ArticlePDF Available

The Contributions of Breast Density and Common Genetic Variation to Breast Cancer Risk



We evaluated whether a 76-locus polygenic risk score (PRS) and Breast Imaging Reporting and Data System (BI-RADS) breast density were independent risk factors within three studies (1643 case patients, 2397 control patients) using logistic regression models. We incorporated the PRS odds ratio (OR) into the Breast Cancer Surveillance Consortium (BCSC) risk-prediction model while accounting for its attributable risk and compared five-year absolute risk predictions between models using area under the curve (AUC) statistics. All statistical tests were two-sided. BI-RADS density and PRS were independent risk factors across all three studies (P interaction = .23). Relative to those with scattered fibroglandular densities and average PRS (2(nd) quartile), women with extreme density and highest quartile PRS had 2.7-fold (95% confidence interval [CI] = 1.74 to 4.12) increased risk, while those with low density and PRS had reduced risk (OR = 0.30, 95% CI = 0.18 to 0.51). PRS added independent information (P < .001) to the BCSC model and improved discriminatory accuracy from AUC = 0.66 to AUC = 0.69. Although the BCSC-PRS model was well calibrated in case-control data, independent cohort data are needed to test calibration in the general population. © The Author 2015. Published by Oxford University Press. All rights reserved. For Permissions, please e-mail:
Received: May 17, 2014; Revised: July 18, 2014; Accepted: October 27, 2014
© The Author 2015. Published by Oxford University Press. All rights reserved. For Permissions, please e-mail:
JNCI J Natl Cancer Inst (2015) 107(5): dju397
First published online March 5, 2015
Brief Communication
1 of 4
 
The Contributions of Breast Density and Common
Genetic Variation to Breast Cancer Risk
Celine M.Vachon, V. ShanePankratz, Christopher G.Scott, LotharHaeberle,
EladZiv, Matthew R.Jensen, Kathleen R.Brandt, Dana H.Whaley, Janet
E.Olson, KatharinaHeusinger, Carolin C.Hack, Sebastian M.Jud, Matthias
W.Beckmann, RuedigerSchulz-Wendtland, Jeffrey A.Tice, Aaron D.Norman,
Julie M.Cunningham, Kristen S.Purrington, Douglas F.Easton,
Thomas A.Sellers, KarlaKerlikowske, Peter A.Fasching*, Fergus J.Couch*
Afliations of authors: Department of Health Sciences Research, Division of Epidemiology, Mayo Clinic (CMV, VSP, CGS, MRJ, JEO, ADN, FJC); Department of Gynecology
and Obstetrics, University Hospital Erlangen Friedrich-Alexander-University Erlangen-Nuremberg, Comprehensive Cancer Center Erlangen-EMN, Erlangen, Germany
(LH, KH, CCH, SMJ, MWB, PAF); Department of Medicine, Institute for Human Genetics, Helen Diller Family Comprehensive Cancer Center, University of California, San
Francisco, CA (EZ); Departments of Medicine and Epidemiology and Biostatistics and General Internal Medicine Section, Department of Veterans Affairs and Division
of General Internal Medicine (EZ, JAT, KK); Division of Breast Imaging, Department of Radiology, Mayo Clinic College of Medicine, Rochester, MN (KRB, DHW); Institute
of Diagnostic Radiology, University Hospital Erlangen, Friedrich-Alexander University Erlangen-Nuremberg, Erlangen, Germany (RS-W); Division of Experimental
Pathology, Department of Laboratory Medicine and Pathology, Mayo Clinic College of Medicine, Rochester, MN (JMC, FJC); Wayne State University School of Medicine
and Karmanos Cancer Institute, Detroit, MI (KSP); University of Cambridge, Centre for Cancer Genetic Epidemiology, Cambridge, UK (DFE); Moftt Cancer Center,
Tampa, Florida (TAS); University of California at Los Angeles, Department of Medicine, Division Hematology/Oncology, David Geffen School of Medicine, Los Angeles,
*Authors contributed equally to this work.
Correspondence to: Celine M.Vachon, PhD, Mayo Clinic, 200 First Street SW, Charlton Building 6–239, Rochester, MN 55905 (e-mail:
We evaluated whether a 76-locus polygenic risk score (PRS) and Breast Imaging Reporting and Data System (BI-RADS)
breast density were independent risk factors within three studies (1643 case patients, 2397 control patients) using logistic
regression models. We incorporated the PRS odds ratio (OR) into the Breast Cancer Surveillance Consortium (BCSC) risk-
prediction model while accounting for its attributable risk and compared ve-year absolute risk predictions between
models using area under the curve (AUC) statistics. All statistical tests were two-sided. BI-RADS density and PRS were
independent risk factors across all three studies (Pinteraction=.23). Relative to those with scattered broglandular densities
and average PRS (2nd quartile), women with extreme density and highest quartile PRS had 2.7-fold (95% condence interval
[CI]=1.74 to 4.12) increased risk, while those with low density and PRS had reduced risk (OR=0.30, 95% CI=0.18 to 0.51).
PRS added independent information (P < .001) to the BCSC model and improved discriminatory accuracy from AUC=0.66 to
AUC=0.69. Although the BCSC-PRS model was well calibrated in case-control data, independent cohort data are needed to
test calibration in the general population.
Mammographic breast density is associated with decreased
diagnostic accuracy of mammography (1,2) and increased
breast cancer risk (3,4). Recent legislation, passed in nineteen
states, mandates radiologists to communicate the importance
of breast density information to patients undergoing mam-
mography. Because 45% to 50% of women have heterogeneously
by guest on November 2, 2015 from
Vachon et al. | 2 of 4
or extremely dense breasts (5), this legislation will result in
increased patient-provider discussions regarding breast screen-
ing frequency, supplemental screening strategies, and risk (6).
Additional factors for further risk stratication, especially for
women with dense breasts, are needed to inform these discus-
sions (7).
