Sabizabulin, a Potent Orally Bioavailable Colchicine Binding Site Agent, Suppresses HER2+ Breast Cancer and Metastasis

Simple Summary The sabizabulin agent represses tumor cell growth, cell migration, and colony formation, and induces cell death in HER2+ breast cancer models. Sabizabulin is comparable to paclitaxel to suppress HER2+ xenograft growth and to inhibit lung metastasis in a HER2+ patient-derived xenograft (PDX) model. Sabizabulin is a promising orally available agent as an alternative to taxanes to target tubulin in breast cancer patients, including the HER2+ breast cancer molecular subtype. Abstract HER2+ breast cancer accounts for 15% of all breast cancer cases. Current frontline therapy for HER2+ metastatic breast cancer relies on targeted antibodies, trastuzumab and pertuzumab, combined with microtubule inhibitors in the taxane class (paclitaxel or docetaxel). It is well known that the clinical efficacy of taxanes is limited by the development of chemoresistance and hematological and neurotoxicities. The colchicine-binding site inhibitors (CBSIs) are a class of promising alternative agents to taxane therapy. Sabizabulin (formerly known as VERU-111) is a potent CBSI that overcomes P-gp-mediated taxane resistance, is orally bioavailable, and inhibits tumor growth and distant metastasis in triple negative breast cancer (TNBC). Herein, we demonstrate the efficacy of sabizabulin in HER2+ breast cancer. In vitro, sabizabulin inhibits the proliferation of HER2+ breast cancer cell lines with low nanomolar IC50 values, inhibits clonogenicity, and induces apoptosis in a concentration-dependent manner. In vivo, sabizabulin inhibits breast tumor growth in the BT474 (ER+/PR+/HER2+) xenograft model and a HER2+ (ER-/PR-) metastatic patient-derived xenograft (PDX) model, HCI-12. We demonstrate that sabizabulin is a promising alternative agent to target tubulin in HER2+ breast cancer with similar anti-metastatic efficacy to paclitaxel, but with the advantage of oral bioavailability and lower toxicity than taxanes.


Protein Extraction and Western Blotting
Cells grown to 80% confluence were incubated with increasing concentrations of sabizabulin (10, 20, 50, 100 nM) or 100 nM colchicine or paclitaxel for 24 h. For timedependent studies, cells were exposed to 100 nM of sabizabulin for 12, 24, 48, or 72 h. Flash-frozen tumors were ground under liquid nitrogen using a mortar and pestle and homogenized. Proteins were extracted in RIPA lysis buffer as in [39] and quantitated by a Bradford assay.

Histology and Immunohistochemistry (IHC) Analysis
Paraffin-embedded specimens were sectioned (5-7 µm) onto positively charged slides at the UTHSC Research Histology Core. For tumor slides, antigens were retrieved in 1× citrate buffer (pH 6.0) in a pressure cooker. After blocking endogenous peroxidase (3% H 2 O 2 /MeOH for 20 min), all slides were blocked in 10% normal serum/PBST for at least 1 h at RT followed by incubation with primary antibody: rabbit anti-Ki67 at 1:800 (ab15580, Abcam, Cambridge, UK), or rabbit anti-cleaved caspase-3 (9661S @1:200, Cell Signaling Technology) overnight in humid chambers at 4 • C. Slides were developed as described in [39] and imaged with a Keyence BZ-X700 microscope (Keyence, Itasca, IL, USA). Quantification of Ki67 was performed as in [39] and cleaved caspase-3 quantitated as the area of positive cells in 6 representative fields per tissue section (@600× magnification) by a blinded observer. Lung metastasis was evaluated following IHC with an anti-mitochondrial antibody (ab92824, Abcam, Cambridge, UK, 1:1000) and the metastatic burden was quantified by a percent positive area algorithm as described in [51]. 50 Testing and Isobologram Assays to Determine a Drug Combination Index (CI) BT474, SKBR3 or AU565 (lapatinib-sensitive) and JIMT (lapatinib-resistant) cells were seeded overnight into 96-well plates at a density of either 7500 cells/well (BT474), 5000 cells/well (SKBR3), 15,000 cells/well (AU565) or 8000 cells/well (JIMT) and treated the next day with fresh media containing increasing concentrations of lapatinib alone or, for isobole testing, lapatinib was added in combination with sabizabulin or paclitaxel and cells treated for up to 96 h. The dose range of lapatinib was as follows per cell line: 1 nM to 600 nM for SKBR3 and BT474; 12.5 nM to 10 µM for AU565; 1 nM to 20 µM for JIMT. At the study endpoint, growth inhibition was measured by the MTS assay; drug response was first normalized to untreated vehicle controls and/or to initial seeding density, plotted on a log scale, and then plotted using non-linear regression best fit analysis in GraphPad Prism 9.0. For drug combination studies, the mean IC 30 , IC 50 and IC 70 values ± SEM for paclitaxel and sabizabulin were first derived from at least 3 biological replicates. Isobole testing was conducted using the IC 30 value of sabizabulin or paclitaxel, which was held constant in the presence of increasing concentrations of lapatinib. The CI index was calculated per methods in [52]. Values < 1.0 indicate synergistic drug action, while values of approximating 1.0 are indicative of an additive effect, whereas values > 1.0 suggest antagonism.

