Transcript Yang Ping, MD, PhD

Genetic and Environmental Determinants in
Lung Cancer Progression and Survivorship
Ping Yang, M.D., Ph.D.
Professor and Consultant
Department of Health Sciences Research
Department of Medicine
Department of Medical Genetics
Mayo Comprehensive Cancer Center
Mayo Clinic College of Medicine
Outline
•
•
Overview of lung cancer prognosis
•
Identify and validate new predictors for
lung cancer survival: ongoing efforts
•
Current research using pharmacogeneticepidemiologic tools: towards
individualized medicine
•
Characteristics of long-term survivors:
a multi-dimensional approach
Known determinants of lung cancer survival:
environment and genes
Acknowledgement: Survivorship Research Team
Medical Oncology
Alex A. Adjei
James R. Jett
Aminah Jatoi
Randolph S. Marks
Julian R. Molina
Pulmonary Medicine
Eric S. Edell
David E. Midthun
Radiation Oncology
Yolanda I. Garces
Bioinformatics
Zhifu Sun
George Vasmatzis
Thoracic Surgery
Mark S. Allen
Stephen D. Cassivi
Claude Deschamps
Francis C. Nichols
Peter C. Pairolero
Victor F. Trastek
Chest Pathology
Marie-Christine Aubry
Molecular Biology
Julie M. Cunningham
Wilma L. Lingle
Wanguo Liu
Stephen N. Thibodeau
Psychology
Matthew M. Clark
Biostatistics
Sumithra J. Mandrekar
V. Shane Pankratz
Jeff A. Sloan (QoL expert)
Pharmocogenomics
Richard M. Weinshilboum
Nicotine Dependence Chaplain
Jon O. Ebbert
Mary E. Johnson
Oncology Nursing
Linda Sarna (UCLA)
Epidemiology
Ping Yang
Overview: An Old Story with Continued Challenge
Carcinoma of the Lung and Bronchus
• High incidence rate:
12-13% cancer diagnosis in U.S.;
>60% diagnosed at a not-curable stage.
• High mortality rate:
5-year survival rate is ~15%.
• Kills more people than any other cancer:
~30% of all cancer deaths in U.S.
Known Predictors of Early-stage Lung Cancer Survival
Tumor-related factors:
Essential
Important
Lymph node involvement, hypercalcemia
Tumor size, pleural involvement, multifocal, cell type, grade, vessel invasion
Over 8 physio-pathological pathways and
more than 30 cellular & molecular markers
Promising
Environment-related factors:
Essential
Promising
Host-related factors:
Essential
Treatment modalities
Smoking history, diet / supplement
Weight Loss
Important
Age, gender
Promising
Marital status, race/ethnicity, mood,
quality of life, metabolizing enzyme genes
Yang et al., 2004, Modified from Brundage et al. 2002
Background: A Lung Cancer Research Infrastructure
Physical & Psychosocial
Status: e.g., symptoms,
comorbidity, & supports
Staging, PS, &
Treatment:
TNM, surgery,
chemotherapy,
& radiotherapy
Health Related
Behaviors: e.g., diet,
smoking, & exercise
Quantity
and
Quality
of Life
Tumor: e.g., histologic cell
type and differentiation grade,
biologic & mechanistic genes
Host Factors:
e.g., genetic
predisposition
and demographic factors
CHEST, 2006
A Prospectively Followed Patient Cohort:
Newly Diagnosed Lung Cancer, 1997-Ongoing
Identification,
Baseline data,
Blood/Tissue
~1000 patients
each year
6 months
follow-up
1 year
follow-up
Annually
after
Progression and Death
Svobodnik A, et al, 2004;
Yang P, et al. 2005.
