THELA USCULTUUTUUDIO 20170319692A1 ( 19) United States (12 ) Patent Application Publication (10 ) Pub. No. : US 2017/ 0319692 A1 Abazeed (43 ) Pub . Date : Nov . 9 , 2017

(54 ) ANTI- AR AGENT AND RADIATION Publication Classification THERAPY FOR ANDROGEN RECEPTOR (51 ) Int. Cl. POSITIVE CANCER A61K 41/ 00 (2006 .01 ) A61K 31 /545 ( 2006 . 01) ( 71 ) Applicant: The Cleveland Clinic Foundation , A61K 41/ 00 (2006 . 01) Cleveland , OH (US ) A61K 51/ 00 (2006 .01 ) ( 72 ) Inventor: Mohamed Abazeed , Cleveland , OH 2 ) U .S . CI. CPC ...... A61K 41 /0038 (2013 . 01 ) ; A61K 51 / 00 ( US ) ( 2013 .01 ); A61K 31/ 545 (2013 . 01 ); A61K ( 21) Appl. No. : 15 /493 , 901 41 /0019 (2013 . 01 ) (57 ) ABSTRACT (22 ) Filed : Apr. 21, 2017 Provided herein are compositions , systems, kits , and meth ods for treating cancer in a subject with androgen receptor Related U . S . Application Data positive cancer cells by sensitizing such cancer cells with (60 ) Provisional application No . 62 /326 ,203 , filed on Apr. anti - androgen receptor therapy ( e . g . , Enzalutamide or anti 22 , 2016 . androgen antibody ) , and then treating with radiation therapy .

DENZ 2 40. ENZ * R + 2 . 0 - ODHT protection pretpersonen 61 DMT IR o Ilia

sensitizationo +

** 4 wit BT474 CAMA1HCC202 *** Patent Application Publication Nov . 9 , 2017 Sheet 1 of 51 US 2017 / 0319692 A1

FIG . IA

Colorectal Ovary Giloma Lung NSC Melanoma

Card Chrasarcc? Pancreas PNET. rost swing 's sarcoma Biliary tract Breast ?? { {{ {{ Q Osteosarcoma NeuroblastomaThyroid Sarcoma Lung SC Erdprurn Kidney Urinary tract Esophagus Stomach Liver Patent Application Publication Nov . 9 , 2017 Sheet 2 of 51 US 2017 /0319692 A1 FIG . 1B

• Neuroblastoma Colon • Melanoma ! Jo breastRhabdomyosarcoma

* R2 = 0 .73 * the With TOHYHome OHNWAutovoHd i ? À 5 ó

High-throughput honthe sense the R2 = 0 .80

i 2 3 4 5 6 Clonogenic Patent Application Publication Nov . 9 , 2017 Sheet 3 of 51 US 2017 /0319692 A1

FIG . 1C

:

)

Survivaldidose) .:mym

.

BiliaryTract Breast? Liver HGG 501PancreasUrinary Tract QUADEsophagusKidney Prostate DIN7LungThyroidNS Stomach Ovary SCLC PNET Melanoma HeadChondrosarcomaand Neck Large Intestine LungAdeno Osteosarcomain situ Soft EndometrialTissue NeuroblastomaSarcoma Ewing Sarcoma Patent Application Publication Nov . 9 , 2017 Sheet 4 of 51 US 2017 /0319692 A1

FIG . 1D 0 .8 DNSCLCs . * ......

.

: cicciaciociacic wenn LUAD

POHOOROOOOOOOOOOOOOOOOOOOOOOOOOOOOOO , #ofcelllinespoijo MONOKOKKOKKOKKOKKOKKOKOKOHOHOHOHOHOHOICICIOIOIOIOIOIOIICIGICICICICICICICICIOCICICIKCIIIIIIIIIIIIIIIII ccccccccccaciaCOCOCHOCOccicciociaciccioCIOCICCIOCCO All 01234567 0 1 2 3 4 5 6 7 Survival d?dose ) Survival d dose)

AN NSCLC 6 LUAD

$ 5 5

4 4

€ SampleQuantile 3 3

Ž STATISTICSDJ * 2 2

porn vivivivivinidivivivivivivivindiriminiminimividwivivivivivivivirtuvininiwiwiwininiiiiiiwidininingLappacocopanonogooooogooooooooooooooo - 3 - 2 - 10 12. Juho 3 wwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwww - 2 - 1 0 171 2 - 2 - 1 0 2 Theoretical Quantile Patent Application Publication Nov . 9 , 2017 Sheet 5 of 51 US 2017 / 0319692 A1