To date, almost 80 conrmed breast cancer susceptibility loci
have been identied (8–23) and explain approximately 14% of
familial breast cancer risk (8). Coupled with established risk fac-
tors like breast density, these loci are likely to increase the utility
and accuracy of clinical risk prediction.
We conducted two studies to evaluate the contribution
of established breast cancer susceptibility loci to the Breast
Imaging Reporting and Data System (BI-RADS) density and
breast cancer association. First, we determined whether a
polygenic risk score (PRS) composed of 76 single-nucleotide
polymorphisms (SNPs) is a statistically signicant risk fac-
tor independent of BI-RADS density in three epidemiologic
studies. Second, we examined whether the addition of this
PRS improves performance of the Breast Cancer Surveillance
Consortium (BCSC) ve-year risk-prediction model in a nested
case-control study (24).
Studies included 456 case patients and 1166 age-matched
control patients nested within the Mayo Mammography Health
Study (MMHS) cohort (25,26) and two clinic-based case-control
studies with 675 case patients and 864 frequency age-matched
control patients (Mayo Clinic Breast Cancer Study [MCBCS]),
and 512 case patients and 367 unmatched control patients
(Bavarian Breast Cancer Cases and Control Study [BBCC]), for
a total 1643 case patients and 2397 control patients (25,27,29)
(Supplementary Table 1 and Methods, available online). All
studies obtained informed consent, ethics, and institutional
BI-RADS breast density was categorized as dened in the
BI-RADS lexicon (30) into one of four categories: 1) almost
entirely fat, 2)scattered broglandular densities, 3)heterogene-
ously dense, 4)extremely dense, by expert radiologists on mam-
mograms close to (MCBCS and BBCC) or years prior (MMHS) to
diagnosis. Genotyping of the 76 SNPs (8–23) was conducted on
a custom Illumina iSelect genotyping array (8) (Supplementary
Methods, available online).
The PRS was formed using published per-minor-allele odds
ratios (ORs) corresponding to the SNP associations with overall
breast cancer (Supplementary Table2, available online) (8–23).
The log OR for each SNP was multiplied by the corresponding
number of minor alleles, summed to generate a unique PRS for
each person (31), and evaluated both as continuous (per stand-
ard deviation) and quartile measures. Logistic regression was
used to examine the association of BI-RADS density, PRS, and
their interaction with breast cancer risk, adjusting for age, 1/
BMI (4,26), and study. ORs, area under the curve (AUC) (32,33),
and 95% condence intervals (CIs) were estimated. Alikelihood
ratio test (LRT) assessed statistical signicance of associations
between PRS and breast cancer while accounting for BI-RADS,
age, and1/BMI.
We formed the BCSC-PRS risk model by estimating the OR
corresponding to a one-unit increase in the PRS with data from
the BBCC and MCBCS studies only, and added this estimate
directly into the original BCSC model (Supplementary Methods,
available online) (24). We then estimated ve-year risk of inva-
sive cancer for the BCSC and BCSC-PRS models within the
MMHS cohort (334 invasive case patients) using a resampling
approach to obtain fty sets of 334 genotyped control patients
whose age distribution matched the full cohort. We compared
the performance of the BCSC-PRS vs BCSC model using: 1) a
LRT, 2)AUCs and corresponding 95% CIs, 3)change in AUC and
95% CIs based on standard errors estimated from 200 bootstrap
samples. We assessed the calibration of BCSC and BCSC-PRS
ve-year risk by comparing observed and predicted numbers
of cancers, adjusting for the case-control design (34), using the
Hosmer-Lemeshow test (35) as done in other risk models using
case-control data (31,36). All analyses were performed within
the 50 resampled data sets, and results were combined across
them using approaches developed for multiple imputation (37)
(Supplementary Methods, available online). We evaluated the
net reclassication of case patients (change in true-positive
rate) and control patients (change in false-positive rate) for a
ve-year risk of 3% or greater (where there is greater absolute
benet of chemoprevention [38]) for the BCSC-PRS compared
with the BCSC model. Estimates and corresponding 95% CIs
were obtained using bootstrapping. All statistical tests were
BI-RADS density, adjusted for age and 1/BMI, was a statisti-
cally signicant risk factor for breast cancer within and across
studies (Table1; Supplementary Tables 3–5). The PRS was not
strongly correlated with age, BMI, or density (all statistical
correlations <.05). PRS was associated with breast cancer risk
within and across the three studies and was a statistically sig-
nicant risk factor independent of BI-RADS density (Pinteraction
=.23) (Tables 1 and 2; Supplementary Tables 3–5, available
online). PRS improved the t of models with BI-RADS density,
age, and 1/BMI (PLRT < .001), resulting in an AUC of 0.69 (95%
CI = 0.67 to 0.71) (Table 1, Supplementary Figure1, available
Importantly, the PRS further stratied risk associated with
extremely dense breasts, such that those with the lowest PRS
had an odds ratio of 0.91 (95% CI=0.53 to 1.56), while those in
the highest PRS had a 2.7-fold (95% CI=1.74 to 4.12) increased
risk compared with women with scattered broglandular
densities and average PRS (2nd quartile) (Table 2). Women
with fatty breasts and in the lowest PRS quartile had the low-
est risk (OR = 0.30, 95% CI = 0.18 to 0.51). Associations were
similar across studies (Pinteraction=.24) and within menopausal
subgroups (Pinteraction = .86) (Supplementary Table6, available
We next incorporated the OR corresponding to a one-unit
increase in the PRS estimated from the MCBCS and BBCC stud-
ies only (OR= 1.83, 95% CI = 1.59 to 2.11) into the BCSC model
and evaluated the impact within the MMHS (Supplementary
Table7, available online). The BCSC and BCSC-PRS models were
well calibrated when t on the case-control sets sampled from
MMHS (Supplementary Figure2, available online). Addition of
the PRS to the BCSC model provided a statistically signicant
improvement to the model t (PLRT < .001) and improved discrim-
ination between case patients and control patients (AUC=0.69,
95% CI = 0.64 to 0.73) compared with the BCSC model alone
(AUC=0.66, 95% CI = 0.61 to 0.70) (∆AUC=0.028, 95% CI = 0.007 to
0.049) (Supplementary Figure3, available online). The BCSC-PRS
model resulted in 36.8% of case patients exceeding the ve-year
risk threshold of 3% or greater where chemotherapy should be
discussed, compared with 25.7% of cases using the BCSC model,
for a net reclassication of 11% (95% CI= 7% to 15%) of cases.
