Preclinical Evaluation of the ATR Inhibitor BAY 1895344 as a Radiosensitizer for Head and Neck Squamous Cell Carcinoma

Abstract

Purpose: Despite aggressive multimodal treatment that typically includes definitive or adjuvant radiation therapy (RT), locoregional recurrence rates approach 50% for patients with locally advanced human papillomavirus (HPV)-negative head and neck squamous cell carcinoma (HNSCC). Thus, more effective therapeutics are needed to improve patient outcomes. We evaluated the radiosensitizing effects of ataxia telangiectasia and RAD3-related (ATR) inhibitor (ATRi) BAY 1895344 in preclinical models of HNSCC.

Methods and materials: Murine and human HPV-negative HNSCC cells (MOC2, MOC1, JHU-012) were treated with vehicle or ATRi with or without 4 Gy. Checkpoint kinase 1 phosphorylation and DNA damage (γH2AX) were evaluated by Western blot, and ATRi half-maximal inhibitory concentration was determined by MTT assay for HNSCC cells and immortalized murine oral keratinocytes. In vitro radiosensitization was tested by clonogenic assay. Cell cycle distribution and mitotic catastrophe were evaluated by flow cytometry. Mitotic aberrations were quantified by fluorescent microscopy. Tumor growth delay and survival were assessed in mice bearing MOC2 or JHU-012 transplant tumors treated with vehicle, ATRi, RT (10 Gy × 1 or 8 Gy × 3), or combined ATRi + RT.

Results: ATRi caused dose-dependent reduction in checkpoint kinase 1 phosphorylation at 1 hour post-RT (4 Gy) and dose-dependent increase in γH2AX at 18 hours post-RT. Addition of RT to ATRi led to decreased BAY 1895344 half-maximal inhibitory concentration in HNSCC cell lines but not in normal tissue surrogate immortalized murine oral keratinocytes. Clonogenic assays demonstrated radiosensitization in the HNSCC cell lines. ATRi abrogated the RT-induced G2/M checkpoint, leading to mitosis with unrepaired DNA damage and increased mitotic aberrations (multinucleated cells, micronuclei, nuclear buds, nucleoplasmic bridges). ATRi and RT significantly delayed tumor growth in MOC2 and JHU-012 in vivo models, with improved overall survival in the MOC2 model.

Conclusions: These findings demonstrated that BAY 1895344 increased in vitro and in vivo radiosensitivity in HPV-negative HNSCC preclinical models, suggesting therapeutic potential warranting evaluation in clinical trials for patients with locally advanced or recurrent HPV-negative HNSCC.

Figure 1:. BAY 1895344 inhibits Chk1 phosphorylation (pChk1) in a dose- and time-dependent manner and causes dose-dependent increase in DNA damage (γH2AX) at 18h post-RT.

(A) MOC2 (top), MOC1 (middle), and JHU-012 (bottom) cells were preincubated with vehicle or 0.2μm BAY 1895344 for 30 min before sham RT or preincubated with vehicle or serial BAY 1895344 dilutions (0.002–2μm) for 30 min before RT (4Gy). Cell lysates harvested at 1h post-RT/sham RT were probed for pChk1 and GAPDH (loading control), and lysates harvested at 18h post-RT/sham RT were probed for γH2AX and GAPDH. (B) MOC2 (top), MOC1 (middle), and JHU-012 (bottom) cells were treated with vehicle or 0.2μm BAY 1895344 for 30 min before sham RT or RT. Cell lysates were harvested after sham RT (baseline), 1h post-RT, or 6h post-RT and probed for pChk1 and GAPDH.

Figure 2:. RT lowers the IC50 of BAY 1895344, and BAY 1895344 radiosensitizes HNSCC cells.

(A) HNSCC or IMOK cells were treated with vehicle or serial dilutions of BAY 1895344 (ATRi) for 30min before sham RT (blue circle) or RT (4Gy, red square). After 72h, metabolic activity was evaluated by MTT assay. IC50s between ATRi and ATRi + RT were compared for each cell line by unpaired t-test (ns, not significant, *p<0.05, **p<0.01). Clonogenic assays were performed on MOC2 (B), MOC1 (C), and JHU-012 (D) cells pre-incubated with vehicle (RT, blue circle) or 100nM BAY 1895344 (ATRi + RT, red square). Vehicle/drug-containing media was replaced with fresh media after 4h, given that clones would not form with longer BAY 1895344 treatment duration. After 7–10 days, fixed cells were stained with Coomassie blue and colonies counted. Percent surviving fraction for each RT dose is shown, with RT compared to ATRi + RT by 2-way ANOVA.

Figure 3:. BAY 1895344 abrogates RT-induced G2/M arrest, promoting mitotic catastrophe.

