the dna repair inhibitor dt01 as a novel …small molecule therapeutics the dna repair inhibitor...

Small Molecule Therapeutics

The DNA Repair Inhibitor DT01 as a NovelTherapeutic Strategy for Chemosensitizationof Colorectal Liver MetastasisNirmitha I. Herath1,2, Flavien Devun1,2, Marie-Christine Lienafa1,2, Aur�elie Herbette1,2,Alban Denys3, Jian-Sheng Sun2, and Marie Dutreix1,4,5

Abstract

Metastatic liver disease from colorectal cancer is a significantclinical problem. This is mainly attributed to nonresectablemetastases that frequently display low sensitivities to availablechemotherapies and develop drug resistance partly via hyperacti-vation of some DNA repair functions. Combined therapies haveshown somedisease control; however, there is still a need formoreefficient chemotherapies to achieve eradication of colorectalcancer liver metastasis. We investigated the tolerance and efficacyof a novel class of DNA repair inhibitors, Dbait, in associationwith conventional chemotherapy. Dbait mimics double-strandbreaks and activates damage signaling, consequently inhibitingsingle- and double-stranded DNA repair enzyme recruitment. Invitro, Dbait treatment increases sensitivity of HT29 and HCT116colorectal cancer cell lines. In vivo, the pharmacokinetics, biodis-tribution and the efficacy of the cholesterol-conjugated clinical

form of Dbait, DT01, were assessed. The chemosensitizing abil-ities ofDT01were evaluated in associationwith oxaliplatin and 5-fluorouracil in intrahepatic HT29 xenografted mice used as amodel for colorectal cancer liver metastasis. The high uptake ofDT01 indicates that the liver is a specific target. We demonstratesignificant antitumor efficacy in a liver metastasis model withDT01 treatment in combination with oxaliplatin and 5-fluoro-uracil (mean: 501 vs. 872 mm2, P ¼ 0.02) compared to chemo-therapy alone. The decrease in tumor volume is further associatedwith significant histologic changes in necrosis, proliferation,angiogenesis and apoptosis. Repeated cycles of DT01 do notincrease chemotherapy toxicity. Combining DT01 with conven-tional chemotherapy may prove to be a safe and effective ther-apeutic strategy in the treatment of metastatic liver cancer. MolCancer Ther; 15(1); 15–22. �2015 AACR.

IntroductionSecondary hepatic tumors or liver metastases account for

approximately 95% of all hepatic malignancies. A majority ofliver metastases arise through primary tumors of the gastrointes-tinal tract (1). Liver resection remains the principle choice oftreatment for early stage liver metastases (2, 3). However, mostpatients are diagnosed at advanced stages making these patientsunsuitable for surgery. Treatment alternatives such as local abla-tive and transarterial therapies demonstrate a limited effect withlow survival benefit (3, 4).

Chemotherapy is widely used in the treatment of metastaticcolorectal carcinoma. However, patients respond poorly to che-motherapy regimens combining folinic acid, 5-fluorouracil and

either oxaliplatin (FOLFOX) or irinotecan mainly due to multi-drug resistance, preventing the eradication of metastatic disease(5, 6). Therefore, improvements in overcoming colorectal cancerchemoresistance are an urgent need and should be a focus ofinvestigation. Here, we investigate how combination with therecently developed DNA repair inhibitors (Dbait) could improveefficacy of conventional chemotherapy in xenografted animalmodels.

Chemoresistance presents a major obstacle to the efficacy ofcancer treatment. DNA repair plays a key role in chemoresistanceby eliminating the damage induced on chromosomes by thechemotherapeutic agents and inhibitors of DNA repair pathwaysmay provide novel opportunities for restoring tumor sensitivity tothese treatments (7). Dbait molecules are a new class of DNArepair inhibitors triggering false DNA damage signaling in cancercells (8). These molecules are short double-stranded DNA with afree double-strand blunt end, which target key damage signaltransducers such as DNA-dependent protein kinase (DNA-PK;ref, 9) and PARP (10), triggering their activation and amplifyingfalse damage signaling. Consequently, the recruitment of down-stream DNA repair enzymes is impaired, inhibiting several DNArepair pathways such as homologous recombination (9), non-homologous end joining (8, 9), base excision repair, and single-strand break repair (10) leading to an accumulation of unrepaireddamage causing cell death.

We previously showed Dbait to be effective in combinationwith radiotherapy on several radio-resistant tumors, both in vitroand in vivo (8, 11–13). Furthermore, studies on an intestinaltumor rodent model revealed that oral administration of Dbait,

1Recombination, Repair and Cancer Laboratory, Institut Curie, Centrede Recherche, Orsay, France. 2DNA Therapeutics, Genopole, Evry,France. 3Department of Radiology and Interventional Radiology,Centre Hospitalier Universitaire Vaudois, Lausanne, Switzerland.4CNRS UMR3347, Centre Universitaire, Orsay, France. 5INSERMU120, Centre Universitaire, Orsay, France.

Note: Supplementary data for this article are available at Molecular CancerTherapeutics Online (http://mct.aacrjournals.org/).

Corresponding Author:Nirmitha I. Herath, Institut Curie – Section Recherche, 15rue Georges Clemenceau, Centre Universitaire Bat.112, 91405 ORSAY Cedex,France. Phone: 3301-6986-3143; Fax: 3301-6986-3141; E-mail:[email protected]

doi: 10.1158/1535-7163.MCT-15-0408

�2015 American Association for Cancer Research.

