Non-small Cell Lung Cancer

1
Non-small Cell Lung Cancer

• Eva Szabo, MD
• Division of Cancer Prevention, NCI

TRACO 11-14-11
2
US Lung Cancer Statistics, 2010227,890 estimated
new cases

• 158,070 estimated deaths
• leading cause of cancer deaths
• greater than breastprostatecolon
• Death rate per 100,000 decreasing (90.56 in 1990
  67.45 in 2006)
• Still increasing in women (37.61 in 199140.17 in
  2006)
• 16 five year survival
• 5 in 1950s, 13 in 1970s
• 29 of all male cancer deaths, 26 of all female
  cancer deaths

3
Tobacco Use in the US, 1900-2003
Per capita cigarette consumption
Male lung cancer death rate
Female lung cancer death rate
4
Radiographic Evidence Linking Tobacco Use to Lung
Cancer -?McMullen, DM Cohen GA, NEJM 354397,
2006
5
Risk Factors

• Tobacco, tobacco, tobacco (85 lung ca.)
• Including passive smoking
• Prior aerodigestive malignancy
• COPD
• Other exposures
• Asbestos, radon, polycyclic aromatic
  hydrocarbons, chromium, nickel, inorganic arsenic
  mining, ship building, oil refining
• Genetic predisposition
• Familial lung cancer 6q23-25 (Am J Hum Gen,
  9/04)
• 15q24-25.1 nicotinic acetylcholine receptor
  subunits CHRNA3 and CHRNA5, OR1.3, attributable
  risk 14
• Amos et al., Nat Gen 200840616, Hung et al.
  Nature 2008452633, Thorgeirsson et al. Nature
  2008452638
• CH3NA3/5 is also susceptibility locus for COPD
• Pillai et al. PLoS Genet 200951

6
Five-year survival by TNM status in NSCLC (old
staging system). ?In stage 1A, TNM
classification T1N0M0 the 5 year survival is
61. ? In stage 1BA, TNM classification T2N0M0
the 5 year survival is 38.?In stage IIA, TNM
classification T1N1M0 the 5 year survival is
34.?In stage IIB, TNM classification T2N1M0 or
T3N1M0 the survival is 24.?In stage IIIA, TNM
classification T1-3NanyM0 or TanyN3M0 the 5 year
survival is 13.?In stage IIIB, TNM
classification T4NanyM0 or T3N1M0, the 5 year
survival is 5.?In stage IV, TNM classification
TanyNanyM1 the 5 year survival is 1.
7
Pathology Non-small Cell Lung Cancer
?Adenocarcinoma, inc
bronchoalveolar 40 ? Squamous cell carcinoma
20 ? Large cell carcinoma 15 ? Others
(carcinoid, etc.)
8
Pathology Small Cell Lung Cancer
? Small cell lung
cancer - 20
9
The Continuum of Lung Carcinogenesis
Opportunities for Intervention Normal to
Hyper/Metaplasia to Dysplasia to Early-Late
Cancer
10
Treatment Strategies for Lung Cancer

• Treatment based on stage
• Early stage (Stage I) surgery
• Early stage (Stage II, IIIA resected)-surgery
  adjuvant chemo
• Regional spread (IIIA/IIIB) combined modality
  (chemoradiation, /- surgery for IIIA)
• Metastatic (IIIB wet/IV) chemotherapy,
  radiation as needed for local control, occasional
  resection of isolated mets
• Small cell lung cancer chemotherapy (thoracic
  radiation for limited stage prophylactic cranial
  radiation to prevent brain mets)

11
Personalizing Therapy for NSCLCSignificantly
mutated genes in adenocarcinomas
12
Epidermal Growth Factor (EGFR) Signaling
13
EGFR as a Target for NSCLCStandard of Care in
2010

