Transcript Using the BCSC Research Infrastructure as a Junior Investigator

Using the BCSC Research Infrastructure
as a Junior Investigator
BCSC Meeting: Celebrating 15 Years of Accomplishment
Bethesda, MD April 27, 2010
E. Shelley Hwang MD, MPH
Chief, Division of Breast Surgery
UCSF Helen Diller Family Comprehensive Cancer Center
Personal Background
Training:
•Surgical oncology
•Joined UCSF faculty at UCSF after completing fellowship
•Little prior background in epidemiology/research methods
Research interest:
•DCIS and preinvasive breast cancer
•Obtained training grant 2004
•Interested in working with large datasets as part of master’s
thesis for MPH
BCSC and me
Advantages to a junior investigator
•Limited funding
•Large existing datasets
•Research mentorship
•Statistical support
•High likelihood of a completing a successful product
Junior researcher responsibilities:
•Define research question
•Identify time, resources
•Bring energy, enthusiasm, focus to project
BCSC and me
Introduction to BCSC
•Research infrastructure: UCSF Women’s Health Research
Center (PI: Grady)
•BCSC primary mentor (Kerlikowske):
•Identify opportunities; bring together junior researchers and
data
•Navigate process of data request, analysis, interpretation
•BCSC senior staff (Miglioretti, Ballard-Barbash):
•Guide the analysis and presentation of data
•Facilitate access to data dictionary, dataset
•Establish team of BCSC researchers with like interests
•Create opportunities to present research
BCSC AB71: Association Between Breast Density
and Recurrence Following Treatment for DCIS
 Breast density strongly correlated with breast cancer risk (RR 4-6)
 Heritable component (twin studies): 60%
 Responsive to changes in exogenous and endogenous hormones
 HRT
 Luteal phase of menstrual cycle
 Menopause
 Lifestyle/modifiable
component:
 Late age at first birth
 Nulliparity
 HRT
 Is increased breast density associated with a higher risk of
invasive recurrence in women following lumpectomy for DCIS?
Study Design
4431 women undergoing screening mammography at a BCSC site
diagnosed with DCIS, 1995-2005
179 women excluded for diagnosis of
ipsilateral invasive cancer within 60
days of DCIS
899 women excluded for mastectomy
Effect of radiation: Association of breast density
and risk of subsequent breast events*
Breast Densit y
No Radiat ion
Low
(BIRADS 1,2)
High
(BIRADS 2, 3)
95% CI
Radiat ion
Low
(BIRADS 1,2)
High
(BIRADS 2, 3)
95% CI
Any Inv asiv e
Canc er
(n=133)
Ipsilat eral
Cont ralat eral
Inv asiv e
Inv asiv e Canc er
Canc er (n=83)
(n=52)
Cont ralat eral
DCIS (n=30)
1.0
1.0
1.0
1.0
1.2
(0.7-2.0)
0.8
(0.4-1.6)
2.7
(1.0- 7.5)
1.6
(0.5-4.7)
1.0
1.0
1.0
1.0
1.7
(0.8-3.3)
1.0
(0.4-2.4)
3.6
(1.1- 11.3)
0.8
(0.1-4.4)
*all HR adjusted for age
Conclusions
 Women with higher breast density are not more likely to develop
invasive cancer in the ipsilateral breast following treatment for DCIS
 High breast density is associated with a 3-fold higher risk of
contralateral invasive cancer compared to women with low density
 Women undergoing treatment for DCIS with increased breast density
may benefit most from strategies aimed towards contralateral risk
reduction
I had such a great experience that I’m working
with the BCSC again!
 AB81: The Association of Breast Density and Contralateral
Breast Events in Women Undergoing Lumpectomy and
Radiation
 --Do women undergoing radiation as part of treatment for index
cancer have increased risk of CBC compared to women who did not
receive radiation?
 --Are young age or breast density associated with this risk?
AB81: The Association of Breast Density and Contralateral
Breast Events in Women Undergoing Lumpectomy and Radiation
Ipsilateral Events
Overall (n=1,885)
Breasy density
1: Almost entirely fat (<25%)
2: Scttrd fibroglandular tiss (25%-50%)
3: Heterogeneously dense (50%-75%)
4: Extremely dense (>75%)
Radiation
No
Yes
HR
95% CI
p-value
REF
0.94
1.00
1.11
0.36
0.74
0.79
0.84
1.19
1.26
1.45
REF
0.66
0.60
0.72
< 0.01
Contralateral Events
Overall (n=634)
Breasy density
1: Almost entirely fat (<25%)
2: Scttrd fibroglandular tiss (25%-50%)
3: Heterogeneously dense (50%-75%)
4: Extremely dense (>75%)
Radiation
No
Yes
HR
REF
1.54
2.08
1.80
REF
0.94
95% CI
p-value
< 0.01
0.87
1.18
0.96
2.71
3.66
3.38
0.492
0.79
*Adjusted for age, race, menopausal status, HT use
1.12
BCSC Publications
Recent High Impact BCSC publications by Junior
Investigators (success is not the exception!)

Are there racial/ethnic disparities among women younger than 40 undergoing
mammography? Kapp JM, Walker R, Haneuse S, Buist DS, Yankaskas BC. Breast Cancer
Res Treat. 2010 Mar 4. [Epub ahead of print]

Rates of atypical ductal hyperplasia have declined with less use of postmenopausal
hormone treatment: findings from the Breast Cancer Surveillance Consortium. Menes TS,
Kerlikowske K, Jaffer S, Seger D, Miglioretti DL. Cancer Epidemiol Biomarkers Prev. 2009
Nov;18(11):2822-8.

Using clinical factors and mammographic breast density to estimate breast cancer risk:
development and validation of a new predictive model. Tice JA, Cummings SR, SmithBindman R, Ichikawa L, Barlow WE, Kerlikowske K. Ann Intern Med. 2008 Mar
4;148(5):337-47.

Influence of computer-aided detection on performance of screening mammography. Fenton
JJ, Taplin SH, Carney PA, Abraham L, Sickles EA, D'Orsi C, Berns EA, Cutter G, Hendrick
RE, Barlow WE, Elmore JG. N Engl J Med. 2007 Apr 5;356(14):1399-409.
Working with the BCSC
 Outstanding resource for investigators, particularly junior researchers
 Access to BCSC mentoring, large datasets, analytical support
(Statistical Coordinating Center); ease of application process for data
requests
 Important to have clear research question, clear definitions and
parameters for variables (years of diagnosis, definition of “recurrence”)
 Potential resource to gather preliminary data for grant submissions
• Specific Data Requests
• Risk Estimation Data Set
• Cancer Incidence Data
Summary
 BCSC has made key contributions in breast screening and breast
cancer outcomes research; this is expected to continue well into the
future as new questions emerge
• cost-effectiveness
• resource allocation
• quality metrics
 Excellent ROI as many projects attain funding apart from BCSC;
BCSC essential to providing the resources and data to secure such
funding
 The resource and infrastructure have made important contributions to
academic training and career advancement; this resource is vital in
institutions with strong BCSC mentorship
 Need to continue outreach efforts to non-BCSC investigators
Acknowledgements
 Karla Kerlikowske
 Diana Miglioretti
 Rachel Ballard-Barbash
 Donald Weaver
 Ed Sickles
 Steve Taplin
 Staff and Researchers of the Statistical Coordinating Center
 Sebastian Haneuse
 Ina Gylys-Colwell
 Patients who continue to contribute their valuable time
and data to support the BCSC
Thank you!