Translational exposure-efficacy modeling to optimize the ... · 28/06/2014 · 3 Abstract Aurora A kinase orchestrates multiple key activities allowing cells to transit successfully

1

Translational exposure-efficacy modeling to optimize the dose and schedule of taxanes

combined with the investigational Aurora A kinase inhibitor MLN8237 (alisertib)

Jessica Huck1, Mengkun Zhang1, Jerome Mettetal2, Arijit Chakravarty2, Karthik

Venkatakrishnan3, Xiaofei Zhou3, Rob Kleinfield5, Marc L. Hyer1, Karuppiah Kannan1,

Vaishali Shinde6, Andy Dorner7, Mark Manfredi1, Wen Chyi Shyu2, Jeffrey A. Ecsedy7*

Departments of 1Cancer Pharmacology, 2DMPK, 3Clinical Pharmacology, 4Clinical

Research, 5Drug Development Management, 6Molecular Pathology, 7Translational

Medicine

Takeda Pharmaceuticals International Co.

Cambridge, MA 02139, USA

* For correspondence:

Takeda Pharmaceuticals International Co

35 Landsdowne Street

Cambridge, MA 02139

phone: 617-444-1541

FAX: 617-551-8905

email: [email protected]

Conflict of interest – None

on June 5, 2021. © 2014 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on June 30, 2014; DOI: 10.1158/1535-7163.MCT-14-0027

http://mct.aacrjournals.org/

2

Financial Support – All research was supported by Takeda Pharmaceuticals

International Co

Running Title: Dose schedule optimization with combined MLN8237 and taxanes

Keywords: Aurora, Alisertib, Paclitaxel, Docetaxel, Combination therapy




3

Abstract

Aurora A kinase orchestrates multiple key activities allowing cells to transit successfully

into and through mitosis. MLN8237 (alisertib) is a selective Aurora A inhibitor that is

being evaluated as an anticancer agent in multiple solid tumors and heme-lymphatic

malignancies. The antitumor activity of MLN8237 when combined with docetaxel or

paclitaxel was evaluated in in vivo models of triple negative breast cancer grown in

immunocompromised mice. Additive and synergistic antitumor activity occurred at

multiple doses of MLN8237 and the taxanes. Moreover, significant tumor growth delay

relative to the single agents was achieved after discontinuing treatment; notably,

durable complete responses were observed in some mice. The tumor growth inhibition

data generated with multiple dose levels of MLN8237 and paclitaxel was used to

generate an exposure-efficacy model. Exposures of MLN8237 and paclitaxel achieved

in patients were mapped onto the model after correcting for mouse-to-human variation

in plasma protein binding and maximum tolerated exposures. This allowed rank

ordering of various combination doses of MLN8237 and paclitaxel to predict which pair

would lead to the greatest antitumor activity in clinical studies. The model predicted that

60 and 80 mg/m2 paclitaxel (QW) in patients lead to similar levels of efficacy, consistent

with clinical observations in some cancer indications. The model also supported using

the highest dose of MLN8237 that can be achieved regardless of whether it is combined

with 60 or 80 mg/m2 of paciltaxel. The modeling approaches applied in these studies

can be used to guide dose-schedule optimization for combination therapies using other

therapeutic agents.




4

Introduction

Antimitotics are among the most successful classes of chemotherapy used in cancer

care. This class of agents, including the taxanes, vinca alkaloids and epothilones, is

used to treat diverse solid and hematological malignancies as single agents or as part

of combination regimens. Paclitaxel (brand name Taxol®), a taxane, identified in the

1960s and was first approved for use in metastatic ovarian cancer patients in 1992 and

in 1994 in metastatic breast cancer patients. Paclitaxel binds to β-tubulin and prevents

the disintegration of spindle microtubules during mitosis [1], thereby preventing the

normal assembly/disassembly dynamics necessary for spindle microtubules to

appropriately attach chromosomal kinetochores and subsequently segregate the sister

chromatids to the daughter cells. As a result, cells treated with paclitaxel arrest in

mitosis via the activation of the spindle assembly checkpoint and either undergo

apoptosis directly out of mitosis or exit mitosis without completion of cytokinesis, a

process known as mitotic slippage [2, 3]. In the latter case, these cells can die by

apoptosis, arrest by senescence or reenter the cell cycle by endoreduplication.

Docetaxel (brand name Taxotere®) is a more soluble and potent synthetic derivative of

paclitaxel and was approved for use in breast cancer in 1998. At the cellular level,

docetaxel and paclitaxel share a similar mechanism of action.

In addition to agents that directly perturb microtubule dynamics, anti-cancer

therapies are being developed to directly inhibit enzymes that drive normal mitotic

progression [4]. Among these targets are the Aurora kinases, a family of

serine/threonine kinases that comprises three isoforms, Aurora A, Aurora B and Aurora

C. As Aurora C expression is predominantly limited to germ cells, most attempts at




5

targeting the Aurora kinases have focused on developing selective inhibitors of Aurora

A, Aurora B or both [5, 6].

Aurora A mediates multiple steps throughout mitosis, including centrosome

maturation and separation, mitotic entry, formation of mitotic spindle poles and spindles,

alignment of chromosomes during metaphase and their subsequent separation during

anaphase [7-11]. The outcomes associated with targeted inhibition of Aurora A kinase

have been studied using multiple experimental modalities including RNA interference,

antibody microinjection, targeted knockout in mice, and with use of small molecule

inhibitors [12, 13]; [14-17]. In mitosis, Aurora A inhibition causes abnormal formation of

the mitotic spindles, resulting in mitotic arrest which is mediated by activation of the

spindle assembly checkpoint. The fate of these arrested cells can vary, and includes

apoptosis directly out of mitosis, exit from mitosis without undergoing cytokinesis

resulting in G1 tetraploidy, or completed cytokinesis albeit with severe chromosome

segregation defects. In the latter two outcomes, the abnormal mitotic divisions can lead

to deleterious aneuploidy resulting in cell death or senescence [13, 18].

MLN8237 (alisertib) is a selective ATP competitive inhibitor of Aurora A kinase

[19] studied in a number of Phase 1 and 2 clinical trials as a single agent and in

combination with other therapeutics, including paclitaxel in recurrent ovarian cancer

(NCT01091428) and with docetaxel in prostate and other advanced solid cancers

(NCT01094288). Multiple preclinical studies demonstrated beneficial antitumor activity

in cultured tumor cells and in efficacy studies in vivo when combining Aurora kinase

inhibitors or Aurora kinase targeted RNA interference in a variety of solid and heme-

lymphatic cancer models with paclitaxel and docetaxel [20-30]. Aurora A inhibition




6

using the selective Aurora A kinase inhibitor MLN8054 or RNA interference was shown

to abrogate the spindle assembly checkpoint mediated mitotic delay induced by

paclitaxel [31]. These cells rapidly exited mitosis without completing cytokinesis via

mitotic slippage and enter the G1 portion of the cell cycle with a tetraploid DNA content.

Interestingly, Aurora A overexpression also abrogated the spindle assembly checkpoint

in the presence of microtubule perturbing agents [32, 33].

Here, we demonstrate in preclinical models that the Aurora A kinase inhibitor

MLN8237 significantly enhances the preclinical antitumor activity of docetaxel and

paclitaxel in triple-negative breast cancer models. Triple-negative breast cancers are

characterized as not expressing estrogen receptor, progesterone receptor or HER-2;

therefore these tumors are not susceptible to hormone or HER-2 targeted therapies.

Treatment strategies for triple-negative breast cancer include multiple chemotherapeutic

agents, including taxanes [34]. Though these agents do provide some benefit to

patients, there remains a significant unmet need in this population; therefore alternative

options need to be tested including combination therapy. Here, we build a quantitative

translational exposure-efficacy model using both pre-clinical and clinical data for

MLN8237 and paclitaxel to guide combination dose and schedule strategies for these

agents in patients in order to optimize the potential antitumor activity.