In control patients, these numbers were 13.0% and 10.7% for
BCSC-PRS and BCSC, respectively, for a net 2% (95% CI=-1% to
5%) reclassication of control patients to a risk of 3% or greater,
although not statistically signicant (Supplementary Table 8,
available online). The increase in number of cases with the
BCSC-PRS at or above the 3% threshold may represent improved
by guest on November 2, 2015 from
3 of 4 | JNCI J Natl Cancer Inst, 2015, Vol. 107, No. 2
discrimination, but might also reect poor model calibration in
the upper tails, to some degree, especially as control patients
were also upwardly reclassied.
Our ndings that the PRS and BI-RADS density were inde-
pendent risk factors and that incorporating the PRS into the BCSC
risk model improved model t and net reclassication for case
patients suggest that both breast density and common genetic
variation are important for risk prediction. Risk models with good
discrimination and accuracy for predicting breast cancer are
important for targeted screening and prevention (39). Tamoxifen,
raloxifene, and aromatase inhibitors have been shown to be
effective for primary prevention but are rarely used in practice
(40) because of side effects and low interest of moderate-risk
women in taking prevention medication for breast cancer (41,42).
The highest-risk women may be more motivated to take preven-
tive therapies and accept their potential complications (38). Our
results demonstrate that the set of 76 SNPs improves the iden-
tication of women at the highest risk. Along with the increase
seen in AUC, the net-reclassication of 11% of case patients (95%
CI=7% to 15%) to a risk level where women are more likely to
benet from chemoprevention suggests that SNPs could be use-
ful clinically.
The main limitation of this study was lack of independent
cohort data to check the calibration of the new BCSC-PRS model
in the general population. Further, our studies consisted of pri-
marily white women, and the translation of these ndings to
other racial and ethnic groups is unknown. A strength of our
work was independent conrmation in three data sets that PRS
adds discriminatory accuracy to BI-RADS breast density.
In summary, we found that BI-RADS breast density and a PRS
composed of 76 SNPs are both important risk factors for breast
cancer that can be incorporated into breast cancer risk models.
If these models are used to estimate population risk, rening the
high- and low-risk risk groups could result in more appropriate
tailoring of screening and prevention interventions.
This work was supported by the National Cancer Institute
(R01 CA128931, R01 CA128978, R01 CA97396, P50 CA116201, R01
CA240386, K24 CA169004, R21 CA179442, P01 CA154292, R01
CA140286, Cancer Center Support Grant CA15083) and the Breast
Cancer Research Foundation. BBCC was supported in part by
the Erlanger Leistungsbezogene Anschubsnanzierung und
Table1. Contribution of continuous and quartile polygenic risk score measures to the Breast Imaging Reporting and Data System breast
density and breast cancer association
Odds ratios (95% CIs) corresponding to BI-RADS breast density and PRS measures in 4 models*
BI-RADS alone
Quartiles PRS alone
Quartiles of PRS †
Continuous PRS†‡
BI-RADS density
Almost entirely fat 0.55 (0.45 to 0.68) –§ 0.57 (0.46 to 0.70) 0.56 (0.45 to 0.70)
Scattered broglandular
1.00 (Ref.) 1.00 (Ref.) 1.00 (Ref.)
Heterogeneously dense 1.25 (1.07 to 1.46) 1.24 (1.06 to 1.45) 1.24 (1.06 to 1.45)
Extremely dense 1.77 (1.38 to 2.26) 1.75 (1.36 to 2.24) 1.73 (1.35 to 2.23)
PRS quartiles
<0.50 –§ 0.63 (0.52 to 0.78) 0.62 (0.51 to 0.77) –§
0.50–0.86 1.00 (Ref.) 1.00 (Ref.)
0.87–1.27 1.52 (1.26 to 1.83) 1.48 (1.23 to 1.79)
1.28+ 1.79 (1.50 to 2.14) 1.74 (1.45 to 2.09)
PRS continuous‡ –§ –§ –§ 1.48 (1.38 to 1.58)
AUC 0.66 (0.64 to 0.68) 0.68 (0.66 to 0.69) 0.69 (0.67 to 0.70) 0.69 (0.67 to 0.71)
* All three studies combined (n=1643 case patients, 2397 control patients). Models adjusted for age, 1/BMI, study. AUC = area under the curve; BI-RADS = Breast Imag-
ing Reporting and Data System; CI = condence interval; PRS = polygenic risk score.
† Likelihood Ratio Test PLRT < .001 for models with PRS and BI-RADS density compared with model with BI-RADS alone. All statistical tests were two-sided.
‡ Continuous PRS measure evaluated as per 1 standard deviation. Model with PRS continuous measure alone had OR=1.49 (95% CI=1.39 to 1.59) and AUC=0.68
(95% CI=0.66 to 0.69).
§ Variable not evaluated in this model.