Scratch Assay in Response to Drug Combinations
JIMT cells were plated at a density of 38,000 cells/well into a 96-well ImageLock plate (Essen Biosciences) to achieve confluence within 24 h. The following day, the WoundMaker tool was used to scratch the cell monolayer. The cells were washed twice with growth media and then fresh growth media (vehicle) or growth media containing drug were added to the cells in the presence of Cytotox Green (4663, Essen Bioscience) reagent to measure cell viability throughout the wound healing process. Drug doses were based on isobole testing and were selected to be minimally toxic for the duration of the assay time. The plate was imaged every four hours to calculate the wound width. The wound size was normalized to the initial wound width (t = 0 h) and then normalized relative to the untreated, vehicle control.

Statistical Analysis
Unless otherwise stated, all raw data were entered into Prism 9.0 and analyzed using one-way or two-way ANOVA followed by multiple pairwise comparison tests (t-tests). Unless otherwise indicated, on all graphs, the mean ± SEM is shown. Significance was determined at a 95% confidence threshold, and p-values indicated by asterisks defined as: * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001.

Anti-Proliferative Effect of Sabizabulin on HER2+ Cells In Vitro
We have previously shown that sabizabulin ( Figure 1A) has nanomolar potency in TNBC, with IC 50 values between 8-9 nM [39]. To compare HER2 levels in cell line models used herein, western blotting was performed. AU565 cells served as a positive control and MDA-MB-231 TNBC cells as a negative control. All models that were selected for evaluation of sabizabulin efficacy express HER2 ( Figure 1B). To optimize downstream assays for evaluating growth inhibition, we first characterized each cell line's doubling time, which was 84 h for BT474  Figure 1C). In both BT474 and SKBR3 cells, growth inhibition by sabizabulin was observed in the low nanomolar range ( Figure 1D and Table 1), with a similar potency observed for both HER2 + cell lines.

Anti-Proliferative Effect of Sabizabulin on HER2+ Cells In Vitro
We have previously shown that sabizabulin ( Figure 1A) has nanomolar potency in TNBC, with IC50 values between 8-9 nM [39]. To compare HER2 levels in cell line models used herein, western blotting was performed. AU565 cells served as a positive control and MDA-MB-231 TNBC cells as a negative control. All models that were selected for evaluation of sabizabulin efficacy express HER2 ( Figure 1B). To optimize downstream assays for evaluating growth inhibition, we first characterized each cell line's doubling time, which was 84 h for BT474 and 40 h for SKBR3 cells, respectively ( Figure 1C). In both BT474 and SKBR3 cells, growth inhibition by sabizabulin was observed in the low nanomolar range ( Figure 1D and Table 1), with a similar potency observed for both HER2 + cell lines.

Sabizabulin Induces Concentration-Dependent Growth Inhibition with Minimal Cytotoxicity
BT474 and SKBR3 cells were exposed to sabizabulin or paclitaxel at increasing concentrations and imaged over time for changes in confluence by phase contrast ( Figure 1E,F) while in the presence of Cytotox Green to simultaneously enumerate dead cells ( Figure 1G,H). By 72 h, sabizabulin and paclitaxel had significantly inhibited the growth of BT474 and SKBR3 cells in a concentration-dependent manner ( Figure 1E,F). Growth inhibition was similar at equivalent concentrations of paclitaxel and sabizabulin for SKBR3 cells ( Figure 1F). In contrast, for BT474 cells, sabizabulin was less effective to inhibit growth relative to the same concentrations of paclitaxel. Drug-induced cytotoxicity was significant in BT474 cells at 12 nM or 16 nM ( Figure 1G). In contrast, in SKBR3 cells, paclitaxel was cytotoxic beginning at 8 nM, whereas sabizabulin was cytotoxic beginning at 12 nM ( Figure 1H).