Identifying and Validating New Prognostic Factors
1 of 4 groups
Tumor-related factors:
Essential
Important
Lymph node involvement, hypercalcemia
Tumor size, pleural involvement, multifocal, cell type, grade, vessel invasion
Over 8 physio-pathological pathways and
more than 30 cellular & molecular markers
Promising
Environment-related factors:
Essential
Promising
Host-related factors:
Essential
Treatment modalities
Smoking history, diet / supplement
Weight Loss
Important
Age, gender
Promising
Marital status, race/ethnicity, mood,
quality of life, metabolizing enzyme genes
Yang et al., 2004, Modified from Brundage et al. 2002
Example: treatment of recurrent lung cancer
and post-recurrence survival
(continued)
Est. Survival, %
Post-Recurrence Survival by Risk Score Group
100
90
80
70
60
50
40
30
20
10
0
RS4
RS: 6-8
RS: 4-6
RS>8
0
3
6
9
12
15
Months After Recurrence
18
21
24
ATS, 2006
1.6
1.4
(No treatment is the reference group, RR=1.0)
Relative Risk of Post-Recurrence Mortality
2.0
1.8
1.2
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
Surgery
Surgery + Chemo/Radiotherapy
Chemotherapy
Chemo + Radiotherapy
Radiotherapy
0.05
Risk Score < 4
Risk Score 4-6
Risk Score 6-8
Risk Score > 8
Treatment Modality by Risk Score
ATS, 2006
Identifying and Validating New Prognostic Factors
2 of 4 groups
Tumor-related factors:
Essential
Important
Lymph node involvement, hypercalcemia
Tumor size, pleural involvement, multifocal, cell type, grade, vessel invasion
Over 8 physio-pathological pathways and
more than 30 cellular & molecular markers
Promising
Environment-related factors:
Essential
Promising
Host-related factors:
Essential
Treatment modalities
Smoking history, diet/supplement
Weight Loss
Important
Age, gender
Promising
Marital status, race/ethnicity, mood,
quality of life, metabolizing enzyme genes
Yang et al., 2004, Modified from Brundage et al. 2002
Survival by Years Since Quit Smoking, Women
Adjusted for age, packs per day, years smoked,
histology, grade, stage, and treatment
100
90
80
70
60
50
40
30
20
10
0
0-10 yrs
11-20 yrs
21-30 yrs
> 30 yrs
0
1
2
3
4
5
Lung Cancer, 2005
Dietary Supplement of Vitamins and Minerals
• In general population, ~40% take vitamin/
mineral supplements regularly.
• Approximately 80% of cancer patients do so.
• Both clinical and laboratory data have
shown that certain micronutrients effect the
growth of malignant cells:
i.e., vitamins and minerals appear to be
modulators of tumor growth.
• Are these supplements helping or
hurting lung cancer patients?
Dietary Supplement of Vitamins and Minerals:
NSCLC
% SURVIVING
Multivariable Model-Based Survival Curves
100
90
80
70
60
50
40
30
20
10
0
P < 0.01
Vitamin/Mineral Users
Non-Users
0
1
2
3
4
5
Years After Diagnosis
Lung Cancer, 2005
Identifying and Validating New Prognostic Factors
3 of 4 groups
Tumor-related factors:
Essential
Important
Promising
Lymph node involvement, hypercalcemia
Tumor size, pleural involvement, multi-focal,
cell type, grade, vessel invasion
Over 8 physio-pathological pathways and
more than 30 cellular & molecular markers
Environment-related factors:
Essential
Treatment modalities
Promising
Host-related factors:
Essential
Smoking history, diet/supplement
Weight Loss
Important
Age, gender
Promising
Marital status, race/ethnicity, mood, quality
of life, drug metabolizing enzyme genes
Yang et al., 2004, Modified from Brundage et al. 2002
Chemotherapy & Treatment Outcome
• For stage III (and IV) NSCLC and limited stage
SCLC, combined modality of concurrent chemoand radiotherapy is considered as the standard
of care.
• The goal of such treatment is to improve locoregional tumor control and minimize metastases
without increasing morbidity.