Table2. Association (odds ratios and 95% condence intervals) of Breast Imaging Reporting and Data System breast density and polygenic
risk score with breast cancer*
PRS quartiles
Odds ratios (95% CI) corresponding to categories of BI-RADS density and PRS quartiles
Almost entirely fat Scattered broglandular densities Heterogeneously dense Extremely dense Pinteraction
<0.50 0.30 (0.18 to 0.51) 0.65 (0.48 to 0.89) 0.80 (0.57 to 1.14) 0.91 (0.53 to 1.56)
0.50–0.86 0.44 (0.28 to 0.68) 1.00 (Ref.) 1.40 (1.02 to 1.92) 1.56 (0.94 to 2.59)
0.87–1.27 0.93 (0.62 to 1.40) 1.32 (0.99 to 1.76) 1.81 (1.34 to 2.46) 3.59 (2.26 to 5.70)
1.28+ 1.20 (0.81 to 1.78) 1.84 (1.40 to 2.42) 1.91 (1.42 to 2.57) 2.68 (1.74 to 4.12)
* All three studies combined (n=1643 case patients, 2397 control patients). Models adjusted for age, 1/BMI and study. BI-RADS = Breast Imaging Reporting and Data
System; CI = condence interval; OR = odds ratio; PRS = polygenic risk score.
Pinteraction from logistic regression model including all three studies.
by guest on November 2, 2015 from
Vachon et al. | 4 of 4
Nachwuchsfoerderung program of the Medical Faculty, University
Hospital Erlangen, Friedrich-Alexander University Erlangen-
Nuremberg. Funding for the genotyping of BBCC and MCBCS, as
well as the custom Illumina iSelect genotyping array was provided
by grants from the EU FP7 programme (Collaborative Oncological
Gene-Environment Study) and from Cancer Research UK.
The Collaborative Oncological Gene-Environment Study enabled
the genotyping for this study.
1. Kolb TM, Lichy J, Newhouse JH. Comparison of the performance of screening
mammography, physical examination, and breast US and evaluation of fac-
tors that inuence them: an analysis of 27,825 patient evaluations. Radiol-
ogy. 2002;225(1):165–175.
2. Pisano ED, Gatsonis C, Hendrick E, etal. Diagnostic performance of digi-
tal versus lm mammography for breast-cancer screening. N Engl J Med.
3. McCormack VA, dos Santos Silva I. Breast density and parenchymal patterns
as markers of breast cancer risk: a meta-analysis. Cancer Epidemiol Bio-
markers Prev. 2006;15(6):1159–1169.
4. Kerlikowske K, Cook AJ, Buist DS, etal. Breast cancer risk by breast den-
sity, menopause, and postmenopausal hormone therapy use. J Clin Oncol.
5. National Cancer Institute. Breast Cancer Surveillance Consortium 2011.
Available at:
6. Ikeda D, Lipson J, Daniel B, etal. Breast density, breast cancer risk, and cali-
fornia breast density notication law SB 1538: senarios for clinicians. 2013; Accessed May 15, 2014.
7. Price ER, Hargreaves J, Lipson JA, etal. The California Breast Density Infor-
mation Group: A collaborative response to the issues of breast density,
breast cancer risk, and breast density notication legislation. Radiology.
8. Michailidou K, Hall P, Gonzalez-Neira A, etal. Large-scale genotyping identi-
es 41 new loci associated with breast cancer risk. Nat Genet. 2013;45(4):353–
9. Thomas G, Jacobs KB, Kraft P, etal. A multistage genome-wide association
study in breast cancer identies two new risk alleles at 1p11.2 and 14q24.1
(RAD51L1). Nat Genet. 2009;41(5):579–584.
10. Garcia-Closas M, Couch FJ, Lindstrom S, etal. Genome-wide association
studies identify four ER negative-specic breast cancer risk loci. Nature
Genet. 2013;45(4):392–398.
11. Cox A, Dunning AM, Garcia-Closas M, etal. A common coding variant in
CASP8 is associated with breast cancer risk. Nat Genet. 2007;39(3):352–358.
12. Stacey SN, Manolescu A, Sulem P, etal. Common variants on chromosomes
2q35 and 16q12 confer susceptibility to estrogen receptor-positive breast
cancer. Nat Genet. 2007;39(7):865–869.
13. Ahmed S, Thomas G, Ghoussaini M, etal. Newly discovered breast cancer
susceptibility loci on 3p24 and 17q23.2. Nat Genet. 2009;41(5):585–590.
14. Haiman CA, Chen GK, Vachon CM, etal. A common variant at the TERT-
CLPTM1L locus is associated with estrogen receptor-negative breast cancer.
Nat Genet. 2011;43(12):1210–1214.
15. Bojesen SE, Pooley KA, Johnatty SE, etal. Multiple independent variants at
the TERT locus are associated with telomere length and risks of breast and
ovarian cancer. Nature Genet. 2013;45(4):371–384.
16. Stacey SN, Manolescu A, Sulem P, et al. Common variants on chromosome
5p12 confer susceptibility to estrogen receptor-positive breast cancer. Nat
Genet. 2008;40(6):703–706.
17. Easton DF, Pooley KA, Dunning AM, etal. Genome-wide association study
identies novel breast cancer susceptibility loci. Nature. 2007;447:1087–1093.
18. Turnbull C, Ahmed S, Morrison J, etal. Genome-wide association study identi-
es ve new breast cancer susceptibility loci. Nat Genet. 2010;42(6):504–507.
19. Zheng W, Long J, Gao YT, etal. Genome-wide association study identies a
new breast cancer susceptibility locus at 6q25.1. Nat Genet. 2009;41(3):324–
20. Fletcher O, Johnson N, Orr N, etal. Novel Breast Cancer Susceptibility Locus
at 9q31.2: Results of a Genome-Wide Association Study. J Natl Cancer Inst.
21. French JD, Ghoussaini M, Edwards SL, etal. Functional variants at the 11q13
risk locus for breast cancer regulate cyclin D1 expression through long-
range enhancers. Am J Hum Genet. 2013;92(4):489–503.
22. Ghoussaini M, Fletcher O, Michailidou K, et al. Genome-wide association
analysis identies three new breast cancer susceptibility loci. Nat Genet.
23. Antoniou AC, Wang X, Fredericksen ZS, etal. A locus on 19p13 modies risk
of breast cancer in BRCA1 mutation carriers and is associated with hor-
mone receptor-negative breast cancer in the general population. Nat Genet.