MolecularCancerTherapeutics

www.aacrjournals.org 15

on October 11, 2020. © 2016 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst December 4, 2015; DOI: 10.1158/1535-7163.MCT-15-0408

http://mct.aacrjournals.org/

in association with the chemotherapy agents 5-fluorouracil andcamptothecin leads to increased chemosensitization of intestinaltumors and to a low systemic exposure (14). To increase theefficiency of cellular uptake, the Dbait molecule was modified bycovalently linking a cholesterol moiety to the 50-end (DT01;ref. 11). We demonstrated that local administration of DT01 byintratumoral injection in association with radiotherapy increasessurvival of xenografted humanmelanomamodels (13). Howeverto date, the efficacy of systemic administration of DT01 in asso-ciation with chemotherapy has not been investigated.

The aims of the current study were to firstly demonstrate theefficacy of DT01 in vitro, secondly to assess the pharmacokineticsand the distribution of DT01 in the liver, and thirdly to demon-strate the concomitant impact of systemicDT01 administration incombination with conventional chemotherapy (oxaliplatin with50-fluorouracil) in a colorectal cancer metastatic liver tumormodel.

Materials and MethodsCell culture, constructs, Dbait molecules,immunofluorescence, and Western blotting

Colorectal cancer cell lines; HT29 (mutated p53, ATCC: HTB-38) and HCT116 (wild-type p53, ATCC: CCL-247) were pur-chased directly from ATCC. These cells were authenticated byATCC by generating human short tandem repeat profiles bysimultaneously amplifying multiple STR loci and amelogenin(for gender determination) using the Promega PowerPlex Sys-tems. These cells were cultured in the laboratory for less than 6months from the date of purchase in DMEM medium supple-mented with 10% FBS, 1% sodium pyruvate, 100 mg/mL strep-tomycin, and 100 mg/mL penicillin (Invitrogen), when the cur-rent study was performed. HT29 cell line stably expressing lucif-erase was established in-house using a pGL4.5 luciferase reportervector (luc2/CMV/Hygro; Promega). HT29 luciferase cells weresupplemented with 200 mg/mL hygromycin B. All cell lines wereadditionally subjected to mycoplasma testing in-house and werefree of mycoplasma contamination (Biovalley).

Cellswere transfectedwith 2.5mgs ofDbait (50-GCTGTGCCCA-CAACCCAGCAAACAAGCCTAGA-(H) TCTAGGCTTGTTTGCTG-GGTTGTGGGCACAGC-30; Eurogentec) where H is a hexaethyle-neglycol linker and underlined nucleotides are phosphorothio-ates. The cells were sham transfected with an 8 bp oligonucleotidecontrol (8H) complexed with 11 kDa polyethyleneimine (PEI) aspreviously described (8, 9).

gH2AX immunofluorescence was performed as described pre-viously using a monoclonal anti–phospho-Histone H2A.X(Ser139) antibody, clone JBW301 (1:500 dilution; 05-636, Milli-pore; ref. 9).

In vitro proliferation assayCellswere seeded at adensity of 3�104 cells/60mmdishes and

transfected with Dbait. Following treatment, cells were washedand left untreated or treated with a combination of 5 mmol/L ofoxaliplatin (OXA, Sigma) and 2.5 mmol/L of 5-fluouracil (5-FU,Sigma) and live cell counts were performed on days 1, 3, 5, 6, 7,and 9.

Clonogenic assayCells were transfected with Dbait and left untreated or treated

with 5mmol/L ofOXAand2.5mmol/L of 5-FU for 1hour. The cells

were diluted, allowed to grow for 14 days, and the clones werestained with crystal violet and counted.

In vivo experimentsThe current study was carried out in strict accordance with the

EuropeanUnion guidelines for animal care. All animal experimen-tation was approved by the ethics committees of the Institut Curieand the Frenchministry. Surgical procedureswereperformedunderanesthesia with local analgesia to minimize suffering.

AnimalsSix-week-old female NMRINU/NUmice (Janvier) weighing 20 to

22 g were housed in specific pathogen-free environment on a 12-hour light and 12-hour dark schedule with food and water adlibitum. No more than 6 animals were housed per cage and theywere acclimated for at least one week prior to initiating in vivostudies.

Intrahepatic HT29L grafting. HT29 Luciferase (HT29L) cells wereimplanted by direct injection of cell suspensions (1 � 106/10 mLof PBS) onto the upper surface of the left lobe. Tumor growth wasmonitored through bioluminescence analysis (IVIS, CaliperSciences).

DT01molecule. For in vivo studies, DT01 (Dbait with a cholesteroltriethylene glycol incorporated at the 50-end) was used (Agilenttechnologies; ref. 11).

Pharmacokinetics ofDT01.HT29L-graftedmicewere treatedwith asingle intraperitoneal (i.p., n ¼ 4) or intravenous (i.v., n ¼ 3)injection of 5mg of DT01. Blood samples were harvested prior totreatment and 1, 5, 10, 30 minutes, 1, 2, 4, and 6 hours aftertreatment. Plasma was recovered through centrifugation andassayed by ELISA.

Fluorescence measurement of organs. As the ELISA techniquefailed to produce reliable quantification in tissues, we usedfluorescent imaging, a reliable technique for assessing moleculedistribution (15). NMRINU/NU mice were injected with 1 mg oftheDT01fluorescentmolecule (DT01-Cy5) through i.p. (n¼3)ori.v. (n ¼ 3) administration. The fluorescent DT01 (DT01-Cy5)incorporates a cyanine 5 at the thymidine located immediatelyafter the linker. Six hours after injection, fluorescence imagingwasperformed using a Typhoon scanner (GE Helathcare).