• Epidermal growth factor receptor (EGFR)
  inhibition in advanced NSCLC
• 10 response rate in advanced disease, 30
  prolonged stabilization
• Survival advantage (erlotinib)
• Shepherd, F. A. et al. N Engl J Med
  2005353123-132
• Mutually exclusive with K-ras
• Most benefit for non-smoking related NSCLC, with
  EGFR mutations (females, adenocarcinomas, Asian)
• Lynch et al., NEJM 3502129, 2004 Paez et al.,
  Science 3041497, 2004 Pao et al., PNAS
  10113306, 2004
• Mechanisms of secondary resistance to EGFR
  inhibitors being identified (T790M mutation-50,
  Met amplification-20)
• Pao et al., PLoS Med 2e17, 2005 Engelman et
  al., Science 3161039, 2007
• Erlotinib approved as single agent for 2nd and
  3rd line treatment of NSCLC
• Also for maintenance after 1st line
  non-progression after chemo
• (gefitinib formerly approved in US, failed to
  show survival advantage)

14
Effect of Gefitinib on Progression-free Survival
and Overall SurvivalMaemondo M et al. N Engl J
Med 20103622380-2388 230 patients, EGFR
mutated, randomized to gefitinib or
chemo(carbo/taxol)Results10 vs. 5 mth
PFS74 vs 31 response Median survival 30.5 vs
23.6 mths
15
EML4-ALK Fusion Gene as a Target for NSCLC

• Identified in 2007
• 5 NSCLC, mainly never smokers
• Often with distinct signet cell histology
• Striking response to inhibitor crizotinib- 57
  RR, 33 stable disease
• Kwak EL et al. NEJM 20103631693

16
HER2/neu as a Target for NSCLC

• Mutations in kinase domain in 4 NSCLC
• Hunter et al., Nature 200430431
• Amplification by FISH in 2-5
• Heinmoller P et al. Clin Cancer Res 200395283
• IHC 30
• Langer et al., JCO 2004221180
• Trastuzumab in NSCLC with 2/3 IHC for Her2 not
  active (11 of patients eligible, 1/24 PR)
• Clamon et al., Cancer 20051031670

17
HER2/neu as Target for NSCLCKelly R J et al. JCO
201028e507-e510

• 50 yo AA nonsmoker with stage IV NSCLC (large
  cell) in 11/07
• Rx with carbo/taxol/bevacizumab (PR), erlotinib
  (PD),
• PF-00299804 (pan-HER inh,
• brief PR)
• -found to be HER2 amplified
• -Trastuzumab ?PD
• -Trastuzumab/Vinorelbine?
• near CR x 13 months

18
Advanced Disease Targeted TherapiesVEGF
Inhibition

• Bevacizumab (anti-VEGF antibody) in advanced
  NSCLC frontline treatment in combination with
  chemotherapy (taxol/carbo) E4599 (Sandler et
  al. NEJM 20063552542)
• Median survival 12.3 vs. 10.3 mths
• Response rate 27 vs. 10
• Time to progression 6.4 vs. 4.5 mths
• In non-squamous cancers only (life-threatening
  bleeding higher in squamous cancers)

19
Controversies in NSCLC Treatment

• Choice of agents?
• Platinum vs. not (platinum preferred)
• Single vs. two vs. three agents (2 conventional
  chemos)
• Treatment of elderly Yes if good performance
• Length of treatment probably no more than 4-6
  cycles of cytotoxic conventional chemo
• Second line treatment yes
• Taxotere and pemetrexed better than supportive
  care
• Erlotinib (EGFR inhibitor) for 2nd or 3rd line

20
Controversies in NSCLC Treatment

• Maintenance chemotherapy after first line
  treatment
• Pemetrexed is FDA approved for maintenance in
  non-squamous NSCLC
• Overall survival 13.4 vs. 10.6 mths for all
  subtypes, 15.5 vs. 10.3 mths for non-squamous
  histologies
• Ciuleanu T et al. Lancet 200937414432
• Erlotinib (SATURN trial) - 1 mth improvement in
  survival (12 vs 11 mths)
• Data presented b Cappuzzo et al. at 13th World
  Congress on Lung Cancer, 8/09, San Francisco, CA

21
Approaches to reducing cancer morbidity and
mortality

• Prevention (primary, secondary, tertiary)
• Early detection
• Better therapeutics

22
The Continuum of Lung Carcinogenesis
Opportunities for Intervention Normal to
Hyper/Metaplasia to Dysplasia to Early-Late
Cancer
23
Primary Prevention