24. Tice JA, Cummings SR, Smith-Bindman R, Ichikawa L, Barlow WE, Ker-
likowske K. Using clinical factors and mammographic breast density to esti-
mate breast cancer risk: development and validation of a new predictive
model. Ann Intern Med. 2008;148(5):337–347.
25. Olson JE, Sellers TA, Scott CG, etal. The inuence of mammogram acquisition
on the mammographic density and breast cancer association in the Mayo
Mammography Health Study Cohort. Breast Cancer Res. 2012;14(6):R147.
26. Heine JJ, Scott CG, Sellers TA, etal. A novel automated mammographic den-
sity measure and breast * cancer risk. J Natl Cancer Inst. 2012;104(13):1028–
27. Kelemen LE, Couch FJ, Ahmed S, etal. Genetic variation in stromal proteins
decorin and lumican with breast cancer: investigations in two case-control stud-
ies. Breast Cancer Res. 2008;10(6):R98.
28. Heusinger K, Loehberg CR, Haeberle L, etal. Mammographic density as a risk
factor for breast cancer in a German case-control study. Eur J Cancer Prev.
29. Rauh C, Hack CC, Haberle L, etal. Percent mammographic density and dense
area as risk factors for breast cancer. Geburtsh Frauenheilk. 2012;72(8):727–
30. American College of Radiology. The American College of Radiology Breast
Imaging Reporting and Data System (BI-RADS), 4th ed Reston, VA: American
College of Radiology; 2003.
31. Darabi H, Czene K, Zhao W, Liu J, Hall P, Humphreys K. Breast cancer risk
prediction and individualised screening based on common genetic variation
and breast density measurement. Breast Cancer Res. 2012;14(1):R25.
32. Harrell FE, Jr., Lee KL, Mark DB. Multivariable prognostic models: issues in
developing models, evaluating assumptions and adequacy, and measuring
and reducing errors. Stat Med. 1996;15(4):361–387.
33. Harrell FE Jr, Califf RM, Pryor DB, Lee KL, Rosati RA. Evaluating the yield of
medical tests. JAMA. 1982;247(18):2543–2546.
34. Prentice RL, Pyke R. Logistic Disease Incidence Models and Case-Control
Studies. Biometrika. 1979;66(3):403–411.
35. Hosmer DW, Lemeshow S. Applied Logistic Regression, 2nd ed. New York, NY:
Wiley-Interscience; 2000.
36. Mealiffe ME, Stokowski RP, Rhees BK, Prentice RL, Pettinger M, Hinds DA.
Assessment of clinical validity of a breast cancer risk model combining
genetic and clinical information. J Natl Cancer Inst. 2010;102(21):1618–
37. Schafer JL. Multiple imputation: a primer. Stat Methods Med Res. 1999;8(1):3–
38. Moyer VA. Medications for risk reduction of primary breast cancer in women:
U.S. Preventive Services Task Force recommendation statement. Ann Intern
Med. 2013;159(10):698–708.
39. Vachon CM, Ghosh K, Brandt KR. Mammographic density: potential as a risk
factor and surrogate marker in the clinical setting. Curr Breast Cancer Rep.
40. Visvanathan K, Hurley P, Bantug E, etal. Use of pharmacologic interventions
for breast cancer risk reduction: American Society of Clinical Oncology clini-
cal practice guideline. J Clin Oncol. 2013;31(23):2942–2962.
41. Fasching PA, von Minckwitz G, Fischer T, etal. The impact of breast cancer
awareness and socioeconomic status on willingness to receive breast cancer
prevention drugs. Breast Cancer Res Treat. 2007;101(1):95–104.
42. Loehberg CR, Jud SM, Haeberle L, etal. Breast cancer risk assessment in a
mammography screening program and participation in the IBIS-II chemo-
prevention trial. Breast Cancer Res Treat. 2010;121(1):101–110.
by guest on November 2, 2015 from
... Existing studies of European ancestries have consistently shown that there is at least a twofold difference in breast cancer risk between the lowest and highest quartiles of PRS distributions [1,[3][4][5]. The addition of PRS to clinical risk models has also been shown to improve discrimination and reclassification abilities [6,7]. Ongoing clinical trials are examining the effect of personalized medicine on breast cancer screening and prevention using PRS [8][9][10]. ...
Full-text available
PurposeThis study aimed to develop an ancestry-specific polygenic risk scores (PRSs) for the prediction of breast cancer events in Japanese females and validate it in a longitudinal cohort study.Methods Using publicly available summary statistics of female breast cancer genome-wide association study (GWAS) of Japanese and European ancestries, we, respectively, developed 31 candidate genome-wide PRSs using pruning and thresholding (P + T) and LDpred methods with varying parameters. Among the candidate PRS models, the best model was selected using a case-cohort dataset (63 breast cancer cases and 2213 sub-cohorts of Japanese females during a median follow-up of 11.9 years) according to the maximal predictive ability by Harrell’s C-statistics. The best-performing PRS for each derivation GWAS was evaluated in another independent case-cohort dataset (260 breast cancer cases and 7845 sub-cohorts of Japanese females during a median follow-up of 16.9 years).ResultsFor the best PRS model involving 46,861 single nucleotide polymorphisms (SNPs; P + T method with PT = 0.05 and R2 = 0.2) derived from Japanese-ancestry GWAS, the Harrell’s C-statistic was 0.598 ± 0.018 in the evaluation dataset. The age-adjusted hazard ratio for breast cancer in females with the highest PRS quintile compared with those in the lowest PRS quintile was 2.47 (95% confidence intervals, 1.64–3.70). The PRS constructed using Japanese-ancestry GWAS demonstrated better predictive performance for breast cancer in Japanese females than that using European-ancestry GWAS (Harrell’s C-statistics 0.598 versus 0.586).Conclusion This study developed a breast cancer PRS for Japanese females and demonstrated the usefulness of the PRS for breast cancer risk stratification.