DT01 and chemotherapy treatment. HT29L-grafted animals (n ¼49) were allocated into treatment groups and administered onecycle of treatment (Supplementary Table S1). DT01 was sys-temically administered through i.p. injection at a dose of 5 mg/day for 5 consecutive days starting on day 0 (D0). OXA (6 mg/kg, 1� per cycle, day 1) and 5-FU (25 mg/kg, 3� per cycle, Days1–3) were administered 2 or 4 hours after DT01 treatment.These mice were sacrificed 22 days after treatment.

An additional group treated withDT01 andOXA/5-FU at the 4-hour interval (n ¼ 10) were kept after treatment until the termi-nation guidelines were met to assess the duration of treatmentefficacy.

Liver function assessment. Blood samples were obtained throughsubmandibular bleeding in lithium heparin tubes (Sarstedt) on

Herath et al.

Mol Cancer Ther; 15(1) January 2016 Molecular Cancer Therapeutics16




days 0, 4, and 18 after treatment. Plasma alanine transaminase(ALT), aspartate aminotransferase (ASAT), alkaline phosphatase(ALP), glutamyl transpeptidase (GGT), amylase (AMYL), andtotal bilirubin (TBIL) were measured using an MS-Scan II (MeletSchloesing Laboratories).

Toxicity assays.NMRINU/NU mice (n ¼ 50) were treated with twocycles of DT01 at escalating doses of 3 mg/day (30 mg total),5 mg/day (50 mg total), or 8 mg/day (80 mg total) through i.p.injection in combination with OXA or 50-FU. OXA and 50-FUwere administered through systemic i.p. injection at doses of1 � 6 mg/kg or 3 � 25 mg/kg, 4 hours after DT01 treatment,respectively. Animals were observed regularly for any adverseeffects.

HistologyHematoxylin, eosin, and saffron (HES)-stained tumor sections

were assessed by an experienced pathologist (Dr. Huerre, InstitutCurie, Paris, France) in a blinded fashion. Viable and necroticcomponents (indicated by increased cell size, indistinct cellborder, eosinophilic cytoplasm, loss or condensation of thenucleus, or associated inflammation) were expressed as a pro-portion (%) of the total tumor surface. Apoptosis was estimated(weak, <5%;moderate, 5%–10%; significant 10%–20%; and verysignificant, 20%–50%) from representative nonnecrotic fields athigh power.

Digitization and image capture was performed using a whole-slide scanning system (Philips digital pathology solutions).

Ki67 and CD31 immunohistochemistryImmunohistochemistry was performed using rabbit anti-Ki67

(ab28364, 1/500; Abcam) and rabbit anti-CD31 (ab15580, 1/500; Abcam) antibodies. This was followed by a secondarybiotinylated goat anti-rabbit IgG antibody (BA-1000; Vector) andrevealed using a rabbit-specific HRP/DAB (ABC) Detection Kit.Images were captured using a fluorescence microscope (Eclipse90i, Nikon). The average Ki67 index was scored by establishing aratio between Ki67þ and negative cells, in five randomly selectedmicroscopic fields per section. Average microvessel density wasdetermined byCD31 staining. CD31þ vesselswere counted infiverandomly selected microscopic fields per section.

Statistical analysisIn vitro experiments were performed with a minimum of two

independent experiments. Two-sided unpaired t tests were usedfor comparison of cell mortality and survival. Kruskal–Wallistests were used to compare tumor volumes, and histologic data.Error bars indicate SEM, except when specifically indicated. Allstatistical analyses were performed using StatEL software(adScience) and a P value of �0.05 was considered statisticallysignificant.

ResultsDbait treatment increases sensitivity of colon cancer cell lines tochemotherapy

We have previously shown that Dbait acts by activating DNA-PK kinase, which phosphorylate numerous targets including thehistone variant H2AX (8, 9, 14). We first confirmed the activity of

Dbait in two colorectal cancer cell lines (HCT116 and HT29) bymonitoring the pan-nuclear phosphorylation of H2AX (Fig. 1A).

To first investigate the effects of Dbait on cell survival tochemotherapy, we determined the number of living cells, atdifferent time points after treatment with Dbait or OXA and 5-FU or a combination of Dbait with chemotherapy (Fig. 1B). Asalready observed infibroblasts (9)Dbait alone appears to have noeffect on cell proliferation in both cell lines (Fig. 1B). Treatmentwith OXA and 5-FU resulted in a decrease in cell proliferation.However, the level of proliferation was significantly reduced byday 9 in cells transfected with Dbait prior to chemotherapytreatment in both HCT116 and HT29 cell lines compared withchemotherapy alone (P < 0.001 and P < 0.02, respectively). Thesedifferences become apparent particularly at later time points (>5days after treatment) indicating that the increase of efficacy withDbait may be a slow process.

Dbait + CT

0

20

40

60

80

100

120

8H Dbait

HT29P < 0.05NS

0

0.2

0.4

0.6

0.8

1

1.2

0 2 4 6 8 10

Days after transfection

HT29

A

0

0.2

0.4

0.6

0.8

1

1.2

0 2 4 6 8 10

Days after transfection

HCT116

0

20

40

60

80

100

120

8H Dbait

HCT116

B

C

Clo

noge

nic

surv

ival

(%)

DAPI γH2AX Merge

Sha

m

HCT116 HT29

DAPI γH2AX Merge

Dba

it

Clo

noge

nic

surv

ival

(%)

Live

cel

lnum

ber (

milli

ons)