• Overall, there is a significant benefit in
survival, but only in a subset of 25-30%
among all treated. Who and why?
Chemotherapy Agents (in %) Used at Mayo Clinic
During the Past Eight Years (1997-2004)
All Chemotherapy
First-Line
Chemotherapy
Subsequent
Chemotherapy
Drug Groups
Stage III/IV
NSCLC
Total Count (denominator)
1093
Platinum-containing Agents (P) 90.1
Taxane-containing agents (T) 76.2
Gemcitabine (G)
32.0
EGFR inhibitor (E)
8.0
Either P or T
91.7
Both P and T
74.7
Either P or G
94.0
Both P and G
28.2
Either P or E
92.2
Both P and E
5.9
Either T or G
85.3
Both T and G
23.0
Either T or E
79.2
Both T and E
4.9
Either G or E
35.9
Both G and I
4.1
None of the above
3.1
SCLC
247
94.7
30.8
4.9
0
97.2
28.3
94.7
4.9
94.7
0
31.2
4.5
30.8
0
4.9
0
2.8
Stage III&IV
NSCLC
1093
85.7
66.1
13.0
2.7
88.2
63.7
91.1
7.6
88.2
0.3
78.0
1.1
68.6
0.3
15.6
0.1
4.6
SCLC
247
91.5
10.5
0
0
96.4
5.7
91.5
0
91.5
0
10.5
0
10.5
0
0
0
3.6
Stage III&IV
NSCLC
SCLC
463
51.8
45.8
47.5
12.5
64.4
33.3
76.9
22.5
59.6
4.8
77.8
15.6
54.0
4.3
54.0
6.0
9.7
107
61.7
52.3
11.2
0
84.1
29.9
68.2
4.7
61.7
0
56.1
7.5
52.3
0
11.2
0
14.0
A BRIEF BACKGROUND
• Platinum-based drugs are commonly
used in lung cancer chemotherapy.
• The glutathione metabolic pathway is
directly involved in the inactivation of
platinum compounds.
The Glutathione Pathway and Its Role
in Drug Detoxification – Yang et al., 2006; JCO
Glutathione
GCLC Gene, Platinum-based Drugs, & Lung Cancer Survival
100
Est. Survival, %
80
60
40
20
0
0
1
2
3
4
5
Years After Diagnosis
Plat GCLC-00
Stage III-IV GCLC-00
Plat GCLC-77
Stage III-IV GCLC-77
Yang et al., 2005
Clinical Implications
• Genotypes of glutathione-related enzymes may be
used as host factors in predicting patients’ survival
after treatment with platinum-based drugs.
• The distribution of GCLC repeats marker:
GCLC-77:
19% - not use platinum drugs?
GCLC-7_:
50% - balancing benefit vs. harm?
GCLC-other: 31% - suitable for platinum-drugs?
Yang et al., 2005
Many Shortcomings
Much needed to be done…
Other pathways
Paradoxical “toxicities”
Accurate follow-up data
…
Identifying and Validating New Prognostic Factors
-4Tumor-related factors:
Essential
Important
Lymph node involvement, hypercalcemia
Tumor size, pleural involvement, multifocal, cell type, grade, vessel invasion
Over 8 physio-pathological pathways and
more than 30 cellular & molecular markers
Promising
Environment-related factors:
Essential
Promising
Host-related factors:
Essential
Treatment modalities
Smoking history, diet/supplement
Weight Loss
Important
Age, gender
Promising
Marital status, race/ethnicity, mood,
quality of life, metabolizing enzyme genes
Yang et al., 2004, Modified from Brundage et al. 2002
JTCVS., 2006
Biological Markers:
Promises and Challenges
• Treatment response is generally poor.
• Limited markers to predict prognosis and
apply to individualized management.
• Gene expression profiling, “microarray”,
has been widely used to search for
answers at molecular level for differed
lung cancer survival
• (Note: DNA microarray measures tens of
thousands expressed genes via mRNA
simultaneously in tissue or cells)
Emerging evidence shows that the accuracy of
expression-based outcome prediction varies
greatly among studies.