Sabizabulin Inhibits Colony Formation in HER2+ Cells
BT474 and SKBR3 cells were treated with increasing concentrations of colchicine, paclitaxel, or sabizabulin in a colony forming assay. The phase confluence of vehicle-treated cells at the assay endpoint was set as the maximum colony formation efficiency (100%) and used to calculate the percentage of suppression in drug-treated cells. At a 4 nM concentration, sabizabulin was less effective than paclitaxel in BT474 and SKBR3 cells, whereas at 8 nM and 16 nM concentrations, sabizabulin inhibited clonogenicity relative to vehicle control with similar efficacy to paclitaxel (Figure 2A-D). Overall, sabizabulin significantly inhibited HER2+ breast cancer clonogenicity with higher potency than previously observed for TNBC cell lines [39].

Sabizabulin Inhibits BT474 Primary Tumor Growth
The anti-tumor efficacy of sabizabulin relative to paclitaxel was first assessed in the BT474 orthotopic xenograft model. Sabizabulin (17 mg/kg, PO) significantly inhibited

Sabizabulin Inhibits BT474 Primary Tumor Growth
The anti-tumor efficacy of sabizabulin relative to paclitaxel was first assessed in the BT474 orthotopic xenograft model. Sabizabulin (17 mg/kg, PO) significantly inhibited tumor growth relative to vehicle by day 11 of therapy and relative to paclitaxel (10 mg/kg, IP) on day 14 of therapy, with inhibition maintained until the study endpoint ( Figure 4A). Sabizabulin significantly reduced end-stage tumor volume and tumor wet weight relative to vehicle-treated animals and was more effective than paclitaxel therapy ( Figure 4B,C, Supplemental Figure S1). Minimal toxicity from dosing regimens was evidenced by the increase in animal body weight throughout the study across cohorts ( Figure 4D). Images of excised tumors are shown in Figure 4E.
tumor growth relative to vehicle by day 11 of therapy and relative to paclitaxel (10 mg/kg, IP) on day 14 of therapy, with inhibition maintained until the study endpoint ( Figure 4A). Sabizabulin significantly reduced end-stage tumor volume and tumor wet weight relative to vehicle-treated animals and was more effective than paclitaxel therapy ( Figure 4B,C, Supplemental Figure S1). Minimal toxicity from dosing regimens was evidenced by the increase in animal body weight throughout the study across cohorts ( Figure 4D). Images of excised tumors are shown in Figure 4E. To confirm that sabizabulin induces tumor apoptosis in vivo, sections were immunostained with cleaved caspase-3 antibodies. Significantly more cleaved caspase-3 staining was detected with sabizabulin treatment relative to vehicle treatment. In contrast, there was no significant difference observed between vehicle-treated and paclitaxel-treated tumors, or between paclitaxel-and sabizabulin-treated tumors ( Figure 4F,G). Cleaved PARP in tumors was compared by western blotting ( Figure 4H). Sabizabulin-and paclitaxel-treated tumors show increased cleaved PARP signal relative to the vehicle control, as expected.

