(166) Drug Interactions with Tamoxifen in a Danish Premenopausal Breast Cancer Cohort

Friday, August 26, 2022

8:00 AM – 6:00 PM CEST

Location: Hall C

Publication Number: 497

Presenting Author(s)

Kirsten M. Woolpert
– Department of Clinical Epidemiology, Aarhus University, Aarhus N, Midtjylland, Denmark

Co-Author(s)

Lindsay J. Collin
– University of Utah, Salt Lake City, UT, United States
Thomas P. Ahern
– Department of Surgery and Biochemistry, Larner College of Medicine, University of Vermont, Burlington, VT, USA, United States
Per Damkier, MD, PhD
– Professor, consultant, Department of Clinical Pharmacology, Odense University Hospital, Denmark and Department of Clinical Research, University of Southern Denmark, Denmark, Denmark
Anders Kjærsgaard
– Department of Clinical Epidemiology, Department of Clinical Medicine, Aarhus University and Aarhus University Hospital, Aarhus, Denmark, Denmark
Henrik Toft Sørensen, MD, PhD, DMSc, DSc
– Professor & Chair, Department of Clinical Epidemiology, Aarhus University and Aarhus University Hospital, Aarhus, Denmark, Midtjylland, Denmark
Stephen Hamilton-Dutoit
– Department of Pathology, Aarhus University Hospital, Aarhus N, Denmark, Denmark
Bent Ejlertsen
– Department of Oncology, Rigshospitalet, Copenhagen University, Copenhagen, Denmark, Denmark
Peer Christiansen
– Department of Plastic and Breast Surgery, Aarhus University Hospital, Denmark
Rebecca Silliman
– Section of Geriatrics, Department of Medicine, Boston University School of Medicine, Boston, Massachusetts, USA, United States
Timothy Lash
– Department of Epidemiology, Emory University, Atlanta, GA, United States
Deirdre Cronin-Fenton
– Department of Clinical Epidemiology, Department of Clinical Medicine, Aarhus University and Aarhus University Hospital, Aarhus, Denmark, Denmark

Background: Tamoxifen treatment roughly halves the recurrence rate among estrogen receptor-positive breast cancer patients. Therapeutic efficacy may to some extent hinge on successful biotransformation of tamoxifen to metabolites with higher estrogen receptor affinities. These biotransformation reactions are primarily catalyzed by the CYP2D6, CYP2C19, and CYP3A4 enzymes and each of these enzymes can be inhibited by comedications.

Objectives: We evaluated whether tamoxifen-treated premenopausal breast cancer patients have a higher recurrence rate when concomitantly prescribed a drug that potentially inhibits tamoxifen metabolism.

Methods: We enrolled a cohort of 5,959 premenopausal women diagnosed with nonmetastatic breast cancer between 2002–2011. We divided the cohort into women with estrogen receptor positive tumors who were treated with tamoxifen (ER+/T+) and women with estrogen receptor negative tumors who were not treated with tamoxifen (ER−/T−). Prescription drug exposures were ascertained from the Danish nationwide prescription registry. We fit Cox regression models and used a Bayesian joint modeling approach to estimate associations of exposure to drugs that are substrates for or inhibitors of CYP2D6, CYP2C19, and CYP3A4 with recurrence.

Results: In the conventional Cox regression models, concomitant prescription of inhibitors of CYP2D6 and CYP2C19 were not associated with recurrence in the ER+/T+ group (CYP2D6: HRadj=0.98, 95% CI: 0.74, 1.3; CYP2C19: HRadj=0.99, 95% CI: 0.71, 1.4). Inhibition of CYP3A4 was associated with an increased recurrence hazard among ER+/T+ women (HRadj=1.8, 95% CI: 1.1 to 2.0), but not among ER−/T− women. After incorporating the longitudinal structure of the CYP inhibiting medication with a Bayesian joint modeling framework, users of CYP3A4 inhibiting medication had a reduced estimated rate of recurrence compared with non-users among ER+/T+ women (HRcrude= 0.69, 95% CI: 0.44, 0.97). With this approach, concomitant use of CYP2D6 and CYP2C19 inhibitors were again not associated with recurrence in the ER+/T+ group (CYP2D6: HRcrude=1.01, 95% CI: 0.98, 1.04; CYP2C19: HRcrude=0.97, 95% CI: 0.93, 1.01).

Conclusions: The positive recurrence association seen with CYP3A4 inhibition was specific to the ER+/T+ group, as would be expected for a predictive marker of tamoxifen effectiveness. In the joint modeling approach, CYP3A4 inhibiting medications were associated with a reduced estimated rate of recurrence, suggesting that the association observed in the conventional approach is likely biased. All associations warrant further study with incorporation of functional variants in the genes encoding these enzymes.