Abstract 2697: Inhibition of estrogen receptor signaling as a strategy for radiosensitization of ER+ breast cancers

Purpose: The estrogen receptor (ER) is expressed in over 80% of breast tumors and has been shown to be a significant driver of breast cancer (BC) pathogenesis and therefore a target of effective first-line therapies. While both ionizing radiation (RT) and endocrine therapies (ET) are used for the treatment of ER+ BC, the effect of ET on tumor radiosensitization remains unclear, with concerns it may be radioprotective based on G1 cell arrest with ET treatment. Here we assessed the efficacy and mechanism of ER-mediated radiosensitization using various pharmacologic approaches in ER+ BC.

Methods: Radiosensitization with ER inhibitors (tamoxifen [TAM], fulvestrant [FULV], AZD9496) was assessed using clonogenic survival assays. DNA damage was assessed by the neutral comet assay. Efficiency of homologous recombination (HR) or non-homologous end joining (NHEJ) as well as changes in cell cycle, apoptosis, and senescence were assessed. The efficacy of TAM with RT in vivo was assessed with an MCF-7 xenograft model.

Results: The selective estrogen receptor modulator TAM radiosensitized ER+ MCF-7 (enhancement ratio [enhR]: 1.14-1.50) and T47D (enhR: 1.33-1.60) cells but not ER-negative SUM-159 cells (enhR: 0.99-1.02). The selective estrogen receptor degrader (SERD) FULV had similar radiosensitization effects in MCF-7 (enhR: 1.33-1.76) and T47D cells (enhR: 0.97-2.81) with no radiosensitization observed in SUM-159 cells (enhR: 1.01-1.03). The novel oral SERD AZD9496 radiosensitized MCF-7 cells (enhR: 1.36-1.56). MCF-7 cells treated with TAM and RT had an increase in dsDNA breaks compared to RT alone as measured by the comet assay (p<0.05) and a decrease in NHEJ-mediated repair with TAM (p<0.05). No changes were observed in HR-mediated repair by Rad51 foci or a reporter (p=NS). RT alone and in combination with TAM or FULV induced similar levels of cell cycle arrest, suggesting that radiosensitization with the combination therapy is cell-cycle independent. There were no significant changes in apoptosis with TAM, FULV, RT, or the combination (p=NS). Although TAM or FULV did induce senescence, ET with RT increased senescence induction (p<0.05). In vivo, combination RT and TAM led to a significant delay in days to tumor doubling (control: 17, TAM: 40, RT: 32, TAM+RT: undefined; p<0.0001), and a significant difference in tumor growth between mice treated with TAM or RT alone compared combination treatment, with no increased toxicities or skin lesions from the combination treatment. Conclusion: Our data suggest that TAM, FULV, or AZD9496 can radiosensitize ER+ breast tumors, and these agents with RT may be more effective for radiosensitization. This work also supports further clinical investigation of the timing of RT for patients receiving ET, including using ET during RT, especially as initiating ET prior to RT has been increasingly utilized as a bridging therapy followed by concurrent ET+RT during the COVID-19 pandemic.

Citation Format: Anna R. Michmerhuizen, Lynn M. Lerner, Andrea M. Pesch, Connor Ward, Rachel Schwartz, Kari Wilder-Romans, Meilan Liu, Charles Nino, Kassidy Jungles, Ruth Azaria, Alexa Jelley, Nicole Zambrana Garcia, Alexis Harold, Amanda Zhang, Bryan Wharram, Daniel F. Hayes, James M. Rae, Lori J. Pierce, Corey W. Speers. Inhibition of estrogen receptor signaling as a strategy for radiosensitization of ER+ breast cancers [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 2697.

Abstract 1943: Radiosensitization of PIK3CA wild type triple negative breast cancers with Bcl-family inhibition

Purpose: Compared to other breast cancer subtypes, triple negative breast cancers (TNBC) derive the least benefit from adjuvant radiation (RT) which contributes to higher rates of locoregional recurrence. Thus, there is a critical need to identify clinical strategies to increase the effectiveness of RT therapy in TNBC.

Methods: Alamar blue proliferation assays were used to calculate half maximal inhibitory concentration (IC50) values for each Bcl-2 family inhibitor 72 hours after drug treatment. Clonogenic survival assays were used to evaluate radiosensitivity and to calculate the radiation enhancement ratio (rER) after combination treatment. Apoptosis was assessed through formation of cleaved PARP and annexin V/PI-based flow cytometry. Xenograft models with MDA-MB-231 cells and TNBC patient-derived xenografts (PDX4664) were used to assess radiosensitization in vivo.