... Bei den nicht genetischen Risikofaktoren stehen die mammografische Dichte [16][17][18] und reproduktive Faktoren im Fokus [19][20][21][22][23][24][25][26][27]. Zusätzlich sind insbesondere Alter bei Menarche und bei Menopause und die Anzahl der Kinder ebenso wie die Dauer des Stillens gut untersuchte Risikofaktoren [19,21]. ...
Full-text available
Zusammenfassung In dieser Übersichtsarbeit werden neueste Entwicklungen in der Prävention von Brustkrebs und Behandlung von Patientinnen mit frühen Krankheitsstadien mit Mammakarzinom zusammengefasst. Die Ermittlung von individuellen Erkrankungsrisiken nach molekularen Subtypen wurde in einer großen epidemiologischen Studie untersucht. Im Bereich der Behandlung gibt es neue Daten zur Langzeitnachbeobachtung der Aphinity-Studie ebenso wie neue Daten zur neoadjuvanten Therapie von HER2-positiven Patientinnen mit Atezolizumab. Biomarker wie Residual Cancer Burden wurden im Zusammenhang mit einer Pembrolizumab-Therapie untersucht. Eine Untersuchung des Genomic-Grade-Indexes bei älteren Patientinnen reiht sich ein in die Gruppe von Studien, die versucht, durch moderne Multigentests Patientinnen zu identifizieren, bei denen eine Chemotherapie vermieden werden kann, weil diese eine exzellente Prognose haben. Diese und weitere Aspekte der neuesten Entwicklungen bei der Diagnostik und Therapie des Mammakarzinoms werden in dieser Übersichtsarbeit beschrieben.
... In recent years, some risk models have been updated to include additional genomic information, typically the effects of rare PV in other genes (ATM, CHEK2 and PALB2) [2,3] and the joined effect of single nucleotide polymorphisms (SNPs) summarized in polygenic risk scores (PRS) [4][5][6]. Indeed, in the general population, some studies suggested that stratification of women according to their risk of breast cancer based on their PRS could personalize screening and prevention strategies [7][8][9][10]. Several PRS for breast cancer have been defined and validated in women of European ancestry from the large multi-centric and multicountry studies conducted by the Breast Cancer Association Consortium (BCAC). ...
Full-text available
Background Three partially overlapping breast cancer polygenic risk scores (PRS) comprising 77, 179 and 313 SNPs have been proposed for European-ancestry women by the Breast Cancer Association Consortium (BCAC) for improving risk prediction in the general population. However, the effect of these SNPs may vary from one country to another and within a country because of other factors. Objective To assess their associated risk and predictive performance in French women from (1) the CECILE population-based case-control study, (2) BRCA1 or BRCA2 (BRCA1/2) pathogenic variant (PV) carriers from the GEMO study, and (3) familial breast cancer cases with no BRCA1/2 PV and unrelated controls from the GENESIS study. Results All three PRS were associated with breast cancer in all studies, with odds ratios per standard deviation varying from 1.7 to 2.0 in CECILE and GENESIS, and hazard ratios varying from 1.1 to 1.4 in GEMO. The predictive performance of PRS313 in CECILE was similar to that reported in BCAC but lower than that in GENESIS (area under the receiver operating characteristic curve (AUC)=0.67 and 0.75, respectively). PRS were less performant in BRCA2 and BRCA1 PV carriers (AUC = 0.58 and 0.54 respectively). Conclusion Our results are in line with previous validation studies in the general population and in BRCA1/2 PV carriers. Additionally, we showed that PRS may be of clinical utility for women with a strong family history of breast cancer and no BRCA1/2 PV, and for those carrying a predicted PV in a moderate-risk gene like ATM, CHEK2 or PALB2.
... Several studies have shown that the addition of PRS to the classical breast cancer risk factors improves risk stratification. Vachon et al. [50], based on the analysis of three casecontrol studies (1643 cases and 2397 controls) conducted in Germany and the USA, found that a 76-locus PRS and breast density were independent risk factors and that incorporating the PRS into the BCSC risk model improved the model fit and net reclassification for case patients. For instance, the PRS further stratified the risk associated with extremely dense breasts such that those with the lowest PRS had an odds ratio of 0.91, while those in the highest PRS had a 2.7-fold increased risk compared with women with scattered fibroglandular densities and average PRS. ...
Full-text available
The aim of this study was to assess the acceptability and feasibility of offering risk-based breast cancer screening and its integration into regular clinical practice. A single-arm proof-of-concept trial was conducted with a sample of 387 women aged 40–50 years residing in the city of Lleida (Spain). The study intervention consisted of breast cancer risk estimation, risk communication and screening recommendations, and a follow-up. A polygenic risk score with 83 single nucleotide polymorphisms was used to update the Breast Cancer Surveillance Consortium risk model and estimate the 5-year absolute risk of breast cancer. The women expressed a positive attitude towards varying the frequency of breast screening according to individual risk and, especially, more frequently inviting women at higher-than-average risk. A lower intensity screening for women at lower risk was not as welcome, although half of the participants would accept it. Knowledge of the benefits and harms of breast screening was low, especially with regard to false positives and overdiagnosis. The women expressed a high understanding of individual risk and screening recommendations. The participants’ intention to participate in risk-based screening and satisfaction at 1-year were very high.