Live

cel

lnum

ber (

milli

ons) 8H Dbait

CT

CTCT +

P < 0.01

P < 0.01

* ** *

Figure 1.Dbait treatment increases sensitivity of colorectal cancer and HCC cancer celllines to chemotherapy. Colorectal cancer (HCT116, HT29) cell lines weretransfected with a sham control or Dbait. A, gHA2X immunostaining. Cellswere transfected with Dbait and immunostaining was performed with agHA2X antibody (pink) and DAPI (blue). B, cell proliferation. Cells wereseeded at a density of 3 � 104 cells/60 mm dish left untreated or treatedwith Dbait alone (gray) or in combinationwith CT (5 mmol/LOXAþ 2.5 mmol/L 5-FU; black) for 1 hour. Live cell counts were performed on days 1, 3, 5, 6, 7,and 9 using Trypan blue post Dbait transfection. C, clonogenic survival.Following transfection, cell lines were either left untreated or treated withDbait alone (gray) or treated in combinationwithCT (black). After 14days, theclones were stained with crystal violet and counted manually. Results areexpressed as an average percentage of clones. � , P < 0.01; the level ofproliferationwas significantly reduced by day 9 in cells transfectedwith Dbaitprior to chemotherapy treatment in both HCT116 and HT29 cell linescompared with chemotherapy alone.

DT01 in Preclinical Models of Colorectal Cancer Liver Metastasis

www.aacrjournals.org Mol Cancer Ther; 15(1) January 2016 17




To confirm the chemosensitization effect of Dbait in combi-nation with OXA and 5-FU, clonogenic survival assays wereperformed on HCT116 and HT29. HCT116 cells showed approx-imately 30% (P < 0.01) lethality after Dbait treatment alone(Fig. 1C) revealing their dependency on repair activity for survival,whereas no significant effect was noted in HT29. As the sensitivityof HCT116 to Dbait was not detected during the first 8 days ofproliferation (Fig. 1B), this result suggests that the cells growingwith Dbait accumulate lethal lesions that impair their survivallater on. Treatment with chemotherapy alone (OXA/5-FU)resulted in a significant decrease in the survival of HCT116(P < 0.001), whereas only a trend was observed with the HT29cell line (P ¼ 0.08). However, combination of Dbait with che-motherapy resulted in a significant reduction in survival in bothcell lines (P ¼ 0.05). HCT116 and HT29 differ by many para-meters including their P53 status (HCT116 being proficientwhereas HT29 is mutated). In this instance, despite some differ-ences in their sensitivity to standalone Dbait treatment both celllines were equally sensitive to the combination of chemotherapywith Dbait.

Pharmacokinetic and biodistribution analyses of i.p. versus i.v.administration of DT01

To avoid transfectant adjuvant toxicity, all in vivo studies wereperformed with DT01, a Dbait–cholesterol conjugate facilitatingthe cellular uptake of these molecules without added toxicity(11). To determine the best route for systemic administration

of DT01 mice were treated with either a single i.p. or an i.v.dose of 5 mg of DT01. I.p. administration resulted in a Cmax

of 578 mg/mL, a tmax of 1 hour and an AUC0–6 of 799, whereasi.v. administration led to a Cmax of 1,917 mg/mL, a tmax of0.08 hours and an AUC0–6 of 799 (Fig. 2A). Pharmacokineticanalyses revealed that after i.p. injection, the plasmatic expo-sure of DT01 was longer than that of i.v. bolus injection withan AUC corresponding to approximately 70% of the AUCwith i.v. administration (Fig. 2A).

We used a fluorescent-labeled cy5-DT01 molecule to mon-itor the biodistribution in excised whole organs (Fig. 2B). Bothcy5-DT01 and DT01 have similar properties in terms ofpharmacokinetics and DNA-PK activation (11). The maximalDT01 fluorescence was observed in the liver, intestines, and thekidneys by both routes with the highest intensities observed inthe liver and intestines following i.p. administration. The highfluorescence emitted by the kidneys and urine observed in micesuggest that DT01 is preferentially eliminated by the kidneys.Although there was no measurable DT01 in the blood 6 hoursafter injection (Fig. 2A), significant amounts of DT01 were stilldetectable in the liver indicating a specific retention in thisorgan (Fig. 2C).

As already demonstrated in vitro, DT01 activation ofDNA-PK intissue can be revealed by the phosphorylation of the histoneH2AX (9). Wemonitored DT01 activity by analyzing distributionof H2AX phosphorylation in livers bearing HT29-grafted tumors.Interestingly, a high level of gH2AX was specifically observed in

0

5

10

15

20

25

Li I K L S

Ave

rage

fluo

resc

ence

(A

U)

Thou

sand

s

NT 1 h3 h 6 h

0

1,000

2,000

0 2 4 6

DT0

1 pl

asm

a co

nc.

(μg/

mL)

Time (h)

i.v. 5 mg

i.p. 5 mg

L Li S I K Bl H B P M

NT

i.p

i.v

A B

C D

Normalliver

Tumor

Glucose DT01

Normalliver

TumorP = NS

P < 0.05

P < 0.05P = NS

P < 0.05

P < 0.01

P = NS

P = NS

P < 0.01P < 0.01

Figure 2.Pharmacokinetics and bio-distribution of DT01. A, DT01 pharmacokinetics. Intrahepatic tumor bearing NMRINU/NU mice were treated with a dose of 5 mgof DT01 through a single i.p. (black) or i.v. (gray) bolus administration. DT01 concentration in plasma was measured at 1, 5, 10, 30 minutes and 1, 2, 4,and 6 hours post treatment using an ELISA assay. B, DT01 biodistribution with fluorescent DT01-cy5. Mice were injected with 1 mg of DT01-cy5 through i.p.or i.v. administration. Six hours after injection, the animals were sacrificed, organs were excised, and scanned using a Typhoon scanner. The organsfrom left to right include lung (L), liver (Li), spleen (S), intestine (I), kidney (K), bladder (Bl), heart (H), brain (B), pancreas (P), and muscle (M). C, DT01-cy5fluorescence quantitation of organs scanned using a Typhoon scanner. The relative fluorescence measurements were recorded and expressed in arbitraryunits (AU) and the results are expressed as an average � STD. The organs from left to right include liver (Li), intestine (I), kidney (K), lung (L), and spleen (S).D, DT01 induction of H2AX phosphorylation in liver tumor cells. Intrahepatic tumor bearing mice were treated with 1 mg of DT01 and the tumors wereharvested 6 hours after treatment and subjected to immunolabeling with a gHA2X antibody (green) and DAPI (in blue).