FIG . 1E

Teniposide - Clofarabine Topotecan Cytarabine me Irinotecan Vincistines t imi Chlorambuci Paclitaxel Etoposide Doxorubicin Gemcitabins Mitomycin Decitabine RawCorrelation Bleomycin

0 250 500 Patent Application Publication Nov . 9 , 2017 Sheet 6 of 51 US 2017 / 0319692 A1

FIG . 2A reunion

: :: *

ool C * og2(mutation) onw * fSCNA win :: to ???????????????????????????????????????????? w 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 survival d( dose ) Patent Application Publication Nov . 9 , 2017 Sheet 7 of 51 US 2017 /0319692 A1

AMP.3251924- 1931-32.3AMP

5035DEL

16024,1DEL ...... *** * *

Survivald{dose) 2B.FIG Lamin

1

19912DEL

DEL15011

+ + + + + ' + ' t 10p11-12DEL 7931-32AMP 7911-11.23AMP 10023-24DEL 7935AMP Patent Application Publication Nov . 9 , 2017 Sheet 8 of 51 US 2017 /0319692 A1

...... Hinnasticatanto ASpearmanbetweengroupmeans mwand

wiiihiiiiiiiiiiiiiiiiiiiiinnnnnnnn ' * * yurit 050.0005.10-015-0.sensitiveresistant:50.1005.0

19012DEL:16924.1, . . . . . my 10023-34DEL:16924.1 DEL116q24.:10p1112-frammistonument 10011-12DEL:5935

FIG.2C

000000 wiwi. kokoond mining wwwww mmmmmmmm ASpearmanbetweengroupmeans 300000000000000000 nimileinimese won wwwman 6. opowa of mon sono con

000.05-

735AMP:16013,37 7035AMP:1931-32.1 7035AMP:1924-25,3 200AMP:16p13,3 200AMP:1931-32,1 200AMP1924-25.3 7931-32AMP16013,3 7931-32AMP1931.1AMD 7931-32AMP:192425.3 7011-11:23AMP192425,3 50142-151AMP:1931321 50142-15,1AMP:1924253 70111-1.23AMP160133 700-13.28AMP:1931321 50142-15,1AMP16033 sensitiveyesistant Patent Application Publication Nov. 9 , 2017 Sheet 9 0f51 US 2017/ 0319692 Al

FIC. 20 ' }keyifge R - ? 037 A2 ? 0 26 88 ?? 0 .23

.: X ?

:

:

:

:

: : .. . ?. ??????

:

: ?.. ????

. . #

??????? Survivaldedose) } Q . ; . 2 ;, ???? 8 6 3. 8 ? . 0 Q. ; 02 {Q . ) ! ;. . 8 ? . ; 00204 06 3 3 8feast 86 Pancreas 82 ?? 0 . 003 ?? 32 ? , 02 R ? Q &

?? ?? . ? ?

-, , ? * ? 00 0. 204 06 008 10 00 02 04 06 08 _ 000510 15 ? . ? Patent Application Publication Nov. 9 , 2017 Sheet 10 of 51 US 201710319692 A1

FIG. 2E Uterine Colorectal

.- ??? ?????

????? ?? ????? . ?????? ????4 : * . ????

21 20 21 22 23 24 20 21 22 23 J Survival${ } } } { y} , tidose) Patent Application Publication Nov . 9 , 2017 Sheet 11 of 51 US 2017 /0319692 A1

FDRICp-val 0.39200103167 FDRval-ICA -0.309026861 *0.338010861 -0.397000515167 -0.3600013900556 -0.3660008320536 -0.3670008320536-0.3740005550476 -0.3840005550476 -0.392001030167 -0.38500103 -0.412000277 -0.413000277 -0.443000277 mutation FIG.2F Uterine Colorectal

Survivaldidose) SULF1 SMG1 RANBPO BRCA1 MLH1 STAG3 Survivaldidose) SMC2ATR MYH1 SMARCA4CHD1 TFDP1 NIPBL Patent Application Publication Nov . 9 , 2017 Sheet 12 of 51 US 2017 /0319692 A1

. 0 0.0167 0.0167 FOR .