• Smoking cessation
• Decline in California lung cancer rates
  1988-1997 declined 14, compared with 2.7 in
  non-California SEER sites, coincident with
  declining smoking rates probably due to
  California tobacco control initiatives
• Cowling DW et al., MMWR 491066-9, 2000

24
Effect of Smoking Cessation on Lung Cancer
DeathsLung Health Study, 14.5 yr F/U
25
Cancer Chemoprevention
The use of natural or synthetic agents to
suppress or reverse carcinogenesis

• Regress existing neoplastic lesions (treat
  intraepithelial neoplasia)
• Prevent development of new neoplastic lesions
  (preneoplastic and cancer)
• Suppress recurrence of neoplastic lesions

26
Rationale for Lung Cancer Prevention

• Metastatic cancer is rarely curable
• US lung cancer 5 yr survival is 15 (5 1950s,
  13 1970s)
• Cancer is preventable
• P1, STAR breast cancer prevention trials with
  tamoxifen and raloxifene (Fisher B et al., JNCI
  19981901371 Vogel, VG et al., JAMA
  20062952727)
• Multiple animal studies with multiple agents
• Long preclinical phase with increasing histologic
  and molecular abnormalities, identifiable
  populations at risk

27
When is the best time to intervene during
carcinogenesis?

• Efficacy of intervention
• Early stage cancer is more curable than late
• Are precursor lesions more curable than invasive
  cancer?
• Can carcinogen-induced DNA damage be prevented?
• Multiple pathways of carcinogenesis
• Toxicity of intervention
• High toxicity acceptable short-term, in setting
  of cancer
• Target population size and ability to identify
• Many at risk (smokers), relatively few get
  cancer/yr
• Inability to identify non-smokers at risk
• Cost (resources, psychological impact, etc.)

28
Minimal Requirements for Preventive Strategies

• Benefit
• Efficacy in preventing cancer and associated
  morbidity/mortality
• Risk
• Lack of adverse side effects that increase
  morbidity/mortality from other diseases, short-
  and long-term (major side effects)
• Tolerability of intervention (minor side effects
  affecting compliance)

29
Efficacy How Do We Identify New Agents?

• Knowledge of mechanism
• Example HPV vaccine and cervical cancer
• Need understanding molecular pathogenesis
• Preclinical (in vitro and animal models)
• Example NSAID treated carcinogenesis and
  transgenic models
• Need models reflective of complexity of human
  disease
• Observational epidemiology (cohort and
  case-control studies)
• Example NSAIDs and colon cancer
  incidence/mortality
• Secondary endpoints from clinical trials
  (including other diseases)
• Example Tamoxifen/raloxifene and breast cancer

30
Optimizing the Risk-Benefit Balance

• Identify individuals most likely to develop
  cancer in short time frame
• Highest risk
• Homogeneous cohorts (current vs. former smokers,
  FAP vs. HNPCC vs. sporadic colorectal adenomas)
• Pharmacogenetic considerations
• Minimize toxicities from drug (e.g., route,
  schedule, modulators of toxicity)
• Barriers
• Lack of adequate risk assessment models for most
  cancers
• Incomplete understanding of carcinogenesis at
  different target organs

needed to treat
NL Initiated Dysplasia Carcinoma
Stages of Carcinogenesis
31
Targeting Inflammation for Lung Cancer
Prevention Rationale

• Animal data showing role for steroids in cancer
  prevention
• 1970s skin
• Early 1990s lung (oral steroids)
• Late 1990s lung (inhaled steroids)
• Epidemiology/Human data
• Mainly negative (but studies of short exposure
  duration)
• VA cohort with COPD (n10,474) HR 0.39 (95 CI,
  0.16-0.96)
• Parimon T et al., AJRCCM 175712, 2007

32
Effect of Budesonide on Mouse Lung Tumorigenesis
Pereira et al., Carcinogenesis 2002 -
? 82 decrease in
tumors
? -Shift from adenoma to carcinoma
33
DCP Phase IIb Trial of Budesonide Lam et al.
Clin Cancer Res 106502, 2004 ?112 smokers with
dysplasia (Bronch)?(Spiral CT) Budesonide vs.
Placebo x 6mths (Bronch,Spiral CT)
?1o Endpoint bronchial dysplasia
(sites/grade)2o Endpoints multiple biomarkers
? Screened (sputum)
1040Cancers detected 13 (3.1)
34
Effect of Budesonide on Bronchial Histology