Herath et al.





the tumor and not in the surrounding healthy tissues (Fig. 2D)indicating a preferential uptake or activity of the DT01 moleculesin the tumor cells of the liver.

DT01 significantly increases sensitivity to OXA and 5-FU in vivoTo explore the interest of associating DT01 with the front-line

treatment for metastatic colorectal cancer, we used a HT29 xeno-grafted liver tumormodel, as previous reports and our in vitro datademonstrate this line to be highly chemoresistant mainly due tothe V600E BRAFmutation (16, 17). The animalswere treatedwithOXA and 5-FU, a treatment close to the traditional FOLFOXprotocol for patients, using two different schedules based onbiodistribution data (Fig. 3A). The two schedules consisted ofeither 2 or 4-hour intervals between the two treatments, as themaximumDT01 levels in the liver were observed at 1 and 3 hoursafter treatment (Fig. 2C). In previous studies we established thatDT01 must be administered prior to the genotoxic treatment inorder to act as a chemosensitizer.

As previously observed in vitro, the tumors were highly resistantto chemotherapy alone and DT01 had only a moderate effectwhen administered alone (Fig. 3B and C). Interestingly, theassociation of DT01 to OXA and 5-FU significantly decreased theliver tumor size in both combination treated groups comparedwith chemotherapy alonewhen administered at 2 (mean volume:525.80 vs. 872.01 mm2, P ¼ 0.03) and 4 hours (mean volume:501.05 vs. 872.01 mm2, P ¼ 0.02) before chemotherapy (Fig. 3Band C). This effect was not observed when DT01 was associatedwith a single chemotherapy agent, either OXA or 5-FU (Supple-mentary Fig. S1). Detailed blinded histologic analyses includingmeasures of the viable tumor area, necrosis, and apoptosis wereassessed in hematoxylin–eosin–saffron (HES)-stained sections,by an experienced pathologist. Both groups with DT01 andchemotherapy combined treatment showed higher treatmentefficacy than the groups receiving single treatment, with amarked increase in necrosis in the group treated with a 4-hourinterval between chemotherapy and DT01 (P < 0.0001) than

0

20

40

60

Mea

n m

icro

vess

elde

nsity

(5 fi

elds

) P = 0.02P < 0.001

DT01 + + + + + +CT

2 h 4 h

0

20

40

60

80

100

Mea

n K

i67

inde

x (%

)

P = 0.02P < 0.001

DT01 + + + + + +CT

2 h 4 h

A

D

B

E

0

400

800

1,200

1,600

Mea

n liv

er tu

mor

vol

ume

(mm

3 )

P = 0.03P = 0.02

DT01 + + + + + +CT

2 h 4 h

Vehicle DT01 CT DT01+CT DT01+CT2 h 4 h

F

0

20

40

60

80

100

Nec

rosi

s (%

)

DT01 + + + + + +CT

2 h 4 h

P < 0.01

P < 0.0001

C

OXA + 5-FUDT01

Day 22Sacrifice

Treatment day 0 – day 4

Figure 3.DT01 significantly increases sensitivity to chemotherapy in a colorectal cancer (HT29) liver metastatic model. Intrahepatic tumor-bearing NMRINU/NU mice weretreated as described in the Materials and Methods, and sacrificed 22 days after treatment. Livers were sampled for macroscopic and microscopic examination.A, sequence of DT01 and CT therapy. Chemotherapy was administered 2 or 4 hours after DT01 treatment. B, mean liver tumor volume (mm3) in each treatmentgroup. C, representative macroscopic and HES sections of liver tumors in each group. D, tumor necrotic component assessed through HES staining.Necrosis is expressed as a proportion (%) of the total tumor surface of the tissue section analyzed. E and F, immunohistochemistry. Proliferation ofthe viable tumor component and the average microvessel density were determined by Ki67 (E) and CD31 (F) staining, respectively. Results are expressedas an average � SEM.






2 hours (P < 0.01), compared with chemotherapy alone(Fig. 3D). Furthermore, a high apoptotic index was apparentin both groups treated with DT01 and chemotherapy (Supple-mentary Fig. S2). Similar to other histologic parameters, theextent of apoptosis was elevated in animals treated with a 4-hour delay (P < 0.0001). These histologic findings were notapparent in the DT01 or chemotherapy alone–treated groups.