Results: A novel radiosensitizer screen identified Bcl-2 family inhibition as a potentially effective treatment strategy in radioresistant breast cancer models. Single-agent response to pan Bcl-2 family inhibition (ABT-263) or Bcl-xL inhibition (WEHI-539, A-1331852) was more effective in PIK3CA wild type (wt) TNBC (IC50 < 1µM) compared to PIK3CA mutant TNBC. Inhibition of apoptosis with ABT-263 led to radiosensitization of PIK3CA/PTEN wild-type TNBC cell lines (rER: 1.09-1.74), but had no effect on PIK3CA/PTEN mutant TNBC (rER: 0.87-1.18). Radiosensitization was observed to be Bcl-xL-dependent, with Bcl-xL inhibitor-specific radiosensitization (rER: 1.12-2.38) but a lack of Bcl-2 inhibitor (ABT-199, rER: 0.94 - 1.21) or MCL-1 inhibitor-mediated radiosensitization (S63845, rER: 0.91 - 1.06). In PIK3CA wt TNBC, combination treatment of Bcl-2 family inhibition and RT significantly increased the percent of apoptotic cells (p < 0.001) and led to increased formation of cleaved PARP 48 hours after RT. Sensitivity to RT was dependent on expression of MCL-1, an anti-apoptotic protein that is overexpressed in PIK3CA/PTEN mutant TNBC. Overexpression of MCL-1 in PIK3CA/PTEN wild type TNBC rescued radioresistance (rER: 0.99-1.09), whereas co-inhibition of MCL-1 and Bcl-xL in PIK3CA/PTEN mutant TNBC was sufficient to overcome radioresistance (rER: 2.32 - 2.35). In vivo, nonspecific Bcl-2 family inhibition or specific Bcl-xL inhibition in combination with RT decreased tumor growth and increased time to tumor tripling (p < 0.0001) in PIK3CA wt models of TNBC.

Conclusions: In this study, we demonstrated that inhibition of Bcl-2 family proteins in combination with RT led to increased levels of apoptosis and cell death in PIK3CA/PTEN wt - but not PIK3CA/PTEN mutant - TNBC and we identified MCL-1 as a critical mediator of this radiosensitIvity. Together, these results indicate that Bcl-xL inhibition may be a feasible clinical strategy for the radiosensitization of PIK3CA/PTEN wild-type TNBC.

Citation Format: Andrea M. Pesch, Benjamin C. Chandler, Anna R. Michmerhuizen, Nicole Hirsh, Kari Wilder-Romans, Meilan Liu, Tanner Ward, Dana Messinger, Charles Nino, Cassandra Ritter, James M. Rae, Corey W. Speers. Radiosensitization of PIK3CA wild type triple negative breast cancers with Bcl-family inhibition [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 1943.

Interactions of anti-sigma factor antagonists of Mycobacterium tuberculosis in the yeast two-hybrid system

Anti-sigma factor antagonists (anti-anti-sigma factors) play critical roles in regulating the expression of alternative sigma factors in response to specific stress signals. The Clusters of Orthologous Groups (COG) database has identified the existence of six genes, Rv0516c, Rv1364c, Rv1365c, Rv1904, Rv2638 and Rv3687c (grouped under the cluster COG1366), encoding potential anti-sigma factor antagonists in Mycobacterium tuberculosis. These molecules are speculated to regulate the expression of sigma factor SigF of M. tuberculosis in response to stress signals. Since signaling occurs via physical interactions of proteins (protein-protein interaction), we investigated whether the anti-sigma factor antagonists of M. tuberculosis interact with anti-sigma factor RsbW (Rv3287c) or the sigma factor SigF (Rv3286c) in the yeast two-hybrid system. The results revealed that most of the anti-sigma factor antagonists interact with either RsbW or SigF or both. In addition, some anti-sigma factor antagonists also displayed limited interactions between themselves. These interactions suggest that they possibly transduce some signals to SigF and between themselves.

Scotopic threshold response (STR) of the human electroretinogram

We recorded the human ERG using full-field stimuli at light intensities near absolute threshold to examine characteristics of the scotopic threshold response (STR). The human STR was seen below PII threshold and was the only component of the ERG evident near absolute rod psychophysical threshold. The human STR was detectable in the corneal ERG at stimulus intensities 0.6-1.0 log units above psychophysical threshold and had about a 2 log unit range to apparent saturation. Maximum STR amplitude was 12-20 microV. The STR latency ranged from 100-185 msec, depending on stimulus duration and intensity. The STR returned to baseline by 300 msec after onset, for very brief flashes, but it was prolonged with longer flashes. The spectral characteristics of the human STR matched rods and not cones. The STR followed Bloch's law and exhibited temporal integration for at least 80 msec. At 2.5-3 log units above visual threshold, corneal positive PII (b-wave and d.c. component) progressively obscured the negative STR. We propose that the human STR reflects post receptor processing in the retina. This is based on the similarity of the human STR to the STR of the cat and monkey, both of which originate at postreceptoral sites. Thus the human STR may find clinical use to evaluate the rod pathway in the proximal retina.