7

Materials and Methods

Tumor cell culture and primary human tumors. MDA-MB-231 cells were obtained

from ATCC (Manassas, VA) and cultured in DMEM medium supplemented with heat

inactivated 10% FBS and 1% L-glutamine (Life Technologies Grand Island, NY). MDA-

MB-231 cells were purchased in 2002 and in-house testing showed them to be free of

mycoplasma and murine pathogens. All experiments were conducted with low passage

cells from recently resuscitated frozen stocks. MDA-MB-231 cells (2x10^6) were

injected orthotopically into the mammary fat pad of NCr nude mice. The primary human

tumor xenografts PHTX-02B and PHTX-14B were developed at Takeda

Pharmaceuticals International Co. from tumors that were originally obtained from breast

cancer patients through the Cooperative Human Tissue Network (Brockville, MD) and

were passed by trocar subcutaneously into the flank of NOD SCID mice.

In vivo efficacy studies. MDA-MB-231, PHTX-02B or PHTX-14B tumor-bearing mice

(n=10 animals per group) were dosed orally (PO) with vehicle (10% HPbCD + 3.5%

NaHC03) or MLN8237 (3, 10, or 20 mg/kg) for 21 days using a once daily schedule

(QD) or for 3 days on / 4 days off over 3 consecutive weeks. Docetaxel (5, 10 mg/kg)

and paclitaxel (5, 10, 15, 20, 30 mg/kg) were administered intravenously (IV) on a once

weekly schedule (QW) for a total of three doses. Tumor growth was measured using

vernier calipers. Tumor growth inhibition (TGI) was determined as the average change

in vehicle treated tumors (ΔVehicle) minus the average change in test agents treated

tumors (ΔTreated) divided by ΔVehicle and expressed as a percentage. Tumor growth

delay (TGD) was the difference in the number of days required for each test agent




8

treatment group to reach an average tumor volume of 1000 mm3 relative to the vehicle

treated group. Drug combinations were assessed for synergy using observed area

under the curve (AUC) values from the efficacy studies over the 21 days of dosing. The

change in AUC relative to the control was calculated for both single agent treatment

groups as well as the combination group. The interaction between the two compounds

was then assessed by comparing the change in AUC observed in the combination

group to the sum of the changes observed in both single agents. The synergy score for

the combination of MLN8237 (M) and either taxane (T) was defined as 100 *

(mean(AUCMT) – mean(AUCM) – mean(AUCT) + mean(AUCvehicle)) / mean(AUCvehicle). A

two sided t-test was used to determine if the synergy score was significantly different

from zero. If the synergy score was less than zero and the P-value was below 0.05,

then the combination was considered to be synergistic. If the synergy score was above

zero and the P-value was below 0.05, then the combination was considered to be sub-

additive or antagonistic. Otherwise the combination was considered to be additive.

Immunohistochemistry. MDA-MB-231 or PHTX-14B tumor bearing NCr female nude

or NOD SCID mice were dosed with MLN8237 at 10 mg/kg or docetaxel at 5mg/kg or

the two agents combined. Tumor tissue was harvested after multiple days of treatment

and fixed in 10% neutral buffered formalin. Tumor sections were stained for Phospho-

Histone H3 (Ser10) (pHisH3) (Millipore, Billerica, MA ) and MPM2 ( Millipore) as

described previously [35]. The number of cells positive for phosphorylation of Histone

H3 on serine 10 (pHistH3) were counted and averaged in five fields of view and DAPI

nuclear staining was used to estimate the total number of cells in the fields. Apoptotic




9

cells in tumor xenograft sections were evaluated by immunohistostaining using Cleaved

Caspase3 (Asp 175) antibody (Cell Signaling Beverley, MA) and were quantified using

Aperio Image analysis software (Leica Microsystems Vista, CA). For histopathological

evaluation, 5 μm sections of formalin fixed, paraffin-embedded tumor samples were

stained with hematoxylin and eosin (HE) using a Leica Autostainer XL (Leica

Biosystems Buffalo Grove, IL). Regions of interest were manually drawn on HE images

using Aperio software to exclude areas of artifacts. Definiens Tissue Studio software

(Definiens Carlsbad, CA) was then used to identify tumor versus non-tumor regions.

Pharmacokinetics. Female Balb/c nude mice bearing the MDA-MB-231 tumor

(approximately 500 mm3) received a single dose of vehicle (10% HPβCD) PO,

MLN8237 PO, docetaxel or paclitaxel IV, or a combination of MLN8237 PO with

docetaxel or paclitaxel IV. Whole blood and tumor samples were collected at specified

time points. Blood samples were collected into tubes containing EDTA and placed on

ice then centrifuged for 5 minutes at 10000 rpm. Plasma was removed into fresh tubes

and stored at -80ºC. Tumors were dissected from the mice, weighed and immediately

frozen on dry ice. Homogenates were prepared in diH2O from frozen tumors using a

FastPrep24 tissue homogenizer. Plasma and tumor homogenates were thawed at

room temperature and the concentration of MLN8237 and paclitaxel or docetaxel in

mouse plasma and tumor samples was determined by HPLC with MS/MS detection.

Pharmacokinetic (PK) analysis was performed in NONMEM (Icon plc, Dublin,

Ireland). Plasma concentrations after a single dose of MLN8237 were fitted to a two-




10

compartment model with absorption, and plasma concentrations after a single dose of

paclitaxel were fitted using a two-compartment model with saturable clearance.

Exposure-efficacy modeling. Total MLN8237 exposures (AUC0-21d) on each dosing

regimen (QD and 3 days on / 4 days off) , and total paclitaxel exposures (AUC0-21d) from

QW dosing, were calculated from simulations of plasma concentration based on the

fitted PK models. The simulated plasma concentration time-courses for paclitaxel after

IV dosing and for MLN8237 after PO dosing are shown in Supplementary Figures 1A

and B respectively, and the pharmacokinetic parameters used are shown in

Supplementary Figure1C. Free fractions in female Balb/c Nude mice of 3.4% and 4.2%

for paclitaxel and MLN8237 respectively were determined using rapid equilibrium

dialysis. Kinetic tumor xenograft volume data were converted to TGI using Equation 1.

The TGI values were fit using Pharsight Phoenix (Certara, St Louis, MO) software with a

combination Emax model defined by Equation 2 where AUCMLN and AUCTax represent the

total cycle free exposures of MLN8237 and paclitaxel, respectively. Supplementary

Table 1 contains a list of the best fit parameters. The resulting best fits are shown in

Supplementary Figure 2.

(eq. 1) %)100(%0,21,

0,21, ⋅−−

=dcontroldcontrol

dtreateddtreated

VVVV

TGI .

(eq. 2) γγγ

TaxTax

TaxTax

MLNMLN

MLNMLNTaxMLN ECAUC

AUCEECAUCAUCE

AUCAUCTGI50

max50

max),(%

++

+=

Clinical exposures were estimated for MLN8237 and paclitaxel as follows. The total

cycle AUC of MLN8237 for doses of 10-50 mg BID administered on Days 1-3, 8-10, 15-

17 of 28-day cycles was calculated based on a previously reported geometric mean




11

apparent oral clearance of 4.45 L/hr in patients with advanced non-hematologic

malignancies [36]. The corresponding unbound plasma exposures were calculated

based on a free fraction of 2.5% in human plasma using rapid equilibrium dialysis. The

total cycle AUC of paclitaxel for doses of 60-90 mg/m2 administered on Days 1, 8, 15 of

28-day cycles was calculated based on a previously reported clearance of 5.5

mL/min/kg (~ 223 mL/min/m2) [37]. The corresponding unbound plasma exposures

were calculated based on a free fraction of 4.4% in human plasma using rapid

equilibrium dialysis. For comparison with unbound preclinical exposures, a scaling was

applied to the unbound clinical exposures factor based on the unbound exposure ratio

between mouse and human as well as on the maximum tolerated dose between mouse

and humans for MLN8237 (20 mg BID mg/kg QD21 days and 50 mg BID for 7 days on

a 21 day schedule, respectively) and paclitaxel (30 mg/kg QW for 3 weeks and 80

mg/m2 QW for 3 weeks on a 28 day schedule, respectively).




