
acquired resistance to prior
EGFR TKI
(N = 456)
Key outcomes: safety and tolerability, PK profile, ORR
Sequist LV, et al. ASCO 2015. Abstract 8001.
(n = 236)
(n = 95)
Phase II expansion cohorts
 Upon progression on EGFR TKI
 T790M+ biopsy at entry
 Stable CNS metastases ok
Phase I dose escalation
Rociletinib BID
21-day cycles
escalate to MTD

TIGER-X: Tumor Response Across Rociletinib Dosing
Sequist LV, et al. ASCO 2015. Abstract 8001. Reprinted with permission.
100
80
60
40
20
0
-20
-40
-60
-80
-100
SLDChangeFromBaseline(%)
500 mg BID HBr
625 mg BID HBr
750 mg BID HBr
1000 mg BID HBr
Ongoing
Individual Patients

TIGER-X: Safety Across Rociletinib Dosing
 Improved safety profile with 500 mg BID vs higher doses
‒ Grade 3 QTc prolongation 2.5%
‒ Discontinuation due to treatment-related AEs 2.5% vs 4% overall
 Once recognized, hyperglycemia manageable with oral agents
Sequist LV, et al. ASCO 2015. Abstract 8001.
AE (All Grades), %
500 mg
(n = 119)
625 mg
(n = 236)
750 mg
(n = 95)
1000 mg
(n = 6)
Hyperglycemia
 Grade 3/4
35
17
45
24
59
36
67
33
Diarrhea 33 40 30 67
Nausea 19 34 37 50
QTc prolongation 13 23 26 50

TIGER-X: Plasma Testing for EGFR Mutations
 Plasma EGFR mutation assessed via digital PCR/flow cytometry assay shows good sensitivity and
specificity vs standard tissue testing
‒ 81% agreement for T790M and 87% for activating mutations
‒ Identified several T790M+ samples missed by tissue testing
‒ T790M status confirmed in subsequent tissue testing
‒ Plasma-based test overcomes limitations of tissue specimen availability
 Similar ORR in samples identified as T790M+ with plasma vs tissue
‒ 53% (78/147) vs 53% (85/160)
 Rociletinib also active in subset of pts confirmed T790M WT
‒ ORR: 32% to 39%
Sequist LV, et al. ASCO 2015. Abstract 8001. Reprinted with permission.

TATTON: EGFR + MET Inhibition in Previously Treated
EGFR Mut+/MET+ NSCLC
 Phase Ib study
Patients with
EGFR mut+, MET+*
advanced NSCLC,
with progression on
≥ 1 EGFR TKI;
WHO PS 0/1
(N = 66)
Received third-generation
T790M-directed EGFR TKI
(n = 30)
T790M- w/o prior third-generation
(n = 24)
T790M+ w/o prior third-generation
(n = 12)
Osimertinib 80 mg +
Savolitinib 600 mg PO QD
Ahn. WCLC 2017. Abstr OA 09.03.
Treatment
beyond PD
per
investigator
discretion
 Primary endpoint: safety/tolerability
 Secondary endpoints: preliminary assessment of ORR, DoR, change in tumor size;
pharmacokinetics
*MET+ status was to be confirmed centrally by
FISH (MET gene copy ≥ 5 or MET/CEP7 ratio
≥ 2), but also included local FISH, IHC (+3 in
≥ 50% of tumor cells), or NGS.

EGFR + MET Inhibition in EGFR Mut+/MET+ NSCLC:
Efficacy
Ahn. WCLC 2017. Abstr OA 09.03.
Best%ChangeFrom
BLinTumorLesionSize
100
75
50
25
0
-25
-50
-75
-100
Prior third-generation T790M-directed EGFR TKI
No prior third-generation EGFR TKI, T790M+
No prior third-generation EGFR TKI, T790M-
ORR, n (%) Prior Third-
Generation T790M-
Directed EGFR TKI
(n = 30)
No Prior Third-Generation
T790M-Directed EGFR TKI
Total
(N = 64)
T790M+
(n = 11)
T790M-
(n = 23)
ORR 10 (33) 6 (55) 14 (61) 30 (47)