35
Phase IIb Budesonide Chemoprevention TrialPI
Giulia Veronesi, European Institute of Oncology
? 202 participants with
persistent spiral CT-detected peripheral nodules
(Randomize)? ? inhaled budesonide vs. placebo x
1 year? ? repeat spiral CT ?Primary endpoint
shrinkage of lung nodules
36
Phase IIb Budesonide Chemoprevention TrialLesion
Specific Analysis -Overall response negative,
but trend toward regression in non-solid lesions
(putative precursors of adenocarcinoma)
37
Adenocarcinoma Precursor Atypical Adenomatous
Hyperplasia

• Natural history unknown
• Localized ground glass opacities on CT
• AAH 25 bronchoalveolar ca 50 invasive
  adenoca 10 fibrosis 15 (Nakajima et al., J
  Comput Assist Tomogr 200226323)
• AAH 63 bronchoalveolar ca 34 scar 3 (Ohtsuka
  et al., Eur J Cardio-Thor Surg 200630160)

38
Where do we go from here with steroids?

• Potential reasons for negative trial
• Intervention doesnt work (animal and human data
  in conflict)
• Formulation does not penetrate peripherally
  enough
• Intervention given too late
• Intervention focused on wrong cohort (next study
  to focus on ground glass opacities only)
• Plan further molecular characterization of
  AAH/ground glass opacities
• Not in current plan phase III

39
myo-Inositol

• Glucose isomer
• Source of several second messengers signaling
  molecules
• Dietary sources (grains, beans, fruits, rice)
• Studied in psychiatric conditions (/-), diabetic
  neuropathy(/-), polycystic ovary syndrome ()

40
Rationale for myo-Inositol in Lung Cancer
Prevention

• Efficacy
• Multiple animal studies show inhibition of
  carcinogen induced tumors in mice (40-50)
• Estensen and Wattenberg, Carcinogenesis
  1993141975
• Hecht et al., Carcinogenesis 2002231455
• Inhibits carcinogenesis in mainstream/sidestream
  smoke-exposed A/J mice by 53
• Witschi H et al., Carcinogenesis 1999201375
• Combination with budesonide ?? efficacy up to 80
  
• Estensen and Wattenberg, Carcinogenesis
  1993141975
• Witschi et al. Carcinogenesis 1999201375
• Wattenberg et al. Carcinogenesis 200021179
• Safety
• Used in multiple short term trials for
  psychiatric and diabetic neuropathy indications
  no toxicity reported
• Generally Regarded as Safe (GRAS) by US FDA
  terminology

41
Phase I Study of Myo-inositol in Bronchial
Dysplasia-Lam et al., CEBP 2006151526
?Inhibits BaP carcinogenesis in mice
(53) combination with budesonide ???Phase I
study (26 participants)tolerable 18 g/d91 vs.
48 regression dysplasia, P0.014 (10
participants)
42
PI3K pathway activation in the airways of smokers
with dysplasiaGustafson A M et al. Sci Transl
Med 2010226ra25-PI3K pathway is activated in
smokers with dysplasia in airway
plt0.001-Myo-inositol inhibited PI3K activation
in normal bronchial airways in smokers with
regression of dysplasia (p0.04)
43
Why is this study so important?

• Does PI3K activation truly identify smokers at
  risk for cancer?
• Easier to get normal brushings than to identify
  dysplasia (sampling bias) do not remove
  biomarker with procedure
• Potential to identify the right cohort
• New potential clinical trial model pathway
  analysis pre- and post-treatment, smaller
  participants, shorter interventions
• Identify mechanisms of interventions
• Needs validation!