12

Results

The antitumor activity of MLN8237 combined with docetaxel was tested in xenograft and

primary human tumor derived triple-negative breast tumor models in

immunocompromised mice. In most mouse strains, the maximum tolerated dose for

MLN8237 is 30 mg/kg dosed daily (QD) or 20 mg/kg dosed twice daily (BID) with an 8

hour break between doses. The maximum tolerated dose for docetaxel was determined

to be 15 mg/kg dosed once weekly (QW) as doses above this led to body weight loss

exceeding 10%. In the MDA-MB-231 xenograft, 3 and 10 mg/kg MLN8237

(QDx21days) combined with 5 and 10 mg/kg docetaxel (QW, days 1, 8 and 15) led to

synergistic antitumor activity (Figure 1A, Table 1). Importantly, MLN8237 at 3 and 10

mg/kg combined with 10 mg/kg docetaxel led to regressions and prolonged tumor

growth delay (difference in days between the control and treated groups to reach 1000

mm3), and in some mice tumors never reformed even after discontinuing treatment. In

comparison, MLN8237 dosed at the maximum tolerated dose in mice bearing the MDA-

MB-231 xenograft did not lead to regressions ([19]). In the primary human tumor

xenograft PHTX-02B, 10 and 20 mg/kg MLN8237 (QDx21days) combined with 5 mg/kg

docetaxel (QW days 1, 8 and 15) led to additive or synergistic antitumor activity (Figure

1B, Table 1). Additive or synergistic antitumor activity also occurred in the PHTX-14B

xenograft with 10 and 20 mg/kg (QDx21days) MLN8237 and 5 and 10 mg/kg (QW days

1, 8 and 15) docetaxel, with sustained tumor regressions occurring only in the

combination regimens (Figure 1C, Table 1). The doses for all treatment regimens

tested were well tolerated as total body weight loss in the mice never exceeded 10%.




13

To ensure that the beneficial antitumor activity observed with the MLN8237 and

docetaxel combination was not due to an increase in the exposure of one drug in the

presence of the other, the pharmacokinetic profile of both drugs was tested in mice

bearing the MDA-MB-231 xenograft after a single dose of 10 mg/kg MLN8237, 5 mg/kg

docetaxel or the combination of both (Figure 2A). In plasma and tumor tissue, the

exposures of both MLN8237 and docetaxel were similar whether dosed alone or in

combination with the other agent.

MLN8237 and docetaxel lead to a transient accumulation of cells in mitosis.

Therefore, the effect of a single dose of MLN8237 and docetaxel alone or combined on

the tumor mitotic index was evaluated in mice bearing the MDA-MB-231 xenograft.

Tumor mitotic index was evaluated using two independent markers for mitotic cells,

phosphorylation of Histone H3 on serine 10 (pHistH3) and a mitotic specific antigen

MPM2. In all cases, the mitotic index increased within two hours after dosing; however

there were no notable differences in the mitotic index in the combination relative to the

single agents (Figure 2B).

As no marked change in the mitotic index was observed with combined

MLN8237 and docetaxel relative to the single agents in MDA-MB-231 model, the effect

of the MLN8237 and docetaxel combination on tumor morphology was evaluated after

treating PHTX-14B and MDA-MB-231 tumor bearing mice with 10 mg/kg MLN8237

(QDx10days), 5 mg/kg docetaxel (QW days 1 and 8), or the combination of MLN8237

and docetaxel (Figure 3). In the PHTX-14B tumors, the combination of MLN8237 and

docetaxel led to marked changes in tumor morphology, with increased mitotic and

multinucleated cells and significant fibrosis (Figure 3A). In regions of the tumors where




14

viable cells remained, there was a significant increase in non-tumor tissue which

comprised necrotic and fibrotic regions along with stromal infiltrate (Figure 3B). Only a

modest increase in apoptotic cells as determined by cleaved caspase 3 staining was

observed with single agent docetaxel and the combination (Supplementary Figure 3),

potentially due to the transient nature of this marker during the apoptotic cascade. The

morphological effects of the combination were less evident in the MDA-MB-231 model,

however mitotic and multinucleated cells were observed (Figure 3C and 3D).

One of the primary dose limiting toxicities of MLN8237 in cancer patients is

myelosuppression [38, 39]. Therefore, there is risk for overlapping toxicity when

combining MLN8237 with docetaxel or paclitaxel as myelosuppression is a common

dose limiting toxicity for taxanes as well. One path towards reducing the risk of

overlapping toxicity is to decrease the dosing frequency for MLN8237. Therefore, the

antitumor activity of MLN8237 dosed intermittently (3 days on / 4 days off) with

docetaxel once per week (QW) for three consecutive weeks was evaluated. In the

PHTX-02B model, 20 mg/kg MLN8237 dosed 3 days on / 4 days off with 5 mg/kg

docetaxel (QW days 1,8 and 21) resulted in significant tumor growth inhibition and

tumor growth delay relative to the single agents (Figure 4A, Table 1). MLN8237 dosed

3 days on / 4 days off with docetaxel dosed weekly resulted in synergistic antitumor

activity in the PHTX-14B model as well (Figure 4B, Table 1). In fact, the extent of

antitumor activity (TGI and TGD) achieved with the combination in both models with the

3 days on / 4 days off schedule equaled that observed when MLN8237 was dosed

consecutively for 21 days at an identical total daily dose. These data support dosing

MLN8237 on an intermittent schedule as a potential means to minimize overlapping




15

dose-limiting toxicities while maintaining antitumor activity when combined with weekly

taxanes.

The in vivo antitumor activity of MLN8237 was also tested in combination with

paclitaxel in the MDA-MB-231 and PHTX-14B xenografts (Figure 5, Table 2). The

maximum tolerated dose for paclitaxel was determined to be 30 mg/kg dosed once

weekly (QW) as doses above this led to body weight loss exceeding 10%. In the MDA-

MB-231 xenograft, 3, 10 and 20 mg/kg MLN8237 (QDx21days) combined with 5, 10, 15,

20 or 30 mg/kg paclitaxel (QW days 1, 8, and 15) led to additive or synergistic antitumor

activity (Figure 5A, Table 2). With 10 and 20 mg/kg MLN8237, paclitaxel at 20 and 30

mg/kg led to substantial tumor growth delay (Table 2). In the primary human tumor

xenograft PHTX-14B, 20 mg/kg MLN8237 (QD) combined with 10 and 20 mg/kg

paclitaxel (QW) led to synergistic antitumor activity (Figure 5B, Table 2) with tumor

growth delay extending beyond the observation period. The antitumor activity of

MLN8237 dosed intermittently (3 days on / 4 days off) with paclitaxel once per week

(QW) for three consecutive weeks was also evaluated. In the MDA-MB-231 model, 20

mg/kg MLN8237 dosed 3 days on / 4 days off with 20 mg/kg paclitaxel (QW days 1,8

and 21) resulted in additive tumor growth inhibition relative to the single agents

(Supplementary Figure 4). Total body weight loss in mice for all MLN8237 and

paclitaxel treatment regimens never exceeded 10%. In addition, no pharmacokinetic

drug-drug interaction between MLN8237 and paclitaxel was observed in mice as the

pharmacokinetic profiles for each agent in plasma and MDA-MD-231 tumor tissue were

similar whether dosed alone or in combination with the other agent (Supplementary

Figure 5).




