AZD1775 in combination with chemotherapy and targeted agents in TNBC PDX models
In an effort to minimize the number of animals required to evaluate multiple combinations with AZD1775, we performed an initial screen using 4 TNBC PDX models treated with AZD1775, capecitabine, paclitaxel, gemcitabine, doxorubicin, navitoclax (BCL-2 inhibitor), VX970 (ATR inhibitor), and romidepsin (HDAC inhibitor) or combinations of these agents with AZD1775. Each group consisted of 3 mice with 2 tumors each. As demonstrated in Figure 1, we observed an increased TGI with the combination of multiple agents with AZD1775 compared to any single agent alone. In particular, the combination of doxorubicin or paclitaxel and AZD1775 showed an enhanced combination effect in one PDX model each (TNBC009 and TNBC013, respectively). In addition, the combination of gemcitabine and AZD1775 was found to have a better combination effect in two out of the four models (TNBC0002 and TNBC013). However, due to this combination being actively researched in several clinical trials, we decided to not pursue this combination.
The combination of AZD1775 and capecitabine was identified for further evaluation based on enhanced combination effect observed in 2 models (TNBC002, TNBC012) and the fact that both drugs cross the blood/brain barrier. Of note, in TNBC013, single agent capecitabine resulted in a very high TGI of approximately 75% which may have limited the detection of a combination effect.
AZD1775 in combination with capecitabine in PDX models
To confirm potentiation of the activity of AZD1775 with the addition of capecitabine in TNBC, we performed further in vivo studies using 2 TNBC PDX models (TNBC012 and TNBC013) with 5 animals (10 tumors) in each group. These models were selected for confirmation based on known p53 mutations and these tumors were isolated from brain metastasis in patients, which is relevant in TNBC given the high incidence of brain metastasis and CNS penetration of both agents. The dose of capecitabine was lowered in these experiments for TNBC013 based on the TGI observed with a higher dose in Figure 1. As depicted in Figure 2A-B, combination treatment resulted in a statistically significant tumor growth inhibition when compared with either single agent and tumor regression was observed in TNBC013 (Figure 2B).
Antiproliferative effects of AZD1775 with 5FU in TNBC cell lines in vitro
Following confirmation of combination activity in vivo, we performed in vitro experiments to further characterize the antiproliferative activity of the combination using live cell imaging and the SRB assay. We selected 5FU for use in vitro based on the required activation steps for the conversion of capecitabine to 5FU in vivo. We observed a statistically significant decrease in proliferation with the combination as compared to either AZD1775 and 5FU alone as assessed by live cell imaging in two of the four TNBC cell lines (MDA-MB-231 and CAL-51) (Figure 3A, D). In contrast, the HCC1937 only demonstrated an enhanced combination effect when compared to single agent 5-FU and no combination effects were observed in the MDA-MB-468 TNBC cell line (Figure 3 B, C). In the SRB assay, quantification of cellular proteins in cultured cells can be measured and synergistic responses can be calculated using the Chou and Talalay method. Using this assay, we observed synergistic combination effects in the MDA-MB-468, HCC1937, and CAL-51 cell lines at various concentrations or both drugs (* = CI < 1) (Figure 4).
Apoptotic effects of AZD1775 with 5-FU in TNBC cell lines
To determine the effect of AZD1775 and 5FU on apoptosis, the Caspase 3/7 assay was utilized. At 24 hours, we observed no significant apoptosis in the MDA-MB-231 or HCC1937 cell lines with single agent or combination treatment (Figure 5 A, C). However, there was a significant increase in apoptosis with the combination compared to both single agents in the MDA-MB-468 cell line (Figure 5B). In the CAL-51 TNBC cell line, apoptosis was only significantly higher in the combination compared to single agent 5FU (Figure 5D).
Cell Cycle effects of AZD1775 with 5FU in TNBC cell lines
To determine the effects on the cell cycle, cells were exposed to AZD1775, 5FU or the combination for 24 hours and analyzed by flow cytometry. As depicted in Figure 6A and B, there is no significant difference in cell cycle arrest in the MDA-MB-231 or MDA-MB-468 with any treatment. However, we did observe a slight decrease in G2/M in the combination compared to either single agent. In the HCC1937 cell line, there was a significant increase in S-phase arrest in the 5-FU single agent when compared to no drug (p<0.01). Additionally, there was also a statistically significant increase in S-phase arrest in the combination, when compared to the single agent AZD1775 (p<0.05) (Figure 6C). Similarly, the p53 WT cell line, CAL-51, demonstrated similar results as the HCC1937 with a statistically significant increase in S-phase arrest in the 5FU single agent compared to no drug and the combination compared to AZD1775 single agent (p<0.001 and p<0.05, respectively) (Figure 6D).
Effects of AZD1775 with 5FU on downstream effectors
Immunoblotting was performed to elucidate the mechanism of the combination effects of AZD1775 with 5FU in the four TNBC cell lines. Following 48-hour exposure to the drugs, two of the four cell lines (MDA-MB-468 and HCC1937) showed a decrease in p-CDC2 in a dose dependent manner in the single agent treatment of AZD1775, and the decrease was only maintained in the combinations for the HCC1937 cell line. Additionally, an increase in γ-H2AX was observed in the MDA-MB-231 and HCC1937 in a dose dependent manner with single agent AZD1775 that was maintained in the combination treatment, indicating a DNA damage response. In the other two TNBC cell lines, an increase in γ-H2AX was only observed in the 250 nM dose of AZD1775. A decrease in Bcl-xL was observed with the 5-FU and combination treatment in the MDA-MB-468 and CAL-51 cell lines indicating an increase in apoptosis (Figure 7).