TY - JOUR
T1 - Mammographic Density Decline, Tamoxifen Response, and Prognosis by Molecular Characteristics of Estrogen Receptor-Positive Breast Cancer
AU - Abubakar, Mustapha
AU - Mullooly, Maeve
AU - Nyante, Sarah
AU - Pfeiffer, Ruth M.
AU - Aiello Bowles, Erin J.
AU - Cora, Renata
AU - Bodelon, Clara
AU - Butler, Ebone
AU - Butcher, Donna
AU - Sternberg, Lawrence
AU - Troester, Melissa A.
AU - Weinmann, Sheila
AU - Sherman, Mark
AU - Glass, Andrew G.
AU - Berrington De Gonzalez, Amy
AU - Gierach, Gretchen L.
N1 - Funding Information:
Erin Aiello Bowles is supported by the National Cancer Institute R50CA211115. Maeve Mullooly is supported by funding from the Health Research Board in Ireland (Emerging Investigator Award EIA-2019-012).
Publisher Copyright:
© 2022 Published by Oxford University Press.
PY - 2022/6/1
Y1 - 2022/6/1
N2 - Background: Mammographic breast density (MBD) decline post-tamoxifen initiation is a favorable prognostic factor in estrogen receptor (ER)-positive breast cancer (BC) and has potential utility as a biomarker of tamoxifen response. However, the prognostic value of MBD decline may vary by molecular characteristics among ER-positive patients. Methods: We investigated associations between MBD decline (≥10% vs <10%) and breast cancer-specific mortality (BCSM) among ER-positive breast cancer patients aged 36-87 years at diagnosis treated with tamoxifen at Kaiser Permanente Northwest (1990-2008). Patients who died of BC (case patients; n = 62) were compared with those who did not (control patients; n = 215) overall and by tumor molecular characteristics (immunohistochemistry [IHC]-based subtype [luminal A-like: ER-positive/progesterone receptor [PR]-positive/HER2-negative/low Ki67; luminal B-like: ER-positive and 1 or more of PR-negative, HER2-positive, high Ki67] and modified IHC [mIHC]-based recurrence score of ER/PR/Ki67). Percent MBD was measured in the unaffected breast at baseline mammogram (mean = 6 months before tamoxifen initiation) and follow-up (mean = 12 months post-tamoxifen initiation). Adjusted odds ratios (ORs) and 95% confidence intervals (CIs) were computed from logistic regression models. All statistical tests were 2-sided. Results: MBD decline was statistically significantly associated with reduced risk of BCSM overall (OR = 0.38, 95% CI = 0.15 to 0.92). This association was, however, stronger among women with aggressive tumor characteristics including luminal B-like (OR = 0.17, 95% CI = 0.04 to 0.73) vs A-like (OR = 0.74, 95% CI = 0.19 to 2.92); large (OR = 0.26, 95% CI = 0.08 to 0.78) vs small (OR = 0.41, 95% CI = 0.04 to 3.79) tumors; PR-negative (OR = 0.02, 95% CI = 0.001 to 0.37) vs PR-positive (OR = 0.50, 95% CI = 0.18 to 1.40) disease; and high (OR = 0.25, 95% CI = 0.07 to 0.93) vs low (OR = 0.44, 95% CI = 0.10 to 2.09) mIHC3 score. Conclusion: The findings support MBD decline as a prognostic marker of tamoxifen response among patients with aggressive ER-positive BC phenotypes, for whom understanding treatment effectiveness is critical.
AB - Background: Mammographic breast density (MBD) decline post-tamoxifen initiation is a favorable prognostic factor in estrogen receptor (ER)-positive breast cancer (BC) and has potential utility as a biomarker of tamoxifen response. However, the prognostic value of MBD decline may vary by molecular characteristics among ER-positive patients. Methods: We investigated associations between MBD decline (≥10% vs <10%) and breast cancer-specific mortality (BCSM) among ER-positive breast cancer patients aged 36-87 years at diagnosis treated with tamoxifen at Kaiser Permanente Northwest (1990-2008). Patients who died of BC (case patients; n = 62) were compared with those who did not (control patients; n = 215) overall and by tumor molecular characteristics (immunohistochemistry [IHC]-based subtype [luminal A-like: ER-positive/progesterone receptor [PR]-positive/HER2-negative/low Ki67; luminal B-like: ER-positive and 1 or more of PR-negative, HER2-positive, high Ki67] and modified IHC [mIHC]-based recurrence score of ER/PR/Ki67). Percent MBD was measured in the unaffected breast at baseline mammogram (mean = 6 months before tamoxifen initiation) and follow-up (mean = 12 months post-tamoxifen initiation). Adjusted odds ratios (ORs) and 95% confidence intervals (CIs) were computed from logistic regression models. All statistical tests were 2-sided. Results: MBD decline was statistically significantly associated with reduced risk of BCSM overall (OR = 0.38, 95% CI = 0.15 to 0.92). This association was, however, stronger among women with aggressive tumor characteristics including luminal B-like (OR = 0.17, 95% CI = 0.04 to 0.73) vs A-like (OR = 0.74, 95% CI = 0.19 to 2.92); large (OR = 0.26, 95% CI = 0.08 to 0.78) vs small (OR = 0.41, 95% CI = 0.04 to 3.79) tumors; PR-negative (OR = 0.02, 95% CI = 0.001 to 0.37) vs PR-positive (OR = 0.50, 95% CI = 0.18 to 1.40) disease; and high (OR = 0.25, 95% CI = 0.07 to 0.93) vs low (OR = 0.44, 95% CI = 0.10 to 2.09) mIHC3 score. Conclusion: The findings support MBD decline as a prognostic marker of tamoxifen response among patients with aggressive ER-positive BC phenotypes, for whom understanding treatment effectiveness is critical.
UR - http://www.scopus.com/inward/record.url?scp=85142337560&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85142337560&partnerID=8YFLogxK
U2 - 10.1093/jncics/pkac028
DO - 10.1093/jncics/pkac028
M3 - Article
C2 - 35583138
AN - SCOPUS:85142337560
SN - 2515-5091
VL - 6
JO - JNCI Cancer Spectrum
JF - JNCI Cancer Spectrum
IS - 3
M1 - pkac028
ER -