Agenda Mammographic Density What are terminal duct lobular units - - PowerPoint PPT Presentation

▶
agenda
SMART_READER_LITE
LIVE PREVIEW

Agenda Mammographic Density What are terminal duct lobular units - - PowerPoint PPT Presentation

Nov 24, 2023 381 likes •500 views

6/9/2017 Disclosures Involution and Breast Density I do not have conflicts of interest to disclose. Mark Sherman (Presenter) & Gretchen Gierach I will not discuss off-label use of Health Sciences Research, Mayo Clinic Division of

slide-1
SLIDE 1

6/9/2017 1 Involution and Breast Density

Mark Sherman (Presenter) & Gretchen Gierach Health Sciences Research, Mayo Clinic Division of Cancer Epidemiology and Genetics National Cancer Institute

Disclosures

  • I do not have conflicts of interest to

disclose.

  • I will not discuss off-label use of

medications.

Agenda

  • What are terminal duct lobular units

(TDLUs)?

  • Why study TDLUs?
  • How are TDLU content and

mammographic density related?

  • Future aspirations and directions

Molecular Histology of the Breast

Mammographic Density

Increasing Risk

Terminal Duct Lobular Unit (TDLU)

Increasing Risk

Radisky BCRT 2016; Figueroa JNCI 2014

slide-2
SLIDE 2

6/9/2017 2

TDLU Morphology / Microenvironment

TDLU Morphology / Microenvironment

TDLU Morphometry

  • Number TDLU / unit area (“Density”)
  • TDLU span
  • Acini / TDLU

Lobule Type: 1-4

  • Types 1-3 reflect acini content

Reproducibility of TDLU Metrics

Comparison Correlation (Spearman) # TDLUs Acini / TDLU 0.36 # TDLUs Span/ TDLU 0.25 Acini / TDLU Span/ TDLU 0.77

Results similar for mean, median or maximum values

Correlation of TDLU Metrics Is TDLU Status an Intermediate Endpoint?

Carcinogenesis???

Involution Invasion

Breast Cancer Types

  • Luminal (A or B)
  • Basal
  • HER2
  • Claudin Low
  • Physiology – Lactation
  • Other Changes - BBD
  • Carcinogenesis
  • Process with

uncertain outcomes

slide-3
SLIDE 3

6/9/2017 3

Colony Forming Cells

Most proliferating cells are ER-, suggesting paracrine signaling

Kannan et al Stem Cell Reports 2013

Agenda

  • What are terminal duct lobular units

(TDLUs)?

  • Why study TDLUs?
  • How are TDLU content and

mammographic density related?

  • Future aspirations and directions

Why Study TDLUs?

  • Predict breast cancer risk (benign biopsy)
  • Predict local recurrence (e.g. DCIS)
  • Prevent cancer by inducing involution
  • Understand link: breast density and cancer risk
  • Study mechanisms of carcinogenesis
  • Intermediate endpoint in prevention trials

Age Related TDLU Involution

  • Hypothesized that TDLU involution is related to breast

density and cancer risk (Henson & Tarone Cancer 1993)

  • Lack of involution is a risk factor for breast cancer
  • Independent risk factor after benign biopsy
  • Mayo Benign Breast Disease (BBD) Cohort
  • Nurses’ Cohort
  • Association is robust; shown with different methods

Biopsy BBD TDLU Involution Cancer Vs. No Cancer Milanese JNCI 2006, McKian JCO 2009, Ghosh JCO 2010, Baer Cancer 2008

slide-4
SLIDE 4

6/9/2017 4 TDLU Involution Modifies Risks Associated with Other Factors

Milanese JNCI 2006

Involution TDLUs N= None ↑ P=Partial ↔ C=Complete ↓

TDLU Involution in Repeat Biopsies

20 40 60 80 100 0-25% 26-50% 51-75% >75% >75% 51-75% 26-50% 0-25% Biopsy 2 Non-progressive TDLU involution is related to increased breast cancer risk HRAdj = 1.63 (95% CI: 1.03-2.57)

Radisky BCRT 2016

Biopsy 1

Breast Density Modifies Risk in BBD

0.5 1 1.5 2 2.5 3 3.5 NP PDWA AH Low Medium High Standard Incidence Ratio Density

∗

Assessed via parenchymal pattern or BIRADS; independently confirmed with automated quantitative measurement

Vierkant BMC Cancer 2017

Involution, Mammographic Density and Breast Cancer Risk

Involution Density HRAdj (95% CI) Complete Low 1.00 Complete High 1.66 (0.75-3.70) Partial Low 1.57 (0.73-3.36) Partial High 2.70 (1.32-5.53) None Low 3.24 (1.05-9.98) None High 4.08 (1.72-9.68)