Sabizabulin Suppresses Tumor Growth and Inhibits Lung Metastases in the HCI-12 HER2+ PDX Model
In general, PDX models replicate clinical metastatic patterns in mice with high genetic similarity to the original patient material, including models derived from the HCI preclinical resource [50,53,54]. HCI-12 was selected for evaluation since, unlike BT474 cells, it is ER-/PR-and HER2+ and since it robustly metastasizes to the lungs [50].
Although HCI-12 PDX is derived from a patient treated with taxanes, its response to paclitaxel in NSG mice was uncharacterized. In contrast to in vitro studies using conventional HER2+ cell lines, in the HCI-12 PDX, paclitaxel was significantly more effective than sabizabulin to repress tumor growth over time ( Figure 5A) and at study endpoint ( Figure 5B,C), suggesting that HCI-12 PDX model is not paclitaxel-refractory. Whereas paclitaxel completely inhibited tumor growth, sabizabulin was less effective, resulting in a 39% decrease in tumor volume relative to vehicle treatment ( Figure 5A). Notably, there was no significant body weight loss at the selected dosing regimens for either drug, suggesting minimal dosing-related toxicity ( Figure 5D). Representative tumor images from vehicle, paclitaxel, and sabizabulin treatment groups are shown in Figure 5E. Evaluation of apoptosis by cleaved caspase-3 immumohistochemsitry revealed no significant differences relative to vehicle-treated tumors for paclitaxel or for sabizabulin, or between paclitaxel and sabizabulin ( Figure 5F,G).
Since apoptosis was not strongly induced in the HCI-12 PDX model in response to either treatment, we next compared cell proliferation by immunostaining slides for Ki67. Whole stained tumors, after excluding necrotic regions, were subjected to digital scanning and densitometry-based quantification for the Ki67+ tumor area ( Figure 6A). Likely because tumors from the vehicle group were highly necrotic, typical of the HCI-12 model, and only healthy tissue was enumerated, the percentage of the positive area for Ki67 was equal to or higher than vehicle group for the treated tumors.
To assay for sabizabulin anti-metastatic activity, lung sections were stained with antihuman specific mitochondria antibodies. On average, in the vehicle group,~49% of lung area was occupied by metastasis ( Figure 6B), whereas in both paclitaxel and sabizabulin cohorts, metastatic burden was < 3.5% of the area of lungs ( Figure 6B). Since over 95% of breast cancer patients die from metastasis, it is important to note that sabizabulin was equally effective at inhibiting lung metastases as paclitaxel, despite being far less effective against primary tumor growth. breast cancer patients die from metastasis, it is important to note that sabizabulin was equally effective at inhibiting lung metastases as paclitaxel, despite being far less effective against primary tumor growth.  weight for vehicle (n = 11 mice), paclitaxel (n = 11 mice), or sabizabulin (n = 12 mice) relative to body weight at the start of therapy (D); the dashed lines indicate initial body weight (y = 0) with a −10% to +10 % variance over time (days on therapy)(A-C) All pairwise comparisons were made to the vehicle control; **** p < 0.001. (E) Representative images of resected tumors. (F) Representative images of IHC slides stained with cleaved caspase-3 (600× magnification; scale bar 50 μM); yellow arrowheads indicate example apoptotic cells. (G) Scatter plot of % positive area for cleaved-caspase-3. N.S. = not significant.  Areas positive for Ki67 or anti-human mitochondrial staining were quantified after scanning whole tumors (vehicle, n = 10 tumors; paclitaxel, n = 11 tumors; sabizabulin, n = 8 tumors) or whole lungs from each animal (n = 10 vehicle, n = 9 paclitaxel and n = 9 sabizabulin), and the staining intensity analyzed by pixel counts using built-in densitometry software algorithms optimized for nuclear (such as Ki67) localization or cytoplasm localization (such as anti-mitochondria) (3D Histech, Ltd. software, Budapest, Hungary). The percent (%) positive area reflects the number of positive tumor epithelial cells (Ki67) or the % area of metastatic burden (detected by anti-human mitochondria) for whole lungs. All p-values are shown relative to the vehicle control group, the mean is indicated above each group; **** p< 0.0001; N.S. is not significant.

Sabizabulin Synergizes with Lapatinib to Inhibit Tumor Cell Growth
Lapatinib, a reversible EGFR/HER2 tyrosine kinase inhibitor, is paired with capecitabine is a second-line therapy for metastatic HER2+ patients [28]; therefore, we next assayed the lapatinib sensitivity in a panel of HER2+ cell lines. We included AU565 cells and JIMT cells, which, relative to BT474 and SKBR3 cells, are reported to be moderately lapatinib sensitive or lapatinib-resistant, respectively [48]. Growth inhibition was measured by MTS assay in the presence of increasing concentrations of lapatinib, confirming these prior observations ( Figure 7A). Next, AU565 ( Figure 7B) and JIMT ( Figure 7C) cells were used in an isobole assay to determine the combination index (CI) when lapatinib is paired with either paclitaxel or sabizabulin at their respective IC 30 concentration (paclitaxel = 2 nM for AU565, or 3 nM for JIMT; sabizabulin = 10 nM for AU565, or 30 nM for JIMT) The CI values for each combination tested are reported in Table 2. Whereas the addition of paclitaxel to lapatinib was neither additive or synergistic in either AU565 or JIMT cells, the addition of sabizabulin to lapatinib produced synergistic effects in each cell line, with the lowest CI values observed for lapatinibresistant JIMT cells. To determine if drug synergy between sabizabulin and lapatinib was also observed during cell migration, a wound healing assay was conducted in JIMT cells with single agents versus combination therapy. The IC 30 concentrations of either sabizabulin (30 nM) or paclitaxel (3 nM) had equivalent activity to repress wound healing relative to the vehicle control (** p < 0.01, either comparison to vehicle). Although there was a trend towards enhanced repression of wound healing with the sabizabulin combination therapy relative to the paclitaxel combination therapy, these data were not statistically significant (p = 0.085) ( Figure 7D). During the same wound healing experiment shown in Figure 7D, there was minimal cell death over the course of the assay as measured by the incorporation of Cytotox Green dye (Supplementary Figure S2).