For many solid tumors, proliferation and microvasculariza-tion are indispensable prerequisites for tumor development andmetastasis. To further investigate these parameters, immunos-taining for Ki67 and CD31, markers of cell proliferation andangiogenesis respectively, were performed in the viable tumorcomponent (Fig. 3E and F). Ki67 immunoreactivity indicatedthat tumors treated with either DT01 or chemotherapy alonewere densely packed with a high degree of proliferation. Treat-ment with a 2-hour interval between DT01 and chemotherapyresulted in a moderate decrease in proliferating cells (P ¼0.02; Fig. 3E). Strikingly, immunoreactivity of Ki67 was 10-foldreduced in the group treated with a 4-hour interval betweenDT01 and chemotherapy (P < 0.001; Fig. 3E). In this group,immunoreactivity was detected only in the tumor rim due to thehigh degree of necrosis observed in the center core region of thetumor. In addition, diminished intratumoral vessel densitieswere detected in groups treatedwith a combination of DT01 and

chemotherapy, compared with chemotherapy alone (Fig. 3F).However, the mean microvessel density was even more notablyreduced in the group treated with a 4-hour interval (P < 0.001)compared to 2 hours (P ¼ 0.02). Despite similarities in theantitumor effect on tumor growth at both the 2 and 4-hourtreatment schedules, histologically the efficacy was significantlymore pronounced at the 4-hour timepoint, in terms of necrosis,apoptosis, proliferation, and angiogenesis.

Unexpectedly, tumors treated with a delay of 4 hours betweenDT01 and chemotherapy and sampled 22 days after treatmentpresented with a proportion of lysed hepatocytes within thetumor and slight edema in the adjacent nonmalignant liver (Fig.4A), in the absence of further clinical signs of toxicity such as lossof weight (Supplementary Fig. S3). Histologic analyses did notrevealmorphologic signs of toxicity in the other groups (Fig. 3). Inaddition, liver enzyme tests did not reveal significant differencesbetween the control and the combination treated groups (Sup-plementary Fig. S4).

Interestingly, no further edema was observed when animalsreceiving the same treatment were sacrificed between 30 and65 days (Fig. 4A). This suggests that the edema observed at day22, is reversible over time. Despite the significant tumorefficacy observed 22 days post treatment, tumors monitoredafter this time point resumed progression (Fig. 4B). Histologic

50 μm

TNVehicle only

TNDT01+CT (E)

TN

DT01+CT (L)TN

DT01+CT (I)

A

CB

0

1,000

2,000

3,000

4,000

Mea

ntu

mor

volu

me

(mm

3 )

V E I L(22) (22) (30–45) (45–65)

0

20

40

60

80

100

Tum

orsu

rface

com

pone

nts

(%)

V E I L(22) (22) (30–45) (45–65)

ProliferationNecrosis

Figure 4.Tumor properties of mice receiving DT01 and chemotherapy at early, intermediate, and later time points. A, representative HES sections of tumor (T) and adjacentnonmalignant (N) tissues harvested early (E) on day 22, intermediate (I) between days 30 and 45 and late (L) between days 45 and 65, receiving combinationtreatment. B, comparison of mean liver tumor volumes (mm3) in the vehicle-treated (V) and after treatment with DT01 þ chemotherapy with a 4-hour intervalwhen sacrificed at E, I, or L time points. C, restart of proliferation and regrowth of tumor when harvested at I and L time points compared with E time point.

Herath et al.





analysis revealed that the proliferative component reachedapproximately 50% at 30 to 45 days after treatment, onlyslightly below the level observed in nontreated tumors(Fig. 4C).

To confirm that combination treatment did not induce addi-tional toxicity to the liver, we analyzed the tolerability of DT01 inassociation with OXA or 5-FU for extended treatment cycles. Wedetermined the toxicity of escalating doses ofDT01 (total doses of30, 50, or 80mg) following systemic administration for two cycles(5�DT01 administrations per treatment cycle) associated toOXAor 5-FU in a cohort of 50 mice. No loss of weight was observed inanimals during or after treatment (Supplementary Fig. S5A).Similarly, other clinical signs of toxicity such as diarrhea orbehavioral changes were not noted in these mice. At autopsy 6weeks after the second cycle of treatment, all abdominal organs,the thoracic cavity, and contents appeared normal. No majorvariations in liver weights or histology were observed betweenthe vehicle and combination treated groups (SupplementaryFig. S5B).

These results suggest that the reversible edema detected aftercombined treatment (Fig. 4) in animals bearing hepatic tumor islikely an acute reaction to the tumor response to efficient com-bination treatment.

Peritoneal metastasis treatmentColorectal cancer often metastasizes to the liver and the

peritoneum. Interestingly, 90% of the mice intrahepaticallyxenografted with colorectal cancers developed peritonealmetastasis. This property allowed us to monitor the effect ofDT01 not only on liver tumors but also on peritoneal metas-tasis. Animals receiving a combination of DT01 and chemo-therapy displayed significantly decreased peritoneal tumorvolumes when compared with chemotherapy alone at both the2 (mean volume: 300.31 vs. 867.20mm2, respectively, P < 0.01)and 4-hour time intervals (mean volume: 259.51 vs. 867.20mm2, respectively, P < 0.01; Fig. 5A and B). Although a slightdecrease in tumor volume was observed in the group treatedwith DT01 alone, this did not reach statistical significance.

DiscussionApproximately 50% of patients with colorectal cancer will

present either with liver and/or peritoneal metastases or developthem throughout the course of their disease (18). A majority ofpatients with colorectal cancer hepatic metastases present withnonresectable disease and systemic chemotherapy represents themain if not the only form of therapy. However, the therapeuticwindow of chemotherapy is limited due to tumor resistance andhigh toxicity to nontargeted tissue. In such clinical situations, anaggressive chemotherapy regimen alone may not only fail toimprove survival, but may also adversely affect the quality of life.Consequently, the mortality of these patients remains high.Therefore development of new agents' specifically targeting DNArepair to circumvent chemoresistance and sparing healthy tissuesis imperative in the treatment of these cancers. DT01 is anattractive drug candidate based on its central role in DNA repair.