16

The durable antitumor activity of MLN8237 combined with both docetaxel and

paclitaxel in preclinical tumor models presented provided part of the rationale for

evaluating the safety and antitumor activity of MLN8237 combined with paclitaxel in

recurrent ovarian cancer patients (NCT01091428). In this clinical study, MLN8237 was

dosed BID 3 days on / 4 days off concomitantly with paclitaxel dosed weekly (QWx3) at

60 or 80 mg/m2 on a 28-day schedule [40]. In order to guide dose-schedule selection

for combined MLN8237 and paclitaxel in cancer patients, an exposure-efficacy model

based on non-clinical and clinical data was developed to predict which MLN8237 /

paclitaxel combinations results in the greatest antitumor efficacy. An exposure-efficacy

surface plot (Figure 6A) and isobologram (Figure 6B) relating MLN8237 and paclitaxel

exposures to tumor growth inhibition was generated from in vivo efficacy studies in

tumor-bearing mice (Table 2). The free-fraction corrected clinical exposures of

MLN8237 dosed BID based on previously published pharmacokinetic data[36, 39] and

the clinical exposures of paclitaxel at 60 or 80 mg/m2 doses of paclitaxel determined

from its human plasma clearance [37, 41] were mapped onto the isobologram by

correcting for mice-human variation in plasma protein binding and maximum tolerated

exposures for both agents. A combination Emax exposure-efficacy model provided a

reasonable fit to the data, as is shown in the diagnostic plots (Supplementary Figure 2).

This translational approach demonstrated allowed placing the clinically achieved

exposures of the combination in the context of the preclinically observed antitumor

efficacy. Placed in this context, the model predicted that 80 and 60 mg/m2 paclitaxel

lead to similar levels of efficacy (Figure 6B and C), consistent with clinical observations

in some cancer indications [42, 43]. In contrast, placing the MLN8237 exposures in the




17

context of the preclinical antitumor efficacy suggests that increasing the dose of

MLN8237 from 10 up to 50 mg BID would result in increasing antitumor activity (Figure

6B and C). This approach allows for rank ordering various combination doses and

schedules of MLN8237 and paclitaxel to predict which pair leads to the greatest

antitumor activity. For example, overlapping toxicities could prevent escalation of

MLN8237 to a biologically active exposure range when combined with 80 mg/m2

paclitaxel. If reducing the dose of paclitaxel to 60 mg/m2 can mitigate overlapping

toxicities allowing for higher MLN8237 doses, the translational approach demonstrated

here suggests that this would also result in increased antitumor activity relative to 80

mg/m2 single agent paclitaxel.




18

Discussion

Antimitotic therapies are a mainstay for cancer care, as they are used broadly in

both solid and heme-lymphatic cancers. Traditionally, these therapies have comprised

agents that directly target microtubules, and include the taxanes, vinca alkaloids and

epothilones. Recently, encouraging activity has been observed with microtubule-

perturbing agents such as the microtubule destabilizer mono-methyl aurastatin E

conjugated to antibodies, including brentuximab vedotin and trastuzumab-DM1 for

treating CD30+ lymphomas and Her2+ breast cancer respectively [44, 45]. Despite the

success of antimitotic therapies across many indications, strategies to improve

response rates and extend responses in patients are needed. Here, we demonstrated

improved antitumor activity and extended duration of response in multiple triple negative

breast cancer models by combining two classes of antimitotic agents, taxanes and the

Aurora A kinase inhibitor MLN8237.

Mice bearing three xenograft models of triple negative breast cancer including two

primary models were treated with various doses of MLN8237 combined with docetaxel

or paclitaxel. In each tumor model, the combination led to greater tumor growth

inhibition relative to the single agents, additive or synergistic antitumor activity while

dosing, and prolonged tumor growth delay after discontinuing treatment. Notably, in

several cases the combination of MLN8237 and either taxane led to tumor regressions

and in some mice the tumors never reformed after discontinuing treatment; outcomes

that did not occur with the single agents when dosed at the individual maximum

tolerated dose.




19

In MDA-MB-231 tumor xenografts treated with combined MLN8237 and docetaxel,

there was no notable difference in the mitotic index, necrotic and fibrotic content, or

stromal infiltrate compared to tumors treated with the single agents after 10 days of

dosing. It is possible that the timing of this analysis in this tumor model was not optimal

to capture the events underlying the antitumor activity observed after 21 days of dosing.

However, the impact of this combination on tumor morphology was more evident in the

PHTX-14B tumor xenograft as a histopathological assessment revealed distinct tissue

morphology changes in combination treated tumors relative to the single agent treated

tumors, including increased non-tumor tissue (necrotic cells, fibrosis, stromal infiltrate)

and multinucleated tumor cells. The multinucleated phenotype is consistent with

previous observations in cultured tumor cells demonstrating that concurrent Aurora A

kinase inhibition using the selective small molecule inhibitor MLN8054 or siRNA with

microtubule perturbing agents including taxanes caused cells to exit mitosis via mitotic

slippage [31]. Cells that exit mitosis by mitotic slippage enter the G1 stage of the cell

cycle with a tetraploid DNA content often accompanied by multiple nuclei due to hyper-

karyokinesis that can occur when the nuclear membrane reforms [2] [4]. Depending on

several underlying genetic factors, these cells can reenter the cell cycle and undergo

another round of DNA replication through a process known as endoreduplication where

they subsequently are characterized as polyploid (>4N). Therefore, the multinucleated

phenotype observed in the tumor tissue with combined MLN8237 and docetaxel

suggests that the mechanism elucidated in cell culture with Aurora A inhibition

combined with taxanes occurs in in vivo tumor models as well.




20

MLN8237 has been evaluated in multiple PI and PII clinical studies [36] [39, 46, 47].

In the first-in-human P1 study, a partial response was achieved in one patient with

platinum and radiation refractory ovarian cancer that lasted for greater than 1 year [38].

Single agent MLN8237 was subsequently investigated in a phase II study in patients

with platinum-resistant or platinum-refractory ovarian, primary peritoneal and fallopian

tube cancers [48]. In this study, objective responses occurred in 10% of patients (n=3

out of 31) as determined by Response Evaluation Criteria in Solid Tumors (RECIST)

and/or reduction in plasma CA-125, warranting further MLN8237 studies in this

indication in combination with other therapeutics including with taxanes.

MLN8237 was tested in combination with weekly paclitaxel in patients with

recurrent ovarian cancer (NCT01091428) [40]. Given that myelosuppression is a

common adverse event for both MLN8237 and paclitaxel, weekly paclitaxel (QWX3 28

day cycle) was selected for this study as it is known to have a decreased incidence of

myelosuppression relative to paclitaxel dosed once every 3 weeks [49, 50]. For

MLN8237, an intermittent schedule of 3 days on (BID) / 4 days off was selected for

combining with weekly paclitaxel, rather than using the single agent MLN8237

recommended schedule of days 1 through 7 on a 21 day cycle [38, 39]. This

intermittent schedule allows for concurrent administration of MLN8237 with weekly

paclitaxel which may be necessary for amplifying the mitotic defects caused by this

combination. Importantly, this intermittent MLN8237 schedule may also further reduce

the risk for overlapping toxicities such as myelosuppression. Previous studies modeling

hematological toxicity by assessing the PK-absolute neutrophil count (ANC) relationship

in rats using methods similar to those described by Friberg and colleagues [51]




21

predicted MLN8237 dosed on an intermittent schedule of 3 days on (BID) / 4 days off

concomitantly with the weekly taxanes will decrease the incidence of dose limiting

neutropenia compared with a 7 day continuous schedule [52]. Moreover, the

approximate 23 hour mean steady state half life of the drug in patients allows for near

complete MLN8237 clearance during the 4 day break which should allow for reversion

of myelosuppressive effects caused by MLN8237 [38]. In our in vivo efficacy

experiments we showed that MLN8237 dosed 3 days on / 4 days off was synergistic

when combined with weekly docetaxel. Of note, the extent of the antitumor activity for

docetaxel combined with MLN8237 dosed 3 days on / 4 days off was nearly identical to

that of MLN8237 dosed for 21 consecutive days. Importantly the intermittent dosing

schedule enabled a significant decrease in the total dose of MLN8237 by 57%. In the

MDA-MB-231 model, continuous dosing of MLN8237 led to slightly greater antitumor

activity relative to dosing 3 days on / 4 days off when combined with paclitaxel,

however, both MLN8237 schedules led to additive antitumor activity.

