(S025) Radiation-Induced Increases in PARP1 Activity Predict for Long-Term Radiosensitization by PARP1 Inhibition in Preclinical Breast Cancer Models

Felix Y. Feng, MD, Corey Speers, MD, PhD, Meilan Liu, MD, Kari Wilder-Romans, BS, Reshma Jagsi, MD, DPhil, Lori J. Pierce, MD; Department of Radiation Oncology, Comprehensive Cancer Center, University of Michigan

Purpose: Sustained locoregional (LR) control of breast cancer (BCa) is a significant issue, particularly for patients with inflammatory disease or chest wall recurrences. Inhibition of poly(adenosine diphosphate-ribose) (PAR) polymerase 1 (PARP1), a DNA damage–response protein, is a promising strategy for radiosensitization (RS). The goal of this study was to optimize combination therapy with PARP1 inhibition and radiation for clinical assessment by establishing the most effective drug/radiation treatment schedule, determining the degree of PARP1-mediated radiosensitization across a range of drug doses, and identifying early biomarkers predictive of long-term treatment efficacy in preclinical BCa models.

Experimental Design: Clonogenic survival assays in 11 BCa and normal epithelial cell lines were used to determine the degree of RS conferred by the PARP1 inhibitor ABT-888 and to optimize the sequencing of therapy. Immunoblots, immunofluorescent staining, flow cytometry, and xenograft studies were used to evaluate for pretreatment and intratreatment biomarkers predicting for subsequent response to therapy.

Results: Using clonogenic survival assays, we demonstrate that among treatment schedules incorporating neoadjuvant, concurrent, and/or adjuvant PARP1 inhibition in combination with radiation, concurrent, and adjuvant therapy resulted in the highest radiation enhancement ratios (EnhRs). Across a panel of 11 BCa and normal breast epithelial cell lines, the PARP1 inhibitor ABT-888 preferentially radiosensitized C (vs normal) cells with EnhRs of up to 2.3 across different subtypes of breast cancer cell lines. The degree of RS did not correlate with pretreatment markers of PARP1 activity (eg, PAR levels by immunoblot), DNA damage/repair (eg, gamma-H2AX and RAD51 foci by immunofluorescence), or cell cycle distribution (by flow cytometry). However, increases in PARP1 activity from pretreatment to 24 hours after radiation therapy (RT) were associated with long-term clonogenic death after treatment with ABT-888 + RT (Spearman’s coefficient = 0.8: P = .002). The cell lines that were most significantly radiosensitized by PARP1 inhibition (EnhR > 1.5) averaged a twofold increase in PAR levels following RT. Findings were also confirmed in the SKBR3 and MDA-231 BCa xenograft models, in which we demonstrated a significant difference in tumor doubling time between the combination-treated and radiation-alone arms in the sensitive cell line.

Conclusion: Our study demonstrates that PARP1 inhibition improves the therapeutic index of RT in BCa cell lines independent of intrinsic breast cancer subtype or BRCA1 mutational status. Treatment-induced increases in PARP1 activity 24 hours after RT predict for long-term RS by PARP1 inhibition, thus identifying it as a potential biomarker of response. These studies have led to a clinical trial incorporating intratreatment biomarker analyses of PARP1 inhibitors and radiation in BCa patients, which is currently open for accrual through the Translational Breast Cancer Research Consortium (TBCRC 024).

Proceedings of the 96th Annual Meeting of the American Radium Society - americanradiumsociety.org