In the current study, we show for the first time that systemicDT01 treatment sensitizes colorectal cancer cells to conventionalchemotherapies by in vitro and in vivo assays. In a colorectal cancermetastatic model, we demonstrate significant antitumor efficacyin the liver and the peritoneum (regarded as a terminal condition)

with DT01 treatment in combination with OXA and 5-FU. It is ofinterest to note, that the significant antitumor effect was limited toDT01 association with both OXA and 5-FU and not with single-agent chemotherapy (Supplementary Fig. S1). This demonstratesthat in agreement with the clinical conventional setting, combi-nation with DT01 must be associated to double chemotherapyrather than single-agent chemotherapy in the treatment of colo-rectal cancer metastases. This study further highlights that tumorsreceiving double chemotherapy combined with DT01 restartproliferation and regrowth at later time points (post 22 days).Therefore repeated cycles of treatment would be necessary toachieve long-term disease control similar to current conventionalchemotherapy protocols. This would be possible as no addedtoxicity was observed with DT01 alone or in combination withOXA or 5-FU.

DT01 preferentially accumulate in the liver and intestines aftersystemic injection. Although the entire liver appeared to beuniformly fluorescent after Cy5-DT01 injection, the activation ofDNA-PK revealed by the phosphorylation of H2AX was observedexclusively in tumor cells and not in the healthy tissue surround-ing the tumor. This observation indicates that either DT01 doesnot enter nontumor cells and/or that DT01 is not active in healthyliver tissue. DT01 was specifically designed by cholesterol conju-gation, first, to increase the bioavailability and second, to play onthe difference in the substrate uptake between cancer and normalcells. Low-density lipoproteins (LDL) are a major component of

0

400

800

1,200

1,600

Mea

n pe

riton

eal t

umor

vo

lum

e (m

m3 )

P = 0.01

P < 0.01

DT01 + + + + + +CT

2 h 4 h

A

B

Veh. DT01 CT DT01+CT4 h

DT01+CT2 h

Figure 5.Association of DT01 with chemotherapy significantly decreases theperitoneal tumor volume. Intrahepatic tumor-bearing NMRINU/NU mice weretreated asdescribed in theMaterials andMethods, and sacrificed 22days aftertreatment. The presence of peritoneal metastasis was visually monitoredduring the duration of treatment and measured at the time of sacrifice. A,mean peritoneal tumor volume (mm3) in each treatment group. B,representative macroscopic images of peritoneal tumors in each group.Results are expressed as an average � SEM.






the cholesterol pathway (19). High requirement for LDL bymalignant cells and thus the consequent overexpression of LDLreceptors has been shown in many types of cancer cells makingtumor cells specific targets of DT01 (20, 21). In addition, anextensive analysis of normal and cancerous human tissues byimmunohistochemistry revealed that either DNA-PKcs or Ku80were consistently absent in the liver and the mammary epitheli-um, a specific post-transcriptional regulation that was not foundin the other tissues and most of the tumors (22). Taken together,these data highlight that DT01 is likely to be an efficient drug forthe treatment of liver cancers (Supplementary Table S2).

In conclusion, there is anurgentneed for new treatmentoptionstargeting secondary hepatic malignancies, a rapidly progressivedisease with a poor prognosis and an alarming rate of mortality.Our study strongly suggests that combining systemic administra-tion of DT01 with conventional chemotherapy may prove to be asafe andeffective therapeutic strategy in the treatmentof colorectalcancer metastasis of the liver and the peritoneum.

Disclosure of Potential Conflicts of InterestJ.-S. Sun has ownership interest in DNA Therapeutics. M. Dutreix has

ownership interest in aDT01 patent and is a consultant/advisory boardmemberfor DNA Therapeutics. No potential conflicts of interest were disclosed by theother authors.

Authors' ContributionsConception and design: N.I. Herath, M. DutreixDevelopment of methodology: N.I. Herath, F. Devun, M.-C. Lienafa

Acquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): N.I. Herath, F. Devun, A. Herbette, A. DenysAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): N.I. Herath, F. Devun, A. Denys, M. DutreixWriting, review, and/or revision of the manuscript: N.I. Herath, F. Devun,A. Denys, J.-S. Sun, M. DutreixAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): N.I. Herath, M.-C. Lienafa, M. DutreixStudy supervision: N.I. Herath, J.-S. Sun, M. Dutreix

AcknowledgmentsThe authors thank Sophie Dodier (Histology platform of Institut Curie) for

histologic services, Professor Michel Huerre for analyses of tumor sections, andAndr�e Nicolas for scanning of histologic sections (Pathology service, l'hopitalCurie). This study was also supported by the technical staff of the Institut Curieanimal facility.

Grant SupportThis study was supported by the interdisciplinary translational program of

the Institut Curie, the National de la Recherche, the Institut National de la Santeet de la RechercheMedicale, Institut National duCancer (TRANSLA13-081) andthe conseil g�en�eral de l'Essonne (A.S.T.R.E). This study was also funded by theInstitut Curie and DNA Therapeutics.

The costs of publication of this articlewere defrayed inpart by the payment ofpage charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received May 18, 2015; revised September 28, 2015; accepted October 14,2015; published OnlineFirst December 4, 2015.

References1. Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D. Global cancer

statistics. CA Cancer J Clin 2011;61:69–90.2. Adam R, Vinet E. Regional treatment of metastasis: surgery of colorectal

liver metastases. Ann Oncol 2004;15 Suppl 4:iv103–6.3. Lin S, Hoffmann K, Schemmer P. Treatment of hepatocellular carcinoma: a

systematic review. Liver Cancer 2012;1:144–58.4. Parkin DM, Bray F, Ferlay J, Pisani P. Global cancer statistics, 2002. CA

Cancer J Clin 2005;55:74–108.5. Asghar U, Meyer T. Are there opportunities for chemotherapy in the

treatment of hepatocellular cancer? J Hepatol 2012;56:686–95.6. Osman AM, Bayoumi HM, Al-Harthi SE, Damanhouri ZA, Elshal MF.