We developed an exposure-efficacy model to relate MLN8237 and paclitaxel

exposures to antitumor activity using tumor growth inhibition from the efficacy studies

performed with the MDA-MB-231 xenograft. This model was translationally applied in

context of clinical exposures of paclitaxel and MLN8237 after inter-species corrections

for plasma protein binding and maximum tolerated exposures of the two agents. An

isobolographic representation of the response surface was used to rank order pairs of

doses of the two agents in the combination setting. This translational PK-efficacy

analysis predicted that the combination of MLN8237 and paclitaxel at the doses

explored in the clinic will have greater antitumor activity than the single agent standard




22

dose for paclitaxel (80 mg/m2) and MLN8237 (50 mg BID) (Figure 6C). In addition, the

modeling predicted that 80 and 60 mg/m2 paclitaxel lead to similar levels of efficacy

alone or in combination with MLN8237. Several observations have been reported that

paclitaxel as a single agent or in combination with other therapeutics dosed weekly at

60 mg/m2 provides similar efficacy to paclitaxel at 80 mg/m2, however 60 mg/m2 is

better tolerated [42, 43] . The model also predicted that higher doses of MLN8237 with

either dose of paclitaxel will lead to increased antitumor activity. Therefore, in patients,

if higher doses of MLN8237 can be achieved with 60 mg/m2 rather than 80 mg/m2 of

paclitaxel, the model predicts increased antitumor activity. In addition, an exposure

related pharmacodynamic effect in tumors was demonstrated during phase 1 testing of

MLN8237 [39], suggesting that the doses of MLN8237 between 30 and 50 mg BID are

likely to result in biologically active exposures in regard to Aurora A kinase inhibition in

tumors. Therefore, both the exposure-efficacy model and the phase 1

pharmacodynamic results support using higher doses of MLN8237 when combined with

paclitaxel.

Viewed from a broader perspective, more generalized applications of the

quantitative model-based translational pharmacology approach applied in this analysis

of the MLN8237-paclitaxel combination are readily apparent. As dose escalation studies

in patient populations with advanced cancers can be resource and time-consuming and

not all dose pairs can be clinically evaluated, it is envisioned that systematic analysis of

the exposure-efficacy surface for antitumor activity in preclinical xenograft models may

represent a key enabler for clinical development of oncology drug combinations. These

results, when coupled with clinical pharmacokinetic, pharmacodynamic and safety data




23

analyses, offer potential to objectively guide prioritization and optimization of dose-

finding in combination Phase 1 trials to support qualification of the therapeutic window

for optimal benefit-risk balance in anticancer drug development.




24

References

1. Horwitz, S.B. Taxol (paclitaxel): mechanisms of action. Ann Oncol 1994; 5 Suppl 6:S3-6.

2. Yamada, H.Y. and G.J. Gorbsky. Spindle checkpoint function and cellular sensitivity to antimitotic drugs. Mol Cancer Ther 2006; 5(12):2963-9.

3. Gascoigne, K.E. and S.S. Taylor. Cancer cells display profound intra- and interline variation following prolonged exposure to antimitotic drugs. Cancer Cell 2008; 14(2):111-22.

4. Ecsedy, J.A., A. Chakravarty, M. Manfredi, and N. D'Amore. Current and next generation antimitotic therapies in cancer therapy.In: Frank, D. A., editor. Signaling Pathways in Cancer Pathogenesis and Therapy. Springer Publishing; 2012 p. 5-22.

5. Green, M.R., J.E. Woolery, and D. Mahadevan. Update on Aurora Kinase Targeted Therapeutics in Oncology. Expert Opin Drug Discov 2011; 6(3):291-307.

6. Kelly, K.R., J. Ecsedy, D. Mahalingam, S.T. Nawrocki, S. Padmanabhan, F.J. Giles, et al. Targeting aurora kinases in cancer treatment. Curr Drug Targets 2011; 12(14):2067-78.

7. Barr, A.R. and F. Gergely. Aurora-A: the maker and breaker of spindle poles. J Cell Sci 2007; 120(17):2987-96.

8. Bischoff, J.R. and G.D. Plowman. The Aurora/Ipl1p kinase family: regulators of chromosome segregation and cytokinesis. Trends Cell Biol 1999; 9(11):454-9.

9. Carmena, M. and W.C. Earnshaw. The cellular geography of aurora kinases. Nat Rev Mol Cell Biol 2003; 4(11):842-54.

10. Giet, R., C. Petretti, and C. Prigent. Aurora kinases, aneuploidy and cancer, a coincidence or a real link? Trends Cell Biol 2005; 15(5):241-50.

11. Lens, S.M., E.E. Voest, and R.H. Medema. Shared and separate functions of polo-like kinases and aurora kinases in cancer. Nat Rev Cancer 2010; 10(12):825-41.




25

12. Glover, D.M., M.H. Leibowitz, D.A. McLean, and H. Parry. Mutations in aurora prevent centrosome separation leading to the formation of monopolar spindles. Cell 1995; 81(1):95-105.

13. Hoar, K., A. Chakravarty, C. Rabino, D. Wysong, D. Bowman, N. Roy, et al. MLN8054, a small-molecule inhibitor of Aurora A, causes spindle pole and chromosome congression defects leading to aneuploidy. Mol Cell Biol 2007; 27(12):4513-25.

14. Kaestner, P., A. Stolz, and H. Bastians. Determinants for the efficiency of anticancer drugs targeting either Aurora-A or Aurora-B kinases in human colon carcinoma cells. Mol Cancer Ther 2009; 8(7):2046-56.

15. Katayama, H., H. Zhou, Q. Li, M. Tatsuka, and S. Sen. Interaction and feedback regulation between STK15/BTAK/Aurora-A kinase and protein phosphatase 1 through mitotic cell division cycle. J Biol Chem 2001; 276(49):46219-24.

16. Marumoto, T., S. Honda, T. Hara, M. Nitta, T. Hirota, E. Kohmura, et al. Aurora-A kinase maintains the fidelity of early and late mitotic events in HeLa cells. J Biol Chem 2003; 278(51):51786-95.

17. Sasai, K., J.M. Parant, M.E. Brandt, J. Carter, H.P. Adams, S.A. Stass, et al. Targeted disruption of Aurora A causes abnormal mitotic spindle assembly, chromosome misalignment and embryonic lethality. Oncogene 2008; 27(29):4122-7.

18. Huck, J.J., M. Zhang, A. McDonald, D. Bowman, K.M. Hoar, B. Stringer, et al. MLN8054, an inhibitor of Aurora A kinase, induces senescence in human tumor cells both in vitro and in vivo. Mol Cancer Res 2010; 8(3):373-84.

19. Manfredi, M.G., J.A. Ecsedy, A. Chakravarty, L. Silverman, M. Zhang, K.M. Hoar, et al. Characterization of Alisertib (MLN8237), an investigational small-molecule inhibitor of aurora A kinase using novel in vivo pharmacodynamic assays. Clin Cancer Res 2011; 17(24):7614-24.

20. Basso, A.D., M. Liu, K. Gray, S. Tevar, S. Lee, L. Liang, et al. SCH 1473759, a novel Aurora inhibitor, demonstrates enhanced anti-tumor activity in combination with taxanes and KSP inhibitors. Cancer Chemother Pharmacol 2011; 68(4):923-33.

21. Curry, J., H. Angove, L. Fazal, J. Lyons, M. Reule, N. Thompson, et al. Aurora B kinase inhibition in mitosis: strategies for optimising the use of aurora kinase inhibitors such as AT9283. Cell Cycle 2009; 8(12):1921-9.