44
Phase IIb myo-Inositol Chemoprevention TrialPI
Stephen Lam, British Columbia Cancer Agency30
pack yr. Get Bronch, Spiral CT N110myo-inosit
ol 9g bid vs. placebo x 6 mths Get Bronch, Spiral
CT1o Endpoint bronchial dysplasia (
sites/grade)2o Endpoints multiple biomarkers
(gene expression) Clinical sites BCCA, Mayo
Clinic, New Mexico VA
45
Peroxisome Proliferator-Activated Receptor ?
(PPAR ? ) as a Target for Prevention of
Aerodigestive Carcinogenesis

• Pioglitazone PPAR? agonist approved for type II
  DM
• Rationale
• Cell lines induces growth arrest,
  differentiation (NSCLC)
• Animal carcinogenesis models
• 4-NQO rat tongue model incidence and
  multiplicity ? 10-fold
• Yoshida et al., Cancer Sci 94365, 2003
• Epidemiology
• 33 ? lung cancer in diabetics using TZDs
  (RR0.67 95 CI, 0.51-0.87) Nonsignificant
  decrease in colon and prostate cancer
• Govindarajan et al. JCO 2007251476-81
• 41-55? HNSCC in diabetics using TZDs
• Govindarajan R et al. JCO 20072563s

46
Pioglitazone in Oral Leukoplakia

• DCP phase IIa clinical trial - 22 pts., 81
  clinical response rate, 79 average ?size
• F. Ondrey, U Minn
• AACR Frontiers Cancer Prev Res, 2007
• New trial phase IIb, 100 person, 6 mths Rx
• PIs J Boyle, MSKCC and F Ondrey, U Minn

pre post
47
Effect of PPAR? Agonists on NSCLC Animal
ModelsTreatment causes tumor volume ?
66.7-growth delay 104 days. Prevention
causes56-64 ? in tumor burden in wildtype and
p53 mutant animals Ming You et al. Mol Cancer
Ther 2010930742
48
Current Pioglitazone Clinical Trials (NCI
sponsored) Phase IIb oral leukoplakiaPioglitazone
45 mg qd vs placebo x6 months100 participants
11 sites1o Endpoint clinical and pathologic
responsePIs Jay Boyle, MSKCC and Frank Ondrey,
UMinn Pilot trial presurgical NSCLC
trialPioglitazone 45 mg qd for 2-6 weeks prior
to definitive surgery 20 participants biomarker
endpointsPI Dennis Wigle, Mayo
49
The Continuum of Lung CarcinogenesisOpportunities
for Intervention
50
For it happensthat in the beginning of the
malady it is easy to cure but difficult to
detect, but in the course of time, not having
been either detected or treated in the beginning,
it becomes easy to detect but difficult to
cure.-N. Machiavelli, The Prince
51
Issues in Lung Cancer Screening

• Lead-time biasearlier diagnosis but no
  postponement of death (survival appears longer)
• Length biasdiagnosis of more indolent disease
  with longer preclinical phase (better prognosis,
  better outcome)
• Overdiagnosisidentification of clinically
  unimportant lesions that would not be diagnosed
  otherwise
• Morbidity/mortality/cost of screening and
  subsequent work-up

52
PLCO CXR Randomized Trial - MortalityOken, M. M.
et al. JAMA 20113061865-1873
53
Helical CT for Early Lung Cancer Detection

• ELCAP (Lancet 199935499)
• Low dose helical CT in 1000 asymptomatic smokers
• Results
• 2.7 lung cancers diagnosed by CT vs. 0.7 by CXR
• 85 cancers were stage I vs. 22 expected
• 96 were resectable
• I-ELCAP (NEJM 20063551763-71)
• Low dose helical CT 31,567 asymptomatic screened
  at-risk individuals (inc. second hand smoke
  exposed)
• Results
• 85 stage I, estimated 10-yr survival 88

54
NLST (National Lung Screening Trial) NLST
Research Team. N Engl J Med 2011365395-409

• NLST design
• 53,454 smokers (current and former)
• 30 pack-yr smoking hx quit lt15 yrs ago
• Age 55-74
• Helical CT vs. chest X-ray (prevalence, then x2)
• NLST results
• CT - 24.2 positive tests, 354 lung cancer
  deaths
• CXR 6.9 positive tests, 442 lung cancer
  deaths
• 20.0 reduction in lung cancer mortality
• 6.7 reduction in all cause mortality

55
Cumulative Numbers of Lung Cancers and of Deaths
from Lung Cancer
56
NLST Histology
57
An ounce of preventionis worth a pound of
cure -Benjamin Franklin