Flow cytometric cell cycle analysis was performed for MOC2 (A), MOC1 (B), and JHU-012 (C) cells treated with four experimental conditions: Control (30 min pre-incubation with vehicle before Sham RT), ATRi (30min pre-incubation with 100nM BAY 1895344 before Sham RT), RT (4Gy delivered 30min after vehicle) and ATRi + RT (4Gy delivered 30 min after 100nM BAY 1895344). Cells were labeled with EdU for 45 min and harvested at 6h post-RT (or sham RT), and cells stained with propidium iodide (PI) followed by flow cytometric analysis of cell cycle distribution (G0/G1 phase, blue; S phase, red checkered; G2/M phase, green mini-checkered). Percentages of cells in each phase were compared among treatment groups by 2-way ANOVA with Tukey’s post-hoc test, with statistically significant differences indicated by color-coded asterisk(s) corresponding with cell cycle phase (*p<0.05, **p<0.01, ****p<0.0001). MOC2 (D) and JHU-012 (E) cells treated per above were incubated with EdU for 45 min, harvested at 6h post-RT (or sham RT), and cells stained with PI and antibodies against γH2Ax and phospho-Histone H3 (pHH3). The percentage of cells in G2/M phase staining positively for both γH2AX and pHH3 was quantified, indicating mitotic catastrophe. The percentages of G2/M phase cells costaining for γH2AX and pHH3 were compared across groups (Control, blue circle; ATRi, red square; RT, green triangle; ATRi + RT, purple upside-down triangle) by 1-way ANOVA with Tukey’s post-hoc test (*p<0.05, **p<0.01, ***p<0.001).

Figure 4:. BAY 1895344 combined with RT leads to increased mitotic aberrations.

Fluorescent microscopy was performed on MOC2 cells treated with four experimental conditions: Control (30 min pre-incubation with vehicle before Sham RT; blue circle), ATRi (30min pre-incubation with 100nM BAY 1895344 before Sham RT; red square), RT (4Gy delivered 30min after vehicle; green triangle) and ATRi + RT (4Gy delivered 30 min after 100nM BAY 1895344; purple upside-down triangle). Cells were fixed at 1, 6, or 24h post-RT or sham RT and stained with DAPI. Slides were imaged at 63x (≥10 fields of view/condition), and mitotic aberrations counted. (A) Multinucleated cell (blue arrow), micronuclei (red arrow), nuclear bud (green arrow), and nucleoplasmic bridge (purple arrow) examples are shown. Mitotic aberrations per cell were evaluated at (B) 1h, (C) 6h, and (D) 24h post-treatment. For each mitotic aberration, counts per cell were compared across treatment groups by 1-way ANOVA with Tukey’s post-hoc test (*p<0.05, **p<0.01, ***p<0.001, ****p<0.0001).

Figure 5:. BAY 1895344 sensitizes MOC2 and JHU-012 tumors to RT in vivo.

(A) The schematic demonstrates treatment regimens for vehicle or BAY 1895344 in single-fraction and multi-fraction RT experiments in heterotopic MOC2 and JHU-012 models (created with BioRender.com). When hind limb IM tumors reached >45mm³ (Day 0), mice were randomized into four treatment groups: Control (vehicle and Sham RT; blue circle for growth delay; blue dotted-dashed line for survival), ATRi (BAY 1895344 and Sham RT; red square for growth delay; red solid line for survival), RT (vehicle + 10Gy or 8Gy x 3; green triangle for growth delay; green dashed line for survival), and ATRi + RT (BAY 1895344 + 10Gy or 8Gy x 3; purple upside-down triangle for growth delay; purples dotted line for survival). For the single-fraction study, mice received vehicle or 50mg/kg BAY 1895344 for 3 days on/4 days off for three weeks with Sham RT or a single 10Gy fraction of RT after the first two doses. For the first MOC2 multi-fraction study, mice received 50mg/kg BAY 1895344 (3 doses surrounding each RT fraction) with sham RT/8Gy RT for 3 fractions with 3d between each fraction. This was followed by a 72h break from therapy, then another 3 days of twice daily 50mg/kg BAY 1895344. For the other multi-fraction studies (MOC2 and JHU-012), mice received vehicle or 40mg/kg BAY 1895344 (3 doses surrounding each fraction) with sham RT or 8Gy for 3 fractions with 4d between RT fractions (9 total doses of vehicle or ATRi). (B) Relative tumor volumes (mean +/- SD) normalized to Day 0 are shown for MOC2 tumors in the single-fraction RT study. Tumor growth delays across groups were compared by 2-way ANOVA with Tukey’s post-hoc test (ns=not significant, *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001). (C) Kaplan-Meier survival curves are shown for mice with MOC2 tumors in the single-fraction RT study, with pairwise group comparisons evaluated by Log-Rank test. (D, F, H) In the multi-fraction RT regimen, relative tumor volumes (mean +/- SD) normalized to Day 0 are shown for MOC2 (D and F) and JHU-012 tumors (H). Tumor growth delays across groups were compared by 2-way ANOVA with Tukey’s post-hoc test (ns=not significant, ****p<0.0001). (E, G, I) Kaplan-Meier survival curves are shown for mice with MOC2 (E and G) and JHU-012 tumors (I). Tick marks indicate mice censored at the time of euthanasia for issues unrelated to tumor growth or treatment toxicity. Pairwise group comparisons were evaluated by Log-Rank test (ns=not significant, *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001).