Objective To assess effectiveness of a web-based educational intervention on women’s health care provider knowledge of breast cancer risk models and high-risk screening recommendations. Methods A web-based pre- and post-test study including 177 U.S.-based women’s health care providers was conducted in 2019. Knowledge gaps were defined as fewer than 75% of respondents answering correctly. Pre- and post-test knowledge differences (McNemar test) and associations of baseline characteristics with pre-test knowledge gaps (logistic regression) were evaluated. Results Respondents included 131/177 (74.0%) physicians; 127/177 (71.8%) practiced obstetrics/gynecology. Pre-test, 118/177 (66.7%) knew the Gail model predicts lifetime invasive breast cancer risk; this knowledge gap persisted post-test [(121/177, 68.4%); P = 0.77]. Just 39.0% (69/177) knew the Gail model identifies women eligible for risk-reducing medications; this knowledge gap resolved. Only 48.6% (86/177) knew the Gail model should not be used to identify women meeting high-risk MRI screening guidelines; this deficiency decreased to 66.1% (117/177) post-test (P = 0.001). Pre-test, 47.5% (84/177) knew the Tyrer-Cuzick model is used to identify women meeting high-risk screening MRI criteria, 42.9% (76/177) to predict BRCA1/2 pathogenic mutation risk, and 26.0% (46/177) to predict lifetime invasive breast cancer risk. These knowledge gaps persisted but improved. For a high-risk 30-year-old, 67.8% (120/177) and 54.2% (96/177) pre-test knew screening MRI and mammography/tomosynthesis are recommended, respectively; 19.2% (34/177) knew both are recommended; and 53% (94/177) knew US is not recommended. These knowledge gaps resolved or reduced. Conclusion Web-based education can reduce important provider knowledge gaps about breast cancer risk models and high-risk screening recommendations.
Mammographic breast density is widely accepted as an independent risk factor for the development of breast cancer. In addition, because dense breast tissue may mask breast malignancies, breast density is inversely related to the sensitivity of screening mammography. Given the risks associated with breast density, as well as ongoing efforts to stratify individual risk and personalize breast cancer screening and prevention, numerous studies have sought to better understand the factors that impact breast density, and to develop and implement reproducible, quantitative methods to assess mammographic density. Breast density assessments have been incorporated into risk assessment models to improve risk stratification. Recently, novel techniques for analyzing mammographic parenchymal complexity, or texture, have been explored as potential means of refining mammographic tissue-based risk assessment beyond breast density.
Breast cancer is the most common cancer among women in the United States and growing in incidence worldwide, while mortality due to breast cancer is slowly decreasing. The last two decades have witnessed many changes in the diagnostic landscape both in terms of imaging studies and laboratory tests, most notably next‐generation sequencing both on tumor and blood. Our understanding of breast oncogenesis, tumor heterogeneity, and metastasis has deepened. Segmental mastectomy and sentinel lymph node (SLN) biopsy have been used in the majority of cases coupled with radiation therapy. In cases managed with mastectomy, radiation of one to three positive nodes has been utilized more often. Gene expression profiling has been used to identify women with hormone receptor‐positive (HR+) breast cancer who would benefit from adjuvant chemotherapy. Targeted therapy has been developed for HR+, HER2 amplified, and triple‐negative breast cancer. Immune checkpoint inhibitors have improved the management of triple‐negative disease. poly(adenosine diphosphate [ADP]–ribose) (PARP) inhibitors have improved outcomes for breast cancer patients with BRCA1/2 mutations. Combinations of targeted therapies and hormonal therapies have proved effective for adjuvant therapy and in recurrent disease.
Full-text available
This review summarizes recent developments in the prevention and treatment of patients with early-stage breast cancer. The individual disease risk for different molecular subtypes was investigated in a large epidemiological study. With regard to treatment, new data are available from long-term follow-up of the Aphinity study, as well as new data on neoadjuvant therapy with atezolizumab in HER2-positive patients. Biomarkers, such as residual cancer burden, were investigated in the context of pembrolizumab therapy. A Genomic Grade Index study in elderly patients is one of a group of studies investigating the use of modern multigene tests to identify patients with an excellent prognosis in whom chemotherapy may be avoided. These and other aspects of the latest developments in the diagnosis and treatment of breast cancer are described in this review.
There has been increased interest in the use of clinical risk prediction models for decision-making in medicine for patient care. This has been accelerated through the focus on precision medicine, the revolution in omics data, and increasing use of randomized controlled trial and electronic health record databases. These models are expected to assist diagnostic assessment, prognostication, and therapeutic decision-making. Randomized controlled trial data are highly relevant for modeling treatment benefit and treatment effect heterogeneity. The development and validation of prediction models requires careful methodology and reporting, and an evidence-based approach is needed to bring risk prediction models to clinical practice. This chapter provides an overview of the key steps and considerations to develop and validate risk prediction models. We comment on the role of clinical trials throughout the process. A risk prediction model for the occurrence of breast cancer is used as an example.
Full-text available
Description: Update of the 2004 U. S. Preventive Services Task Force (USPSTF) recommendation on screening for glaucoma. Methods: The USPSTF reviewed evidence on the benefits and harms of screening for glaucoma and of medical and surgical treatment of early glaucoma. Beneficial outcomes of interest included improved vision-related quality of life and reduced progression of early asymptomatic glaucoma to vision-related impairment. The USPSTF also considered evidence on the accuracy of glaucoma screening tests. Population: This recommendation applies to adults who do not have vision symptoms and are seen in a primary care setting. Recommendation: The USPSTF concludes that the current evidence is insufficient to assess the balance of benefits and harms of screening for primary open-angle glaucoma in adults. (I statement)
In recent years, multiple imputation has emerged as a convenient and flexible paradigm for analysing data with missing values. Essential features of multiple imputation are reviewed, with answers to frequently asked questions about using the method in practice.
Multivariable regression models are powerful tools that are used frequently in studies of clinical outcomes. These models can use a mixture of categorical and continuous variables and can handle partially observed (censored) responses. However, uncritical application of modelling techniques can result in models that poorly fit the dataset at hand, or, even more likely, inaccurately predict outcomes on new subjects. One must know how to measure qualities of a model's fit in order to avoid poorly fitted or overfitted models. Measurement of predictive accuracy can be difficult for survival time data in the presence of censoring. We discuss an easily interpretable index of predictive discrimination as well as methods for assessing calibration of predicted survival probabilities. Both types of predictive accuracy should be unbiasedly validated using bootstrapping or cross-validation, before using predictions in a new data series. We discuss some of the hazards of poorly fitted and overfitted regression models and present one modelling strategy that avoids many of the problems discussed. The methods described are applicable to all regression models, but are particularly needed for binary, ordinal, and time-to-event outcomes. Methods are illustrated with a survival analysis in prostate cancer using Cox regression.