0

0.01671733082-05-0.1733089050167 p-val O.193308x10-0 0167082050.3-177 008x1091883.- -0.199308x10% -0.172308e05 -0.1753,08805 08X10%01923.- -0.182308X10:52 -0.188308x1040 08x1051993,.-0 -0.193308x10% -0.191308x1050-0.1913:08x10 -0.199308x1050 -0.183308x105 -0.1793,08x102 -0.192308X105 -0.1943:08x109 -0.185308x10% -0.181308x105 -0.184308x1050-0.184308x10$ 0.182308x1050 -0.189308x10 C

TIN U

E

Survivald[dosa) UL L T FIG.JA D TUUL . . D

DDRImplicatedin* FLNA*FLNA FLNC LRPIL SULFI CIC PIK3CAPIK3CA PiK??? TPR*NuclearArchitecture p53bindingproteinTP53BP1* TelomeremaintenanceRIFT* SMG1*Nonsensemediateddecay AlternativesplicingRANBP9* ChromatinremodelingSMARCA4* ChromatidcohesionSTAG3* OxidativestressSHCI SignallingCNTNI EzubiquitinligaseHECW1. Cytoskeleton MCM3ARexportnuclearMRNA LipidmetabolismLipidmetabolismLRPI Cell-celladhesionMLLT4 Patent Application Publication Nov . 9 , 2017 Sheet 13 of 51 US 2017 /0319692 A1

FIG . 3B

. PIK3CA 1200 ,. H1047

.

.

... .* * * ResidueNo. . E542 /5 / 6

ogg * p85BD woj . w 0 1 2 3 4 5 6 SurvivalWW d (dose ) Patent Application Publication Nov . 9 , 2017 Sheet 14 of 51 US 2017 /0319692 A1

FIG . 3C

.

.

.

.

. Splice . Missense Nonsense

42887 . Frameshift . PTENPIK3CA p858D Patent Application Publication Nov . 9 , 2017 Sheet 15 of 51 US 2017 /0319692 A1

FIG . 3D

will HEC59 :: PIK3CA : R38C R880 · · · · · PTEN· ve . : · · · R233 * R233 * -50 so kdakDa

SO koa to 37 kDa GAPDH

FIG . 3E

PIK3CA 085BD

o? WAF H&N LUSCStomach Breast143444 Patent Application Publication Nov . 9 , 2017 Sheet 16 of 51 US 2017 /0319692 A1

FIG . 3F

8007

3 4 5 6 Survival d [dose )

FIG . 3G

Mutation Amp KEAP1alterationfrequency ?????????????????????????????????????????????????????????????????????????????????????????????????????? LUADLUSC HS???Y??& N ????Uterine ACC Patent Application Publication Nov . 9 , 2017 Sheet 17 of 51 US 2017 /0319692 A1

Survivald(dose)

FIG.3H

LUAD KALRNMUTGOLGA6BCAMP AMP-SORCS2 CUBDEL MIR570DEL MAFBAMP CUBNMUT PTPRDMUT KEAPTMUT Patent Application Publication Nov . 9 , 2017 Sheet 18 of 51 US 2017 /0319692 A1

FIG . 31

NFE2L2 600 , .NoResidue

ETGE.. 0 1 2 3 4 5 6 Survival d (dose ) Patent Application Publication Nov . 9 , 2017 Sheet 19 of 51 US 2017 /0319692 A1

FIG . 4A

SQSIMI @ NQO1 expressiongeneorJennnsuogerpesYM

- 0 .31 .???? 5000 10000 15000 20000

FIG . 4B Spearman : 0 .61

SOSTMIMRNA #

- ???4 ???????????????????????????????????????????????????????????????5 6 7 8 9 10 11 12 13 14 NQ01 mRNA. Patent Application Publication Nov . 9 , 2017 Sheet 20 of 51 US 2017 /0319692 A1

Songs

:: . . . Tvival 0 .57 . NFE2L2: 0 . 4

... activity NFE2L2activity: ....

- 0 % ????Neuig LU?CA EsophagusCRC t??????? Pleura! Breast LUAD FIG . 4C FIG . 40 Patent Application Publication Nov . 9 , 2017 Sheet 21 of 51 US 2017 /0319692 A1

FIG . 4E FIG . 4F

P = 0 .023 2104. R2 = 0 . 20 NFE2L2activity SOSTMIMRNA NE os1 2 3 4 5 T? 3 4 5 6 Survival d (dose ) Patent Application Publication Nov . 9 , 2017 Sheet 22 of 51 US 2017 /0319692 A1

FIG . 4G

.