TDLU involution and mammographic density are independent breast cancer risk factors (pInt =0.60)

Ghosh JNCI 2010

slide-5
SLIDE 5

6/9/2017 5

Involution Cancers # (Column %) + LN # ( Row %) SIR (95% CI) None 242 (21.2) 87 (36) 2.05 (1.80-2.33) Partial 690 (60.5) 193 (28) 1.70 (1.57-1.83) Complete 208 (18.2) 41 (20) 1.09 (0.95-1.25) Total 1140 321

  • 321 (28%) of 1140 were LN+
  • Percentage with LN+ by involution : None>Partial>Complete
  • Number with LN+ by TDLU involution: Partial>None>Complete;

partial involution contributes 60% of LN+

  • Data are independent of age

TDLU Involution in Benign Biopsies Predicts LN+

Visscher Cancer 2016

TDLU Involution Metrics Vs. Age

Figueroa J D et al. JNCI J Natl Cancer Inst 2014;106:dju286

Published by Oxford University Press 2014.

TDLU Counts TDLU Span Acini / TDLU (Categorical)

Correlations (Spearman) Counts vs. Span r=0.16 Counts vs. Acini r=0.18 Span vs. Acini r=0.71 TDLU counts per section

  • vs. TDLU #/mm2 highly

correlated – standardized samples

TDLU Counts by Age And Parity Status

Figueroa J Natl Cancer Inst 2014

Nulliparous women had fewer TDLUs than uniparous women: Premenopausal RR=0.79 (0.73-0.85) Postmenopausal RR=0.67 (0.56-0.79) Increasing births related to higher TDLU counts: Premenopausal ptrend = 0.01; Postmenopausal ptrend = 0.007

Pinteraction = <=0.001 Yrs Since Birth RR (85% CI) <5 1.00 5-9 0.59 (0.29-1.21) 10+ 0.43 (0.20-0.93) Ptrend = 0.03 TDLU Counts: Uniparous Women

Age at last birth is not recorded in studies

Breast Density Vs. Age

Ginsburg BJC 2008 (Also summarizing data from Maskarinec CEBP 2006, Vachon CEBP 2007)

slide-6
SLIDE 6

6/9/2017 6

Tissue Composition Vs. Age & Density

Ginsburg BJC 2008, Li CEBP 2005, Caman BMC Public Health 2013, Sun CEBP 2014

TDLU Involution Vs. Mammographic Density

Ghosh JCO 2010

High Density Less Involution Low Density High Involution

Average TDLU Count by Age

More at-risk epithelium may partially explain risk associated with higher BD

TDLU Count / 100 mm2 % Dense Area Age Gierach GLCancer Prev Res 2016 % Dense Area TDLU Count

Average TDLU Count and Average Percent Breast Density by Age

Window of Susceptibility Radiation, smoking, age of immigration low to high risk area

Serum Hormones Vs. TDLU Counts

Serum Levels Premenopausal OR (95% CI) Postmenopausal OR (95CI) Estradiol 0.88 (.80-0.97)

1.61 (1.32-1.97)

Testosterone 0.89 (0.88-1.08) 1.32 (1.09-1.59) Progesterone

0.80 (0.72-0.89)

Prolactin

1.18 (1.07-1.31)

1.29 (1.04-1.59) SHBG 1.04 (0.93-1.16) 1.14 (0.89-1.45)

Adjusted for covariates. Results significant after correction for multiple comparisons bolded

Khodr CEBP 2014 (Komen Tissue bank)

slide-7
SLIDE 7

6/9/2017 7

Breast Radiology Evaluation And Study

  • f Tissues (BREAST) Stamp Project

TDLU Involution vs. Mammographic Density

  • 465 women ages 40-65 years undergoing

biopsy for radiologic finding at the University of Vermont (2007-10)

  • Questionnaires and blood collected
  • Area and volumetric density measured globally

and peri-lesionally (around biopsy target)

  • 348 women with benign biopsies evaluable

Shepherd CEBP 2011, Gierach Cancer Prev Res 2016

Premenopausal Women

5 10 15 20 25 30 35 40 45 50 % Dense-A % Dense-V %Dense-A %Dense-V

TDLU Counts TDLU Span % Mammographic Density

Ptrend 0.09 0.11 0.01 0.003

Gierach Cancer Prev Res 2016

Bars = Quintiles

Premenopausal Women

50 100 150 200 250 300 350 400 450 Non-Dense-A Non-Dense-V Non-Dense-A Non-Dense-V

TDLU Counts TDLU Span Non-Dense Area/Volume (cm2 / cm3 )