26

22. Scharer, C.D., N. Laycock, A.O. Osunkoya, S. Logani, J.F. McDonald, B.B. Benigno, et al. Aurora kinase inhibitors synergize with paclitaxel to induce apoptosis in ovarian cancer cells. J Transl Med 2008; 6:79.

23. Hata, T., T. Furukawa, M. Sunamura, S. Egawa, F. Motoi, N. Ohmura, et al. RNA interference targeting aurora kinase a suppresses tumor growth and enhances the taxane chemosensitivity in human pancreatic cancer cells. Cancer Res 2005; 65(7):2899-905.

24. Mazumdar, A., Y.C. Henderson, A.K. El-Naggar, S. Sen, and G.L. Clayman. Aurora kinase A inhibition and paclitaxel as targeted combination therapy for head and neck squamous cell carcinoma. Head Neck 2009; 31(5):625-34.

25. Qi, W., L.S. Cooke, X. Liu, L. Rimsza, D.J. Roe, A. Manziolli, et al. Aurora inhibitor MLN8237 in combination with docetaxel enhances apoptosis and anti-tumor activity in mantle cell lymphoma. Biochem Pharmacol 2011; 81(7):881-90.

26. Shimomura, T., S. Hasako, Y. Nakatsuru, T. Mita, K. Ichikawa, T. Kodera, et al. MK-5108, a highly selective Aurora-A kinase inhibitor, shows antitumor activity alone and in combination with docetaxel. Mol Cancer Ther 2010; 9(1):157-66.

27. Kumano, M., H. Miyake, T. Terakawa, J. Furukawa, and M. Fujisawa. Suppressed tumour growth and enhanced chemosensitivity by RNA interference targeting Aurora-A in the PC3 human prostate cancer model. BJU Int 2010; 106(1):121-7.

28. Lin, Y.G., A. Immaneni, W.M. Merritt, L.S. Mangala, S.W. Kim, M.M. Shahzad, et al. Targeting aurora kinase with MK-0457 inhibits ovarian cancer growth. Clin Cancer Res 2008; 14(17):5437-46.

29. Tanaka, E., Y. Hashimoto, T. Ito, K. Kondo, M. Higashiyama, S. Tsunoda, et al. The suppression of aurora-A/STK15/BTAK expression enhances chemosensitivity to docetaxel in human esophageal squamous cell carcinoma. Clin Cancer Res 2007; 13(4):1331-40.

30. Jeet, V., P.J. Russell, N.D. Verma, and A. Khatri. Targeting aurora kinases: a novel approach to curb the growth & chemoresistance of androgen refractory prostate cancer. Curr Cancer Drug Targets 2012; 12(2):144-63.

31. Wysong, D.R., A. Chakravarty, K. Hoar, and J.A. Ecsedy. The inhibition of Aurora A abrogates the mitotic delay induced by microtubule perturbing agents. Cell Cycle 2009; 8(6):876-88.




27

32. Jiang, Y., Y. Zhang, E. Lees, and W. Seghezzi. AuroraA overexpression overrides the mitotic spindle checkpoint triggered by nocodazole, a microtubule destabilizer. Oncogene 2003; 22(51):8293-301.

33. Anand, S., S. Penrhyn-Lowe, and A.R. Venkitaraman. AURORA-A amplification overrides the mitotic spindle assembly checkpoint, inducing resistance to Taxol. Cancer Cell 2003; 3(1):51-62.

34. Hudis, C.A. and L. Gianni. Triple-negative breast cancer: an unmet medical need. Oncologist 2011; 16 Suppl 1:1-11.

35. Chakravarty, A., V. Shinde, J. Tabernero, A. Cervantes, R.B. Cohen, E.C. Dees, et al. Phase I assessment of new mechanism-based pharmacodynamic biomarkers for MLN8054, a small-molecule inhibitor of Aurora A kinase. Cancer Res 2011; 71(3):675-85.

36. Dees, E.C., J.R. Infante, R.B. Cohen, B.H. O'Neil, S. Jones, M. von Mehren, et al. Phase 1 study of MLN8054, a selective inhibitor of Aurora A kinase in patients with advanced solid tumors. Cancer Chemother Pharmacol 2011; 67(4):945-54.

37. Thummel, K.E., D. D. Shen, N. Isoherranen. Appendix II: Design and Optimization of Dosage Regimens: Pharmacokinetic Data. In: Brunton, L., Chabner, B., Knollman, B., editors. Goodman and Gilman's The Pharmacological Basis of Therapeutics. McGraw-Hill; 2010 p.1891.

38. Dees, E.C., R.B. Cohen, M. von Mehren, T.E. Stinchcombe, H. Liu, K. Venkatakrishnan, et al. Phase I study of aurora A kinase inhibitor MLN8237 in advanced solid tumors: safety, pharmacokinetics, pharmacodynamics, and bioavailability of two oral formulations. Clin Cancer Res 2012; 18(17):4775-84.

39. Cervantes, A., E. Elez, D. Roda, J. Ecsedy, T. Macarulla, K. Venkatakrishnan, et al. Phase I pharmacokinetic/pharmacodynamic study of MLN8237, an investigational, oral, selective aurora a kinase inhibitor, in patients with advanced solid tumors. Clin Cancer Res 2012; 18(17):4764-74.

40. Falchook, G.S., B.A. Goff, R. Kurzrock, H.J. Gray, L.P. Martin, R.L. Coleman, et al. Phase I/II study of weekly paclitaxel with or without MLN8237 (alisertib), an investigational aurora A kinase inhibitor, in patients with recurrent epithelial ovarian, fallopian tube, or primary peritoneal cancer (OC), or breast cancer (BrC): Phase I results. J Clin Oncol 30:15s, 2012 (suppl; abstr 5021).

41. Fennelly, D., C. Aghajanian, F. Shapiro, C. O'Flaherty, M. McKenzie, C. O'Connor, et al. Phase I and pharmacologic study of paclitaxel administered weekly in patients with relapsed ovarian cancer. J Clin Oncol 1997; 15(1):187-92.




28

42. Rose, P.G., M. Smrekar, and N. Fusco. A phase II trial of weekly paclitaxel and every 3 weeks of carboplatin in potentially platinum-sensitive ovarian and peritoneal carcinoma. Gynecol Oncol 2005; 96(2):296-300.

43. Bookman, M.A. Dose-dense chemotherapy in advanced ovarian cancer. Lancet 2009; 374(9698):1303-5.

44. Pro, B. and G.F. Perini. Brentuximab vedotin in Hodgkin's lymphoma. Expert Opin Biol Ther 2012; 12(10):1415-21.

45. Barginear, M.F., V. John, and D.R. Budman. Trastuzumab-DM1: a clinical update of the novel antibody-drug conjugate for HER2-overexpressing breast cancer. Mol Med 2012; 18:1473-9.

46. Matulonis, U.A., S. Sharma, S. Ghamande, M.S. Gordon, S.A. Del Prete, I. Ray-Coquard, et al. Phase II study of MLN8237 (alisertib), an investigational Aurora A kinase inhibitor, in patients with platinum-resistant or -refractory epithelial ovarian, fallopian tube, or primary peritoneal carcinoma. Gynecol Oncol 2012; 127(1):63-9.

47. Mosse, Y.P., E. Lipsitz, E. Fox, D.T. Teachey, J.M. Maris, B. Weigel, et al. Pediatric phase I trial and pharmacokinetic study of MLN8237, an investigational oral selective small-molecule inhibitor of Aurora kinase A: a Children's Oncology Group Phase I Consortium study. Clin Cancer Res 2012; 18(21):6058-64.

48. Matulonis, U.A., L. Pereira, J. Liu, H. Lee, J. Lee, C. Whalen, et al. Sequential bevacizumab and oral cyclophosphamide for recurrent ovarian cancer. Gynecol Oncol 2012; 126(1):41-6.