Converging questions have been raised from
researchers and clinicians:
• Why does gene-based prediction vary?
• Can DNA expression profiles provide more
•
accurate prediction than conventional predictors?
Are gene panels or molecular signatures
independent predictors or merely surrogates of
conventional factors?
Three Pioneer Studies:
Larger Samples in “Top-Tier” Journals
• Stanford group (PNAS 2001;98(24):13784-9):
56 cases of lung cancer
- 41 AD, 16 SCC, 5 LCLC, 5 SCLC
• Harvard group (PNAS 2001;98(24):13790-5):
186 cases of lung cancer
- 127 AD, 21 SCC, 20 carcinoid, 6 SCLC
• Michigan group (Nat Med 2002;8:816-24):
- 86 cases of lung adenocarcinoma
Survival Prediction on Harvard Data From
50 Genes Selected From Michigan Data
Survival Curves Predicted by Different Gene
Markers on an Independent Sample
A
B
Survival Plot
1.0
H
L
0.9
1.0
H
L
0.9
0.8
0.7
0.7
0.7
0.5
0.4
Surviving
0.8
0.6
0.6
0.5
0.4
0.6
0.5
0.4
0.3
0.3
0.3
0.2
0.2
0.2
0.1
0.1
0.1
0.0
0.0
0.0
0
10
20
30
40
50
60
Survival
70
80
90 100 110
H
L
0.9
0.8
Surviving
Surviving
Survival Plot
Survival Plot
1.0
C
0
10 20 30 40 50 60 70 80 90 100 110
Survival
Top 50 genes selected
Top 50 genes from
from univariate analysis multivariate adjustment
and cross validation
(age, gender, stage, cell
type), original data
0
10 20 30 40 50 60 70 80 90 100 110
Survival
Top 50 genes from
multivariate adjustment
(age, gender, stage, cell
type), Dchip data
Comparison of survival predictions by a 50-gene signature and
combination of clinical and pathologic variables
Sun &Yang, 2006;15:2063-8
A
B
Survival Plot
0.8
0.8
H 1.0
L 0.9
0.8
0.7
0.7
0.7
1.0
H
L
0.6
0.5
0.4
Surviving
0.9
Surviving
0.9
Surviving
Survival Plot
Survival Plot
1.0
C
0.6
0.5
0.4
0.6
0.5
0.4
0.3
0.3
0.3
0.2
0.2
0.2
0.1
0.1
0.1
0.0
0.0
0.0
0
10
20 30 40 50
60 70 80
90 100 110
Survival
Top 50 genes selected from
univariate analysis and
cross validation
H
L
0
10 20 30 40 50 60 70 80 90 100 110
Survival
Top 50 genes from
multivariate adjustment
(age, gender, stage, cell
type), original data
0
10 20 30 40 50 60 70 80 90 100 110
Survival
Top 50 genes from multivariate
adjustment (age, gender,
stage, cell type), Dchip data
Common Genes
1
1
1
7
7
--
--
3
3
3
--
3
Outline
•
•
Overview of lung cancer prognosis
•
Identify and validate new predictors for
lung cancer survival: ongoing efforts
•
Current research using pharmacogeneticepidemiologic tools:
towards individualized medicine
•
Characteristics of long-term survivors:
a multi-dimensional approach
Known determinants of lung cancer survival:
genes and environment
A Brief Background
• Individuals who are alive over 5 years after
a lung cancer diagnosis are referred to as
long-term lung cancer (LTLC) survivors.
• In the U.S., approximately 26,000
individuals become LTLC survivors annually.
• A paucity of information regarding the
quality of life (QoL) among LTLC survivors.