Ptrend 0.004 <0.02 0.01 0.009

Gierach Cancer Prev Res 2016

Breast Radiology Evaluation And Study of Tissues (BREAST) Stamp Project

Premenopausal women:

  • TDLU counts and span are directly related to

global and perilesional area/volume % density

– Associations for global absolute density weaker; associations found for absolute peri-lesional density

  • TDLU metrics are inversely related to global

absolute non-dense area and volume

– Associations for perilesional non-dense content weak

Postmenopausal women: Associations weak/null Acini/TDLU weak/null associations with density

slide-8
SLIDE 8

6/9/2017 8 Relationship of IGFs to Mammographic Density and TDLU Involution

  • In rodents, IGFs affect breast development, including

events at puberty, pregnancy, post-weaning involution

  • IGFs are involved in stromal-epithelial interactions
  • Mammographic density is a risk factor related to

fibroglandular content and tissue organization

  • IGFs have inconsistent relationship with density, but

more consistent relationship with breast cancer risk

  • IGF levels inversely related to level of TDLU involution;

increased IGF1-R associated with increased breast cancer risk (Rice Breast Cancer Res 2012)

Relationship of IGFs, Breast Density and TDLU Involution

  • 228 women (155 pre- and 73 post-

menopausal) with blood, TDLU assessment (corrected for area), not using hormones

  • Serum analysis: IGF-1; IGFBP-3 and ratio
  • Relate IGFs to TDLU involution, stratified by

menopausal status, adjusted for confounders

  • Mammographic density assessed as effect

modifier

0.5 1.5 2.0 2.5 3.0 1.0 0.5 1.5 2.0 2.5 3.0 1.0

Circulating Insulin-like Growth Factors (IGFs) and TDLU Count by Tertiles of Breast Density

IGF-1:IGFBP-3 Density Tertile 1 Density Tertile 2 Density Tertile 3 Premenopausal Postmenopausal RR (95% CI)

Horne H,…Gierach GL. Breast Cancer Res 2016.

P-int P-int

0.006 <0.001

Elevated IGF1 or IGF1:IGFBP3 ratio High TDLU Counts Higher Breast Cancer Risk High Breast Density??? Higher Breast Cancer Risk

High TDLU Counts CancerRisk High MD Cancer Risk High Breast Cancer Risk

Key Pathways IGFs? Hormones? Inflammation? Jak-Stat Signaling Is this local or Systemic?

TDLU Involution & Mammographic Density

slide-9
SLIDE 9

6/9/2017 9 TDLU Involution vs. Density in the Multiethnic Cohort

TDLU Involution Density% (Unadjusted) Density% (Adjusted) None 38.8 32.5 Partial 42.8 39 Complete 40.0 40.2

  • TDLU involution vs. density% associations null
  • TDLU involution was significantly associated with dense area
  • Complete: 48.0 cm2 ; None: 29.7 cm2
  • pUnadjusted = 0.03; pAdjusted = 0.007
  • TDLU involution not associated with non-dense area

Maskarinec BCR 2016

Density Vs. TDLU Involution

Features Ghosh et al n=2,667 Gierarch et al Premenopausal n=226 Postmenopausal n=122 Maskarinec et al N=173 Context BBD Cohort Benign Core Biopsy Cancer Cases Age (years) 51 Premenoapusal 46 Postmenopausal 57 59.7 Race >90% White >90% White 33% White; 46% Japanese BMI 26.4 Obesity (%) Pre-meno 22.1% Post-meno 32.0% 8.7% Mammographic Density Median ~23% Pre-meno 32.8% Post-meno 16.3% 41.2%

Ghosh JCO 2010, Gierach Cancer Prev Res 2016, Maskarinec BCR 2016

Agenda

  • What are terminal duct lobular units

(TDLUs)?

  • Why study TDLUs?
  • How are TDLU content and

mammographic density related?

  • Future aspirations and directions

Future Directions

  • Histopathology, image analysis, AI
  • Multiplex immunochemistry / RNA
  • Functional radiologic imaging methods
  • Experimental techniques (organoids)
  • Preclinical and clinical prevention studies
  • Longitudinal analyses
slide-10
SLIDE 10

6/9/2017 10

Mayo BBD Team: Lynne Hartmann, Amy Degnim, Dan Visscher, Derek Radisky, Celine Vachon, Stacey Winham, Marlene Frost, Rob Vierkant and others… National Cancer Institute: Gretchen Gierach, Ruth Pfeiffer, Britton Trabert, Hannah Yang, Bill Anderson, Goli Samimi, Jonine Figueroa and others… Karmanos Cancer Institute / Delphinus: Neb Duric, Mark Sak, and others … Komen Tissue Bank: Anna Maria Storniolo, Jill Henry and others …