TATTON (Part B): Osimertinib + Savolitinib in Previously
Treated EGFR Mut+/MET-Amplified NSCLC
 Open-label phase Ib study
Patients with
EGFR mut+, MET+*
advanced NSCLC,
with progression on
≥ 1 EGFR TKI;
WHO PS 0/1
Cohort A
Received first-/second-generation
EGFR TKI (T790M negative)
(n = 46)
Cohort B
Received third-generation EGFR TKI
(n = 48)
Osimertinib 80 mg +
Savolitinib 600 mg PO QD
Yu. AACR 2019. Abstr CT032. Sequist. AACR 2019. Abstr CT033.
Treatment
beyond PD
per
investigator
discretion
 Primary endpoint: safety/tolerability
 Secondary endpoints: preliminary assessment of ORR, DoR, change in tumor size;
pharmacokinetics
*MET+ status was to be confirmed centrally by FISH (MET
gene copy ≥ 5 or MET/CEP7 ratio ≥ 2), but also included
local FISH, IHC (+3 in ≥ 50% of tumor cells), or NGS (≥ 20%
tumor cells, ≥ 200x seq depth coverage, and ≥ 5 copies over
tumor ploidy).

TATTON: Preliminary ORR With Osimertinib +
Savolitinib After a First-/Second-Generation EGFR TKI
 Median duration of response: 7.1 mos
Yu. AACR 2019. Abstr CT032.
OSI + SAVO: Prior 1st/2nd-Gen EGFR-TKI, T790M- (n = 46)
Objective response, n (%)
CR
PR
24 (52)
0
24 (52)
Nonresponse, n (%)
Stable disease (≥ 6 wks)
Progressive disease
Not evaluable
22 (48)
16 (35)
3 (7)
3 (7)
Median time to response, days (IQ range) 43 (40-43)
100
80
60
40
20
0
-20
-40
-60
-80
-100
BestChangeinTargetLesionSize(%)
+ = central MET FISH positive
= MET status not centrally confirmed

 Median duration of response: 9.7 mos
TATTON: Preliminary ORR With Osimertinib +
Savolitinib After a Third-Generation EGFR TKI
Sequist. AACR 2019. Abstr CT033.
OSI + SAVO: Prior 3rd-Gen EGFR TKI (n = 48)
Objective response, n (%)
CR
PR
12 (25)
0
12 (25)
Nonresponse, n (%)
Stable disease (≥ 6 wks)
Progressive disease
Not evaluable
36 (75)
21 (44)
6 (13)
9 (19)
Median time to response, days (IQ range) 46 (43-51)
100
80
60
40
20
0
-20
-40
-60
-80
-100
BestChangeinTargetLesionSize(%)
+ = central MET FISH positive
= MET status not centrally confirmed

Many Osimertinib Combinations in Phase I/II
Development for Advanced EGFR Mut+ NSCLC
Combination Partner MoA Patient Population Est. N
Bevacizumab[1] Anti-VEGF First line (TKI naive) 58*
Bevacizumab[2] Anti-VEGF Second line after prior TKI; osimertinib
naive; T790M+; brain mets
98*
Itacitinib[3] JAK1 inhibitor Second line after prior TKI; second phase
requiring T790M+
60*
Ramucirumab[4] Anti-VEGFR2 Second line after prior TKI; T790M+ 74
Gefitinib[5] First-gen EGFR TKI (C797S) First line 64*
Necitumumab[4,6] Anti-EGFR Second line after prior TKI; T790M+[4]
Second line after prior TKI; multicohort[6]
74
82*
Selumetinib[7] MEK inhibitor First line 25*
DS-1205c[8] Axl inhibitor Second line after prior TKI; T790M- 118*
Navitoclax[9] Bcl2/Bcl-xL inhibitor Second line after prior TKI; T790M+ 50*
Sapanisertib[10] TORC1/2 inhibitor Second line after prior TKI; T790M- 36*
1. NCT02803203. 2. NCT02971501. 3. NCT02917993. 4. NCT02789345. 5. NCT03122717.
6. NCT02496663. 7. NCT03392246. 8. NCT03255083. 9. NCT02520778. 10. NCT02503722.
*Recruiting as of March 2019.