Longitudinal Evaluation of Quality of Life
in Long-Term Lung Cancer Survivors
A Short story
Overall QoL change between two time periods:
under 3 years and over 5 years post diagnosis
Multi-dimension Follow-up Measures
Besides medical records, multiple tools:
• SF-8 Health Survey
• ECOG* Performance Status Score
(*Eastern Cooperative Oncology Group)
• Lung Cancer Symptom Scale (LCSS)
• Linear Analogue Self-assessment (LASA)
(modified for lung cancer patients)
• Baecke Questionnaire for Habitual Activities
• FACT-SP Spiritual Well Being Assessment
• Other tools (diet, sleep, cognitive function, etc)
QoL Assessment
• Overall QoL was assessed using LCSS-9:
- scores 0 (worst) to 100 points (best)
- as continuous variable: distance in cm on a VAS
a raw score of the total 100 points
- as a binary variable
a poor QoL defined as <50 points (Sloan, 2004)
• Declining QoL was defined as:
a 10-point or more decrease between
the two time periods
A Prospective Lung Cancer Cohort:
Long-term Survivors, 2002-2004
N = 2837
N = 448, 15.8%
Patients
diagnosed
1997-1999
5-year
follow-up
Annually
after
Declining Overall QoL Over Time:
Higher Proportion with Poor
Overall QoL
QoL Declined
34%
No Change
48%
18%
QoL Improved
Yang et al., 2005
Factors Influencing Overall QoL in
Long-term Lung Cancer Survivors
Characteristics
Poor QoL at
<3 year
>5 year

Age > 75 years
Education < 16 years



TNM staging- Stage I
Histology- Poorly/un-differentiated
Lung cancer treatment
Chemotherapy – Yes
Radiation therapy – Yes
Comorbid conditions
COPD
Heart failure
Recurrent/subsequent
lung cancer





Implications
• Our preliminary results show: among the
LTLC survivors, the mean overall QoL declined
significantly between the two time periods.
This is in a sharp contrast to long-term
survivors of other cancers, e.g., breast
cancer, whose overall QoL are compatible
to their age-matched controls.
• We found substantial differences in factors
contributing to their poor QoL at each time
period.
Future Directions
•
Long-term lung cancer survivors may need
additional help to improve their QoL.
•
Further research efforts are needed. The next
step is to identify factors that are associated
with a declined vs. an improved QoL over time:
environmental, genetic, biological, behavioral,
psychosocial.
•
Ultimately, we aim to define modifiable factors
and improve QoL of “at risk” survivors.
Acknowledgement: Survivorship Research Team
Alex A. Adjei
William R. Bamlet
Jean M. Chovan
Julie M Cunningham
Chiaki Endo
Yolanda I. Garces
Aminah Jatoi
Mary E. Johnson
Wilma L. Lingle
Randolph S. Marks
David E. Midthun
Paul J. Novotny
Peter C. Pairolero
Shawn M. Stoddard
William R. Taylor
Jason A. Wampfler
Diane K. Wilke
George Vasmatzis
Mark S. Allen
Marie-Christine Aubry
Aaron O. Bungum
Stephen D. Cassivi
Matthew M. Clark
Claude Deschamps
Jon O. Ebbert
Eric S. Edell
Susan M. Ernst
Erin E. Finke
Debra L. Hare
Shauna L. Hillman
James R. Jett
Ruoxiang Jiang
Thomas D. Knowlton
Farhad Kosari
Wanguo Liu
Sumithra J. Mandrekar
Sheila R. McNallan
Rebecca L. Meyer
Julian R. Molina
Francis C. Nichols
Janice R. OffordScott H. Okuno
V Shane Pankratz
Jeff A. Sloan
Hiroshi Sugimura
Zhifu Sun
Stephen N. Thibodeau Victor F. Trastek
Richard M. Weihshilboum
Brent A. Williams
Joel B. Worra
Anthony L. Visbal
Xinghua Zhao
ALL STUDY PARTICIPANTS AND SUPPORTERS
THANK YOU!