49. Gonzalez-Angulo, A.M. and G.N. Hortobagyi. Optimal schedule of paclitaxel: weekly is better. J Clin Oncol 2008; 26(10):1585-7.

50. Baird, R.D., D.S. Tan, and S.B. Kaye. Weekly paclitaxel in the treatment of recurrent ovarian cancer. Nat Rev Clin Oncol 2010; 7(10):575-82.

51. Friberg, L.E., A. Henningsson, H. Maas, L. Nguyen, and M.O. Karlsson. Model of chemotherapy-induced myelosuppression with parameter consistency across drugs. J Clin Oncol 2002; 20(24):4713-21.

52. Le, K., L. Yu, M. Manfredi, J. Ecsedy, L. Silverman, K. Cardoza, et al. Evaluation of optimal dosing regimens for investigational drug MLN8237, an Aurora A kinase inhibitor, in combination with docetaxel through pharmacokinetic-




29

pharmacodynamic (PK-PD) modeling of hematological toxicity. Clinical Pharmacology and Therapeutics. 89: s58-s59, 2011 (suppl; abstr PII-67).




30

Table 1. Antitumor activity summary of MLN8237 combined with docetaxel

Model

MLN8237 dose (QD or 3on/4off)

Docetaxel dose(Q7Dx3)

TGIc (%)

TGD (days)d

Outcome (AUC)e

MDA-MB-231 a 3mg/kg 5mg/kg 54.2 7 Synergistic

10mg/kg 5mg/kg 91.8 >48 Synergistic

10mg/kg 5mg/kg 67.3 11 Synergistic




PHTX-02Bb 10mg/kg 5mg/kg 95.2 35 Synergistic


20mg/kg 5mg/kg 99.3 49 Additive

20mg/kg 3on/4off 5mg/kg 103.9 >51 Synergistic

PHTX-14Bb 10mg/kg 5mg/kg 123.4 >45 Synergistic


10mg/kg 10mg/kg 128 >45 Synergistic


3mg/kg 3on/4off 5mg/kg 85.6 38 Additive

20mg/kg 3on/4off 5mg/kg 123 >60 Synergistic

3mg/kg 3on/4off 10mg/kg 121.5 >60 Additive

20mg/kg 3on/4off 10mg/kg 140.1 >76 Synergistic aOrthotopic MDA-MB-231 xenografts were grown in the fat pad of nude mice and treated with MLN8237 administered orally for 21 days with docetaxel dosed IV once per week bPrimary breast cancer models were grown SC in SCID (PHTX-14B) or NOD (PHTX-02B) mice and treated with MLN8237 administered orally for 21 days with docetaxel dosed IV once per week cTumor growth inhibition (TGI) =(Δ treated / Δ control) x 100 / Δ control, was calculated on the last day of treatment dTumor Growth Delay (TGD) The difference in days between the control and the treated groups to reach 1000mm3. > denotes that the treatment group was terminated prior to reaching 1000mm3 eSynergy analysis based on the area under the curve (AUC) values days 0 through 20




31

Table 2. Antitumor activity summary of MLN8237 combined with paclitaxel

Model MLN8237 dose

(QD) Paclitaxel dose

(Q7Dx3) TGIc (%)

TGD (days)d

Outcome (AUC)e

MDA-MB-231a 20mg/kg 30mg/kg 101.4 35 Synergistic














PHTX-14Bb 20mg/kg 20mg/kg 103 >14 Synergistic


3mg/kg 20mg/kg 72 >14 No data aOrthotopic MDA-MB-231 xenografts were grown in the fat pad of nude mice and treated with MLN8237 administered orally for 21 days with paclitaxel dosed IV once per week bPrimary breast cancer models were grown in SCID mice and treated with MLN8237 administered orally for 21 days with paclitaxel dosed IV once per week cTumor growth inhibition (TGI) =(Δ treated / Δ control) x 100 / Δ control, was calculated on the last day of treatment dTumor Growth Delay (TGD) The difference in days between the control and the treated groups to reach 1000mm3. > denotes that the treatment group was terminated prior to reaching 1000mm3 * Treatment group was terminated prior to reaching 1000mm3 eSynergy analysis based on the area under the curve (AUC) values days 0 through 20




32

Figure legends

Figure 1. MLN8237 combined with docetaxel results in antitumor activity in three

models of triple negative breast cancer, including the cell line xenograft MDA-MB-231

(A), the primary human breast tumor xenograft PHTX-02B (B), and the primary human

breast tumor xenograft PHTX-14B (C). Mice were treated for 21 days with MLN8237

(PO, QD), docetaxel (IV, QWx3), or the combination of both at the indicated doses.

Tumors were measured twice weekly with vernier calipers and error bars represent

standard error of the mean. The dotted line box indicates the 21 day treatment period.

Figure 2. MLN8237 and docetaxel exposures are similar when dosed alone or in

combination and result in an increased mitotic index. A single dose of MLN8237 (10

mg/kg), docetaxel (5 mg/kg), or the combination of both was administered to mice

bearing MDA-MB-231 xenografts. Blood and tumor samples were taken at multiple

times out to 24hours. (A) MLN8237 and docetaxel concentrations are shown in both the

plasma and the MDA-MB-231 tumor xenograft. (B) MDA-MB-231 tumor sections

(n=3/group) were stained with fluorescently labeled antibodies directed against

phosphorylated Histone H3 on serine 10 (pHistH3) or MPM2 and the staining was

quantified as described in the Materials and Methods. Error bars represent standard

deviation.

Figure 3. Hematoxylin and eosin (H&E) staining of tumor xenografts results in

morphological changes following single agent or combination treatment with MLN8237




33

and docetaxel. Mice bearing PHTX-14B (A and B) and MDA-MB-231 (C and D)

xenografts were treated with vehicle, MLN8237 (10mg/kg PO, QD) docetaxel (5mg/kg

IV Q7D) or the combination of both for 10 days. On Day 10 tissues were harvested and

fixed in 10% neutral buffered formalin. Tumor sections were stained by H&E and

imaged. Tumor and non-tumor areas were quantified in regions of the tumors that

maintained viable cells using Definiens Tissue Studio software.

Figure 4. MLN8237 administered on an intermittent 3 days on / 4 days off schedule

combined with docetaxel results in significant antitumor activity in two primary breast

tumor models, PHTX-02B (A) and PHTX-14B (B). Tumor bearing mice were treated

with MLN8237 administered once daily for 3 days for three weeks (3on/4off), docetaxel

administered once weekly for 3 weeks, or the combination of both. Tumors were

measured twice weekly with vernier calipers and error bars represent standard error of

the mean. The dotted line box indicates the 21 day treatment period.

Figure 5. MLN8237 combined with paclitaxel results in significant antitumor activity in

the MDA-MB-231 (A) and PHTX-14B (B) tumor models. Tumor bearing mice were

treated for 21 days with MLN8237 (PO, QD), paclitaxel (IV, Q7D), or the combination of

both. Tumors were measured twice weekly with vernier calipers and error bars

represent standard error of the mean. The dotted line box indicates the 21 day

treatment period.




34

Figure 6. A surface response plot relating MLN8237 and paclitaxel exposures to tumor

growth inhibition (%TGI) generated from multiple in vivo efficacy studies in mice bearing

the MDA-MB-231 xenograft (blue dots) (A). An isobologram derived from the surface

response plot (B). Clinically achieved exposures of MLN8237 (10 or 40 mg BID) and

paclitaxel (60 or 80 mg/m2) from the NCT01091428 study represented by the red stars

were mapped onto the isobologram by correcting for mouse-to-human variation in

plasma protein binding and maximum tolerated exposures for both agents (AUCu/CF

(Correction Factor)). Predicted tumor growth inhibition derived from the exposure-

efficacy surface response plot with increasing doses of MLN8237 administered BID with

or without paclitaxel (C).