Discussion
HER2+ breast cancer patients are commonly treated with anti-HER2 immunotherapy in conjunction with a tubulin inhibitor [27]. Taxanes are standard of care agents. While initially successful, they are often subject to efflux pump-mediated chemoresistance, they cause neurotoxicity, and they have low aqueous solubility [36][37][38]. CBSIs are exciting potential replacements for taxanes in breast cancer treatment since they are less likely to be susceptible to taxane resistance mechanisms.
Our group has previously reported the efficacy of a novel CBSI, sabizabulin, in TNBC. Sabizabulin overcomes taxane resistance, is a poor substrate of P-gp mediated drug efflux, is orally bioavailable, and exhibits low levels of toxicity [39,42]. Sabizabulin is highly potent against TNBC both in vitro and in vivo, therefore we determined the efficacy of sabizabulin in HER2+ breast cancer models. The anti-proliferative activity of sabizabulin was determined in two conventional HER2+ breast cancer cell lines, BT474 (ER+/PR+/HER2 + ) and SKBR3 (ER-/PR-/HER2+). The IC 50 values of sabizabulin for HER2+ cells were in the low nanomolar range, like paclitaxel. Sabizabulin effectively inhibited cell growth without gross cytotoxicity, at sub-IC 50 concentrations. Clonogenicity assays confirmed the ability of sabizabulin to inhibit colony formation with similar efficacy to paclitaxel. Sabizabulin was also effective to induce apoptosis in vitro, as validated by induction of PARP and caspase-3 cleavage in a dose-dependent and time-dependent manner. Sabizabulin also induced phosphorylation of BCL2, previously shown to correlate with taxane-induced apoptosis [55][56][57][58].
In vivo, sabizabulin proved effective in inhibiting HER2+ primary tumor growth, demonstrating higher potency than paclitaxel in BT474 xenografts, but with reduced efficacy relative to paclitaxel in the metastatic HCI-12 HER2+ PDX model. Consistent with these observations, the induction of apoptosis by sabizabulin was higher than paclitaxel in the BT474 model. Although there were no significant differences in apoptosis among the cohorts in the HCI-12 PDX, there was a trend that paclitaxel induced apoptosis relative to the vehicle cohort (p = 0.19 for paclitaxel treatment versus p = 0.82 for sabizabulin treatment). Since apoptosis was similar in the HCI-12 PDX, despite differences in tumor volumes, we expected that changes in Ki67 would be observed in response to paclitaxel or sabizabulin treatment; however, no significant changes in Ki67 staining of tumor sections were observed among cohorts. It is possible that since those large regions of HCI-12 tumors are necrotic by study endpoint, in contrast to the BT474 xenografts, and since we excluded necrotic regions during immunohistochemistry analyses, we may have underestimated changes in proliferation or apoptosis as assessed by immunohistochemistry in this model.
Since sabizabulin was less effective to repress primary tumor growth than paclitaxel in the HCI-12 PDX, we expected to observe more lung metastatic lesions in the sabizabulin group relative to paclitaxel. However, we observed that sabizabulin was equally effective as paclitaxel to inhibit lung metastasis, despite animals in the sabizabulin cohort bearing larger tumors, which likely seed more cells into the circulation. Since the BT474 model does not robustly metastasize, we could not use it to assay for anti-metastatic potential. Because metastatic lesions that emerge during chemoresistance are a primary driver of metastatic burden that leads to reduced overall survival, it is notable that sabizabulin has equivalent anti-metastatic efficacy as paclitaxel, with the advantage that sabizabulin also overcomes taxane resistance [39]. These results were surprising; since sabizabulin reduced primary tumor volume by only 40% relative to control tumors.
Preliminary results from our global gene expression and protein target profiling studies of cultured breast cancer cells exposed to paclitaxel versus sabizabulin suggest that sabizabulin has additional molecular targets besides microtubules. Several molecular pathways enriched by sabizabulin appear to be unique to sabizabulin relative to paclitaxelenriched pathways, i.e., sabizabulin shows likely polypharmacology. These studies, which are still in progress, could also explain why we have observed that sabizabulin is synergistic with lapatinib to inhibit cell growth in HER2+ cells, but paclitaxel is not. In contrast to cell growth, for the wound healing assay, the efficacy of combination therapy with lapatinib on the migratory capacity of JIMT cells was equivalent for paclitaxel and sabizabulin. Therefore, the action of each agent is dependent upon the specific biological phenotype assayed. As potential novel molecular targets of sabizabulin besides microtubules are identified, and the putative direct binding partners are then validated by crystallography, the biological activity of any additional sabizabulin targets may provide an enhanced understanding of the mechanisms of sabizabulin action relative to paclitaxel.
Based on the potent anti-tumor efficacy in the BT474 model and the anti-metastatic activity in the HCI-12 PDX, further development of sabizabulin for clinical use in HER2+ patients is warranted. In fact, new derivatives have been recently developed based on the sabizabulin scaffold that have activity against a broad panel of cancer types, including taxane-refractory melanoma [59]. Overall, we have shown that sabizabulin is an effective and well-tolerated agent with activity against four HER2+ breast cancer cell lines, BT474, SKBR3, AU565 and JIMT cells, and in two HER2+ xenograft models, and that sabizabulin, but not paclitaxel, sensitizes AU565 and JIMT cells to lapatinib treatment.