What About Single-Agent Immune Checkpoint Inhibitor
Therapy in EGFR Mutation–Positive NSCLC?
 EGFR mutation only
clinical variable
associated with no
benefit
 Has led to expert
recommendation
to defer single agent
immunotherapy to
late salvage, if ever,
in EGFR mutation–
positive NSCLC
Lee. JAMA Oncol. 2018;4:210.
EGFR Subgroup, Trial HR (95% CI)
EGFR wild type
 OAK 0.69 (0.57-0.83)
 CheckMate 057 0.66 (0.51-0.85)
 KEYNOTE-010 0.66 (0.55-0.79)
 POPLAR 0.70 (0.47-1.04)
 Subtotal 0.67 (0.60-0.75)
EGFR mutated
 OAK 1.24 (0.71-2.18)
 CheckMate 057 1.18 (0.69-2.02)
 KEYNOTE-010 0.88 (0.45-1.72)
 POPLAR 0.99 (0.29-3.40)
 Subtotal 1.11 (0.80-1.53)
Favors PD-1/
PD-L1 Inhibitor
Favors
Docetaxel
HR (95% CI)
0.2 1.0 4.0
Weight, %
32.6
16.2
33.5
7.1
89.4
3.5
3.9
2.5
9.7
10.6

What About Single-Agent Immune Checkpoint Inhibitor
Therapy in EGFR Mutation–Positive NSCLC?
 Phase II study of first-line pembrolizumab in advanced EGFR mutation–
positive NSCLC with PD-L1 ≥ 1% and no prior TKI (planned N = 25)
 Study stopped for futility at 11 patients
‒ 7 with canonical EGFR mutations, 2 with exon 20 insertion, 1 with E330K
mutation, and 1 where EGFR del19 mutation was identified in error
 Only 1/11 patients responded
‒ . . . and the response was in the patient where EGFR mutation was an
error
‒ . . . and despite 8/11 patients having PD-L1 ≥ 50%
Lisberg. ASCO 2018. Abstr 9014.

Garassino. Lancet Oncol. 2018;19:521.
PD-L1 Low/Negative (< 25%)
ALK+ only
EGFRmut only
100
BestChangeFromBaseline
inTargetLesionSize(%)
80
60
40
20
0
-20
-40
-60
-80
-100
ORR*: 3.6% (95% CI: 0.1% to 18.3%)
PD-L1 High (≥ 25%)
ALK+ only
EGFRmut only
100
BestChangeFromBaseline
inTargetLesionSize(%)
80
60
40
20
0
-20
-40
-60
-80
-100
ORR*: 12.2% (95% CI: 5.7% to 21.8%)
ATLANTIC: Response to Durvalumab in Previously Treated
Patients With EGFR mutation-positive NSCLC by PD-L1 level
*Confirmed ORR by independent central review

Pembrolizumab in TKI-Naive Patients With EGFR
Mutation–Positive, PD-L1–Positive (≥ 1%) Adv NSCLC
 Single institution phase II trial
 11 patients enrolled (of planned 25)
‒ 9 patients treatment-naïve
‒ 7 patients with sensitizing EGFR mutations
‒ 8 patients with PD-L1 expression ≥ 50% (22C3 assay)
 1 PR (but not EGFR mutation positive on repeat analysis)
 2/11 died within 6 mos (1 with pneumonitis)
Lisberg. J Thorac Oncol. 2018;13:1138.

What About Combination Immune Checkpoint Inhibition
+ Chemotherapy in EGFR Mutation–Positive NSCLC?
 Carboplatin/pemetrexed ± bevacizumab is a standard first-line
chemotherapy regimen for non-squamous NSCLC with EGFR mutations
 KEYNOTE-189: first-line carboplatin/pemetrexed ± pembrolizumab in
metastatic non-squamous NSCLC
‒ Results: improved OS (HR: 0.49) and PFS (HR: 0.52) with addition of
pembrolizumab
‒ However, patients with EGFR mutations were excluded from this
phase III trial
Gandhi. NEJM. 2018;378:2078.

Role of Chemotherapy, Targeted therapy and Immunotherapy in NSCLC Part I

In this document

More Related Content

What's hot

Similar to Role of Chemotherapy, Targeted therapy and Immunotherapy in NSCLC Part I

More from Mohammed Fathy

Recently uploaded

Role of Chemotherapy, Targeted therapy and Immunotherapy in NSCLC Part I