Figure 1ol

ume

(mm

3 )

1500

2000

A MDA-MB-231

Treatment Period Treatment Period

1000

1500

lum

e (m

m^3

)

Ave

rage

Tum

or V

o

0

500

1000

0 10 20 30 40 50 60Days

Vehicle MLN8237 3mg/kg Vehicle MLN8237 10mg/kg

0

500

0 10 20 30 40 50 60

Days

Ave

rage

Tum

or V

ol

B

MLN8237 3mg/kg Docetaxel 5mg/kg


Docetaxel 5mg/kg Docetaxel10mg/kg

Vehicle

Docetaxel 5mg/kg Docetaxel10mg/kg



MLN8237 10mg/kg

PHTX-02B

500

1000

1500

2000

2500

e Tu

mor

Vol

ume

(mm

3 )

e Tu

mor

Vol

ume

(mm

3 )

500

1000

1500

2000

2500Treatment PeriodTreatment Period

0

500

0 10 20 30 40 50 60Days

Ave

rage

Ave

rag

0

500

0 20 40 60 80Days

Vehicle MLN8237 10mg/kg QD

MLN8237 10mg/kg QD/Docetaxel 5mg/kg

Docetaxel 5mg/kg Q7D

Vehicle

Docetaxel 5mg/kg Q7D

MLN8237 20mg/kg QD

MLN8237 20mg/kg QD/Docetaxel 5mg/kg

C

2000

2500

m^3

) 2000

2500

mm

^3)

PHTX-14B

Treatment Period Treatment Period

0

500

1000

1500

0 20 40 60 80 100

Ave

rage

Tum

or V

olum

e (m

m

0

500

1000

1500

0 20 40 60 80 100

Ave

rage

Tum

or V

olum

e (m

0 20 40 60 80 100

Days Days


MLN8237 10mg/kg QDDocetaxel 5mg/kg

MLN8237 20mg/kg QD Docetaxel 5mg/kg









A

1000

10000

1000

10000

Figure 2

10

100

MLN

8237

(ng/

mL)

MLN8237 Plasma

MLN8237/Docetaxel Plasma

MLN8237 Tumor

MLN8237/Docetaxel Tumor

10

100

Doc

etax

el (n

g/m

L)

Docetaxel Plasma

MLN8237/Docetaxel PlasmaDocetaxel Tumor

MLN8237/Doceta el T mor10 5 10 15 20 25

Time (hr)

10 5 10 15 20 25

Time (hr)

MLN8237/Docetaxel Tumor

7.00

8.00 % MPM2 Positive% pHistH3 Positive

B

3.00

4.00

5.00

6.00

% P

ositi

ve C

ells

0.00

1.00

2.00

0 2 4 6 8 16 24 0 2 4 6 8 16 24 0 2 4 6 8 16 24Hours

%

MLN8237 10 mg/kg Docetaxel 5 mg/kg MLN8237 10 mg/kg Docetaxel 5 mg/kg




PHTX-14B Day 10

Figure 3

A

Vehicle MLN8237

B

90

100

PHTX-14B Day 10

Tumor

CombinationDocetaxel

20

30

40

50

60

70

80

% A

rea

Non-tumor

0

10

Vehicle MLN8237 Docetaxel Combination

Vehicle MLN8237

MDA-MB-231 Day 10C

DD

60

70

80

90

100

a

MDA-MB-231 Day 10

TumorNon-tumor

CombinationDocetaxel

0

10

20

30

40

50

60

Vehicle MLN8237 Docetaxel Combination

% A

rea




APHTX-02B

Figure 4

0

500

1000

1500

2000

2500

Ave

rage

Tum

or V

olum

e (m

m3 )

Treatment Period

00 20 40 60 80

Days

A

Vehicle MLN8237 20mg/kgMLN8237 20mg/kg Docetaxel 5mg/kg

Docetaxel 5mg/kg

B

Treatment Period20003)

BPHTX-14B

2000

3)

Treatment Period

0

500

1000

1500

0 20 40 60 80 100

Ave

rage

Tum

or V

olum

e (m

m^3

0

500

1000

1500

Ave

rage

Tum

or V

olum

e (m

m^3

0 20 40 60 80 100Day

Vehicle MLN8237 3mg/kg

Docetaxel 10mg/kgDocetaxel 5mg/kg MLN8237 3mg/kg +Taxotere 10mg/kg

MLN8237 3mg/kg +Taxotere 5mg/kg

00 20 40 60 80 100 120

Day


Docetaxel 10mg/kgDocetaxel 5mg/kg

MLN8237 20mg/kg +Taxotere 10mg/kgMLN8237 20mg/kg +Taxotere 5mg/kg




2500

3000

m^3

)

Treatment Period

A MDA-MB-231Figure 5

0

500

1000

1500

2000

Ave

rage

Tum

or V

olum

e (m

m

00 10 20 30 40

Days


MLN8237 10mg/kg Paclitaxel 20mg/kg

MLN8237 20mg/kg/Paclitaxel 20mg/kg

MLN8237 10mg/kg/Paclitaxel 20mg/kg

1400Treatment Period

B PHTX-14B

400

600

800

1000

1200

vera

ge T

umor

Vol

ume

(mm

^3)

0

200

0 10 20 30 40 50

Days

Av


Paclitaxel 20mg/kg Paclitaxel 10mg/kg

MLN8237 20mg/kg Paclitaxel 20mg/kg

MLN8237 20mg/kg Paclitaxel 10mg/kgPaclitaxel 20mg/kg Paclitaxel 10mg/kg




Figure 6

A

% T

GI

MLN8237 AUCu (nM.h)Paclitaxel AUCu (ng/ml.h)

500080mg/m2 Paclitaxel

B

Pac

litax

el A

UC

(ng.

hr/m

L)

55 60 65 70 75 80

85 90

2500

3000

3500

4000

450080mg/m2 Paclitaxel

+ 10mg BID 3-3-3 MLN8237

60mg/m2 Paclitaxel 60mg/m2 Paclitaxel+ 40mg BID 3-3-3 MLN8237

Tota

l Cyc

le U

nbou

nd

5 10 15 20 25 30 35 40

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 50

500

1000

1500

2000

Total Cycle Unbound MLN8237 AUC (nM.hr) x 104

70

80

90

100

Inhi

btio

n

C

10

20

30

40

50

60

70

redi

cted

% T

umor

Gro

wth

0 mg/m2 paclitaxel

60 or 80 mg/m2 paclitaxel

00 10 20 30 40 50

Pr

MLN8237 Dose (mg BID)




Published OnlineFirst June 30, 2014.Mol Cancer Ther Jessica Huck, Mengkun Zhang, Jerome Mettetal, et al. Aurora A kinase inhibitor MLN8237 (alisertib)and schedule of taxanes combined with the investigational Translational exposure-efficacy modeling to optimize the dose

Updated version

10.1158/1535-7163.MCT-14-0027doi:

Access the most recent version of this article at:

Material

Supplementary

http://mct.aacrjournals.org/content/suppl/2014/07/01/1535-7163.MCT-14-0027.DC1

Access the most recent supplemental material at:

Manuscript

Authoredited. Author manuscripts have been peer reviewed and accepted for publication but have not yet been

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://mct.aacrjournals.org/content/early/2014/06/28/1535-7163.MCT-14-0027To request permission to re-use all or part of this article, use this link



http://mct.aacrjournals.org/lookup/doi/10.1158/1535-7163.MCT-14-0027http://mct.aacrjournals.org/content/suppl/2014/07/01/1535-7163.MCT-14-0027.DC1http://mct.aacrjournals.org/cgi/alertsmailto:[email protected]://mct.aacrjournals.org/content/early/2014/06/28/1535-7163.MCT-14-0027http://mct.aacrjournals.org/

Article FileFigures 1-6

Translational exposure-efficacy modeling to optimize the ... · 28/06/2014 · 3 Abstract Aurora A kinase orchestrates multiple key activities allowing cells to transit successfully

Documents