Modulation of doxorubicin cytotoxicity by resveratrol in a human breastcancer cell line. Cancer Cell Int 2012;12:47.

7. Helleday T, Petermann E, Lundin C, Hodgson B, Sharma RA. DNA repairpathways as targets for cancer therapy. Nat Rev Cancer 2008;8:193–204.

8. QuanzM, Berthault N, Roulin C, RoyM,Herbette A, Agrario C, et al. Small-molecule drugs mimicking DNA damage: a new strategy for sensitizingtumors to radiotherapy. Clin Cancer Res 2009;15:1308–16.

9. Quanz M, Chassoux D, Berthault N, Agrario C, Sun J-S, Dutreix M.Hyperactivation of DNA-PK by double-strand break mimicking moleculesdisorganizes DNA damage response. PLoS ONE 2009;4:e6298.

10. Croset A, Cordelieres FP, Berthault N, Buhler C, Sun JS, Quanz M, et al.Inhibition of DNA damage repair by artificial activation of PARP withsiDNA. Nucleic Acids Res 2013;41:7344–55.

11. Berthault N, Maury B, Agrario C, Herbette A, Sun JS, Peyrieras N, et al.Comparison of distribution and activity of nanoparticles with shortinterfering DNA (Dbait) in various living systems. Cancer Gene Ther2011;18:695–706.

12. Coquery N, Pannetier N, Farion R, Herbette A, Azurmendi L, Clarencon D,et al. Distribution and radiosensitizing effect of cholesterol-coupled Dbaitmolecule in rat model of glioblastoma. PLoS ONE 2012;7:e40567.

13. Biau J, Devun F, JdeyW, Kotula E, QuanzM, Chautard E, et al. A preclinicalstudy combining the DNA repair inhibitor Dbait with radiotherapy for thetreatment of melanoma. Neoplasia 2014;16:835–44.

14. Devun F, Bousquet G, Biau J, Herbette A, Roulin C, Berger F, et al.Preclinical study of the DNA repair inhibitor Dbait in combinationwith chemotherapy in colorectal cancer. J Gastroenterol 2012;47:266–75.

15. Vasquez KO, Casavant C, Peterson JD. Quantitative whole body biodis-tribution of fluorescent-labeled agents by non-invasive tomographic imag-ing. PLoS ONE 2011;6:e20594.

16. Davies H, Bignell GR, Cox C, Stephens P, Edkins S, Clegg S, et al. Mutationsof the BRAF gene in human cancer. Nature 2002;417:949–54.

17. Benlloch S, Paya A, AlendaC, Bessa X, AndreuM, Jover R, et al. Detection ofBRAF V600E mutation in colorectal cancer: comparison of automaticsequencing and real-time chemistry methodology. J Mol Diagn 2006;8:540–3.

18. Forner A, Llovet JM, Bruix J. Hepatocellular carcinoma. Lancet 2012;379:1245–55.

19. Radwan AA, Alanazi FK. Targeting cancer using cholesterol conjugates.Saudi Pharm J 2014;22:3–16.

20. Rudling MJ, Reihner E, Einarsson K, Ewerth S, Angelin B. Low densitylipoprotein receptor-binding activity inhuman tissues: quantitative impor-tance of hepatic receptors and evidence for regulation of their expression invivo. Proc Natl Acad Sci U S A 1990;87:3469–73.

21. Maletinska L, Blakely EA, Bjornstad KA, Deen DF, Knoff LJ, Forte TM.Human glioblastoma cell lines: levels of low-density lipoprotein receptorand low-density lipoprotein receptor-related protein. Cancer Res 2000;60:2300–3.

22. Moll U, Lau R, Sypes MA, Gupta MM, Anderson CW. DNA-PK, the DNA-activated protein kinase, is differentially expressed in normal and malig-nant human tissues. Oncogene 1999;18:3114–26.


Herath et al.




2016;15:15-22. Published OnlineFirst December 4, 2015.Mol Cancer Ther Nirmitha I. Herath, Flavien Devun, Marie-Christine Lienafa, et al. Chemosensitization of Colorectal Liver MetastasisThe DNA Repair Inhibitor DT01 as a Novel Therapeutic Strategy for

Updated version

10.1158/1535-7163.MCT-15-0408doi:

Access the most recent version of this article at:

Material

Supplementary

http://mct.aacrjournals.org/content/suppl/2015/12/04/1535-7163.MCT-15-0408.DC1

Access the most recent supplemental material at:

Cited articles

http://mct.aacrjournals.org/content/15/1/15.full#ref-list-1

This article cites 22 articles, 3 of which you can access for free at:

Citing articles

http://mct.aacrjournals.org/content/15/1/15.full#related-urls

This article has been cited by 3 HighWire-hosted articles. Access the articles at:

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

Subscriptions

Reprints and

[email protected]

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

Permissions

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

.http://mct.aacrjournals.org/content/15/1/15To 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-15-0408

http://mct.aacrjournals.org/content/suppl/2015/12/04/1535-7163.MCT-15-0408.DC1

http://mct.aacrjournals.org/content/15/1/15.full#ref-list-1

http://mct.aacrjournals.org/content/15/1/15.full#related-urls

http://mct.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://mct.aacrjournals.org/content/15/1/15


the dna repair inhibitor dt01 as a novel …small molecule therapeutics the dna repair inhibitor...

Documents