Estrogen receptor (ER)-negative breast cancer shows a higher incidence in women of African ancestry compared to women of European ancestry. In search of common risk alleles for ER-negative breast cancer, we combined genome-wide association study (GWAS) data from women of African ancestry (1,004 ER-negative cases and 2,745 controls) and European ancestry (1,718 ER-negative cases and 3,670 controls), with replication testing conducted in an additional 2,292 ER-negative cases and 16,901 controls of European ancestry. We identified a common risk variant for ER-negative breast cancer at the TERT-CLPTM1L locus on chromosome 5p15 (rs10069690: per-allele odds ratio (OR) = 1.18 per allele, P = 1.0 × 10?10). The variant was also significantly associated with triple-negative (ER-negative, progesterone receptor (PR)-negative and human epidermal growth factor-2 (HER2)-negative) breast cancer (OR = 1.25, P = 1.1 × 10?9), particularly in younger women (<50 years of age) (OR = 1.48, P = 1.9 × 10?9). Our results identify a genetic locus associated with estrogen receptor negative breast cancer subtypes in multiple populations.
Estrogen receptor (ER)-negative tumors represent 20–30% of all breast cancers, with a higher proportion occurring in younger women and women of African ancestry1 . The etiology2 and clinical behavior3 of ER-negative tumors are different from those of tumors expressing ER (ER positive), including differences in genetic predisposition4 . To identify susceptibility loci specific to ER-negative disease, we combined in a meta-analysis 3 genome-wide association studies of 4,193 ER-negative breast cancer cases and 35,194 controls with a series of 40 follow-up studies (6,514 cases and 41,455 controls), genotyped using a custom Illumina array, iCOGS, developed by the Collaborative Oncological Gene-environment Study (COGS). SNPs at four loci, 1q32.1 (MDM4, P = 2.1 × 10−12 and LGR6, P = 1.4 × 10−8), 2p24.1 (P = 4.6 × 10−8) and 16q12.2 (FTO, P = 4.0 × 10−8), were associated with ER-negative but not ER-positive breast cancer (P > 0.05). These findings provide further evidence for distinct etiological pathways associated with invasive ER-positive and ER-negative breast cancers.
Breast cancer is the most common cancer among women. To date, 22 common breast cancer susceptibility loci have been identified accounting for ∼8% of the heritability of the disease. We attempted to replicate 72 promising associations from two independent genome-wide association studies (GWAS) in ∼70,000 cases and ∼68,000 controls from 41 case-control studies and 9 breast cancer GWAS. We identified three new breast cancer risk loci at 12p11 (rs10771399; P = 2.7 × 10(-35)), 12q24 (rs1292011; P = 4.3 × 10(-19)) and 21q21 (rs2823093; P = 1.1 × 10(-12)). rs10771399 was associated with similar relative risks for both estrogen receptor (ER)-negative and ER-positive breast cancer, whereas the other two loci were associated only with ER-positive disease. Two of the loci lie in regions that contain strong plausible candidate genes: PTHLH (12p11) has a crucial role in mammary gland development and the establishment of bone metastasis in breast cancer, and NRIP1 (21q21) encodes an ER cofactor and has a role in the regulation of breast cancer cell growth.
The Breast Cancer Association Consortium (BCAC) has been established to conduct combined case-control analyses with augmented statistical power to try to confirm putative genetic associations with breast cancer. We genotyped nine SNPs for which there was some prior evidence of an association with breast cancer: CASP8 D302H (rs1045485), IGFBP3 -202 C --> A (rs2854744), SOD2 V16A (rs1799725), TGFB1 L10P (rs1982073), ATM S49C (rs1800054), ADH1B 3' UTR A --> G (rs1042026), CDKN1A S31R (rs1801270), ICAM5 V301I (rs1056538) and NUMA1 A794G (rs3750913). We included data from 9-15 studies, comprising 11,391-18,290 cases and 14,753-22,670 controls. We found evidence of an association with breast cancer for CASP8 D302H (with odds ratios (OR) of 0.89 (95% confidence interval (c.i.): 0.85-0.94) and 0.74 (95% c.i.: 0.62-0.87) for heterozygotes and rare homozygotes, respectively, compared with common homozygotes; P(trend) = 1.1 x 10(-7)) and weaker evidence for TGFB1 L10P (OR = 1.07 (95% c.i.: 1.02-1.13) and 1.16 (95% c.i.: 1.08-1.25), respectively; P(trend) = 2.8 x 10(-5)). These results demonstrate that common breast cancer susceptibility alleles with small effects on risk can be identified, given sufficiently powerful studies.
Breast cancer exhibits familial aggregation, consistent with variation in genetic susceptibility to the disease. Known susceptibility genes account for less than 25% of the familial risk of breast cancer, and the residual genetic variance is likely to be due to variants conferring more moderate risks. To identify further susceptibility alleles, we conducted a two-stage genome-wide association study in 4,398 breast cancer cases and 4,316 controls, followed by a third stage in which 30 single nucleotide polymorphisms (SNPs) were tested for confirmation in 21,860 cases and 22,578 controls from 22 studies. We used 227,876 SNPs that were estimated to correlate with 77% of known common SNPs in Europeans at r(2) > 0.5. SNPs in five novel independent loci exhibited strong and consistent evidence of association with breast cancer (P < 10(-7)). Four of these contain plausible causative genes (FGFR2, TNRC9, MAP3K1 and LSP1). At the second stage, 1,792 SNPs were significant at the P < 0.05 level compared with an estimated 1,343 that would be expected by chance, indicating that many additional common susceptibility alleles may be identifiable by this approach.