Conclusions
Pre-clinical studies of sabizabulin to treat prostate cancer and TNBC, and now in HER2+ breast cancer, confirm its safety, its low nanomolar potency, its ability to inhibit primary tumor growth, and lung metastasis, and its ability to induce apoptosis, and evade taxane-resistance [39,[42][43][44]. We suggest sabizabulin could be an effective replacement therapy for taxanes, or a well-tolerated second line therapy for patients who acquire taxane resistance. Sabizabulin would also be useful for those patients requiring chemotherapy dose escalation. Side effects of sabizabulin are tolerable; diarrhea and fatigue were observed at a phase II MTD of 63 mg [46]. Sabizabulin could potentially substitute for a taxane in combination with current anti-HER2 therapies, which is an area of future study to determine if there is synergy between sabizabulin (a microtubule destabilizer) with trastuzumab or pertuzumab, as has been observed for paclitaxel (a microtubule stabilizer). Additional future studies of interest include evaluating the efficacy of sabizabulin in acquired taxaneresistant HER2+ breast cancer models, and downstream genetic profiling to discover any conserved mechanisms of taxane resistance that are overcome by sabizabulin in the TNBC and HER2+ breast cancer subtypes.
Sabizabulin is currently under investigation in multiple clinical trials, including a Phase 3 clinical trial for metastatic prostate cancer (NCT04844749) and a Phase 2 trial for metastatic breast cancer (NCT05079360). We will continue to screen for improved sabizabulin analogs that bypass taxane resistance. We will also expand drug development efforts for the conjugation of sabizabulin to tissue-enriched targeting moieties to more effectively and specifically deliver sabizabulin to common sites of breast cancer metastasis in patients that are less frequently observed in pre-clinical rodent models, including the brain, bone, and liver.

Patents
The following patent covered sabizabulin and some of its related compounds: "Compounds for Treating Cancer"; inventors: Duane D. Miller, Wei Li, Jianjun Chen, James T. Dalton, Chien-Ming Li, Sunjoo Ahn. United State Patent 9,029,408 B2, issued on 12 May 2015.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/ 10.3390/cancers14215336/s1. Supplementary Figure S1: Sabizabulin inhibits tumor growth of orthotopic HER2+ BT474 xenografts. Individual tumor measurements for each mouse over time are shown for vehicle (A), paclitaxel (B), and sabizabulin (C) treatment groups during the 33 days on therapy. Supplementary Figure S2: Minimal changes in cell viability are observed during wound healing in JIMT cells. Cytotoxicity was measured by the incorporation of Cytotox Green dye concurrent with the measurements of wound width throughout the same assay shown in Figure 7D. Cytotox Green was normalized at each time point relative to vehicle-treated cells (n = 8 well/treatment/time point), and the mean fold-change ± SEM relative to vehicle reported. No significant changes in cell viability were observed by pairwise t-tests with Welch's correction between any of the groups demarked by the grey bracket. A 2.5 µM dose of paclitaxel (blue line) served as the assay positive control (+control).

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.