Autoantibody Reactivity to Tumor Associated Antigens as a Biomarker for Early Lung Cancer

1
Autoantibody Reactivity to Tumor Associated
Antigens as a Biomarker for Early Lung Cancer
J Hung1, M Jagen1, AK Greenberg1, E Tan2, D
Naidich1, H Steck3, G Fung3, M Salganikoff 3, B
Rao3, BK Phalan1, E Eylers1 and WN Rom1 NYU
School of Medicine1, The Scripps Research
Institute2 and Siemens Medical Solutions,
Inc.3 Funded by NIH U01CA86137, M01RR-00096
Methods
Introduction
Summary of Results
Results
Percent of Lung Cancer Cases and Smoking Controls
with positive ELISA (defined as gt 2x NHC mean
3 std dev)
Patient Demographics

• Lung cancer is the leading cause of death among
  all cancer types.
• Early detection of lung cancer in its earliest or
  even pre-neoplastic form may lead to a higher
  cure rate.
• Some preneoplasias may be visualized on CT scan
  as ground glass opacities (GGOs).
• Studies have demonstrated that cancer patients
  sera contain antibodies that reacts with
  autologous intracellular proteins.These proteins
  are known as tumor associated antigens (TAA)

• Sera from individuals enrolled in the NYU Lung
  Cancer Biomarker Center (NYU LCBC) of NCI Early
  Detection Research Network were analyzed for
  reactivity to tumor associated antigens (TAAs)
• Patient population
• Individuals were categorized in the following
  groups
• Non-smoking healthy control (NHC), n40.
• Lung cancer patients, n17.
• Smokers with solid nodules (SN) seen on initial
  screening chest CT, n49.
• Smokers with GGOs seen on initial screening chest
  CT, n47.
• Smokers with normal CT (Smoking controls), n45.
• The cancer patients and smokers with GGOs were
  consecutive cases enrolled in the NYU LCBC.
  Smoking controls and individuals with stable
  solid noduels were matched for age and smoking
  history to the cancer and GGO patients
  (combined).
• Smokers were defined as individuals with gt 20
  pack years of cigarettes smoking history

• Smoking controls had a higher level of reactivity
  to a panel of TAAs than non-smoking controls.
• When compared with smoking controls, reactivity
  to c-myc and p62 was significantly elevated in
  both patients with lung cancer and in individuals
  with possible preneoplastic lesions (GGOs). Both
  the mean levels of reactivity were increased, and
  the number of individuals with reactivity above a
  cut-off value (gt2x mean normal 3SD) were
  increased.
• In individuals with GGOs, significantly more
  people also had increased reactivity to cyclin B1
  and decreased reactivity to cyclin D1 compared
  with smoking controls.
• Individuals with stable solid nodules (presumably
  benign) also had increased reactivity to c-myc
  compared with smoking controls

Smoker Normal CT SN GGO Lung Cancer
Age 57.18.6 58.48.7 58.08.9 68.212.4
Male 45 59 60 47
Pack/ Yrs 46.126.3 47.632.1 49.023.8 56.131.3
Fev1/ FVC 71.38.4 74.48.6 74.07.3 58.714.6
p-value lt0.05 when compared to Smokers with
normal chest CT
Mean ELISA Results in Optical Density
Percent of Smokers with GGOs and Smoking Controls
with positive ELISA (defined as gt 2x NHC mean
3 std dev)
NHC Smoker Normal CT SN GGO Lung Cancer
p53 .020.012 .097.109 .197.500 .105.234 .076.108
c-myc .115.142 .113.109 .065.093 .174.151 .286.222
IMP 1 .063.150 .093.294 .045.037 .074.051 .099.106
p62/IMP 2 .166.372 .065.043 .250.466 .240.478 .261.484
IMP 3/Koc .138.181 .160.149 .160.124 .110.077 .309.459
Cyclin A .105.126 .163.214 .093.053 .062.025 .202.075
Cyclin B1 .009.011 .048.141 .010.142 .124.202 .053.212
Cyclin D1 .054.041 .219.383 .104.090 .029.005 .157.421
CDK 2 .002.037 .047.015 .044.017 .069.074 .148.203
Survivin .063.058 .074.033 .050.013 .038.012 .105.142

• GGOs are defined as nodules with hazy attenuation
  without obscuration of underlying vascular
  markings.
• In high risk patients, ground-glass opacity
  nodules (GGO) and solid nodules are commonly seen
  on Chest CTs.
• In one study (Nakata et al., Chest 2002) all the
  GGOs 2 cm that persisted for 3 months or more
  were malignant or premalignant lesions. Of 43
  persistent focal GGOs, 53 were bronchioalveolar
  carcinoma (BAC), 26 were adenocarcinoma and 21
  were preneoplasias (atypical adenomatous
  hyperplasiaAAH).
• In this study, we determined level of
  autoantibody to a panel of TAAs in individuals
  with lung cancer, GGOs, solid nodules and normal
  CTs.

Ground Glass Nodule
Conclusion
p-value lt0.05 when compared to non-smoking
healthy control

• Our results suggest a pattern of autoantibody
  reactivity to a panel of TAAs may distinguish
  patients with lung cancer and preneoplastic
  lesions from smokers with normal CTs or benign
  lesions.
• The finding of a similar pattern of reactivity in
  both patients with lung cancer and in individuals
  with possible preneoplastic lesions, suggests
  that this assay may prove useful for the
  identification of these very early stages of lung
  cancer.
• Further studies, with increased numbers and
  perhaps additional TAAs, are needed to validate
  these results and increase the sensitivity and
  specificity of the assay.

Percent of Smokers with SNs and Smoking Controls
with positive ELISA (defined as gt 2x NHC mean
3 std dev)

• Enzyme-linked immunosorbent assay (ELISA) to TAAs
  were performed on sera
• Tumor associated antigens (TAA)
• P53
• c-myc
• Insulin-like growth factor 2, binding protein 1
  (IMP1)
• p62/IMP2
• IMP3/Koc
• CyclinA
• Cyclin B1
• Cyclin D1
• CDK2
• Survivin
• Optical density (OD) value at 490 nm were
  recorded
• The cut-off value for positive ELISA was defined
  as having gt2 times of mean OD value of
  non-smoking normal 3 standard deviation.
• Statistical analysis was performed using SPSS
  software

O.D.
Hypothesis

• We hypothesized that autoantibody reactivity
  level to a panel TAAs would be elevated in the
  serum of patients with lung cancer and
  preneoplasia (represented by persistent GGOs on
  CT) and that this reactivity to TAAs could be a
  biomarker for early lung cancer detection.

O.D.