.

.

. . SQSTM1 over -express . . No altern . No alteration

.

.

.

.

. 80 % . . ??? .

.

.

.

.

.

.

.

60 % . . ow .

.

. . h .

.

.

40 % . .

. Surviving 1.

0 20 40 60 80 100 Months Survival total cases deceased median survival z > 1 .51 21 15 1 13 .63 z < 1. 5 101 T XXX 44 30 . 58 Patent Application Publication Nov . 9 , 2017 Sheet 23 of 51 US 2017 /0319692 A1

FIG . 5A

17912 - 22 AMP 17021. 2 - 21. 3 DEL 9p21- 24 DEL 1p36 . 13 DEL * *

***

*** * *

*** * 6q25. 1 AMP

FIG . 5B ERBB2 presupWisku* tie

unempatk ** +

isras ** * 17 www

2 w * * wwwwwwwwwwwwwwww 444444444444444444 ** Fu of s WNYW HARRERA **50 : from **** * * ** * 8 **** *

** **** * - 2 - 10 1 2 3 4 SCNA Patent Application Publication Nov . 9 , 2017 Sheet 24 of 51 US 2017 /0319692 A1

X Aga

100$ , waque:jeadDi. . SO-DISZ2890 100SISL8190 10090-920907 100SISL5090 SOISLVOL0 0-382089'. 0SUOISE8890SOOSLE90 sorsUZ9' 190 1005051690 0190SAISL100 100SO-IST.209'0 SO-S4999'0 50S4sso 49090

(esoooJAIS 2300$paziewION : ...... OSOL

09 WRX hasano FOXPIEJKgPajeriMOP630 dasfeneruponudiswoeg OURO$93XQ20godizans29919umencalonsaugo VEVO*9)umviseasonsuas XXX SzFUKsUk0LOE/LULULIHEMIBAJOnuWasBondueDUEFEHTIQZbe70gumslide Soko200mlSuitUOO NELLOUMOXouxvodaparacadasou9 BoomSS WWW6WYORK ndoe

oseu Patent Application Publication Nov . 9 , 2017 Sheet 25 of 51 US 2017 /0319692 A1

FIG . 5D

.

.

.

. to en Ham water inthe ARMRNA . ething into the :* .

.

. oto ¿ À ó survivalSurviva d(3 .dose Y . . . )

FIG . 5E

.

* ARMRNA *

* - - * - I OsteosarcomaProstate Breast GliomaMesothelioma Ewing 's Patent Application Publication Nov . 9 , 2017 Sheet 26 of 51 US 2017 /0319692 A1

. : : : Dage . . .

...... -/+ CellLineER(ERBB2AR??????????????????????????????????????????????? HMC18 BT474ZR7530 HCC202MDAMB453- INCAP -100kDaMCF7 CAMA1kDa75 -150kDa -37kDa FIG.5F

UNCap - HCC202CAMA1 ZR7530BT474 MCF7 HMC18 Actin- .ERBB2 Patent Application Publication Nov . 9 , 2017 Sheet 27 of 51 US 2017 /0319692 A1

FIG . 6A

8 H1 KH TE RLUi ** * AOO : LIIIIIIIIIIIIIIIIIIIIIII ...... ió- i

W

? A 4 GY ? RLU HH **ttt Å DH

: 0 .0 0 . 5 1 .0 1 . 5 2 .0 log ( ENZ ) (UM ) Patent Application Publication Nov . 9 , 2017 Sheet 28 of 51 US 2017 /0319692 A1

FIG .. 6BOB

DENZ 2 .47 ENZ + IR 20 - 3DHT TijdWOHT # IR . .27 .

. - Rov ona saninnananas

.:

- 0 . 4

n 0 . 2 r00

i

..

t HMC18W M???7 BT474 ZR7530CAMA1 HCC202 MDAMB453 Patent Application Publication Nov . 9 , 2017 Sheet 29 of 51 US 2017 /0319692 A1

FIG . 6C Neutral

.. TailMoment

DHTIR

* F** DHT : : Neutral

6 TailMoment

. ENZAalENZ +IR Patent Application Publication Nov . 9 , 2017 Sheet 30 of 51 US 2017 /0319692 A1

FIG . 6D HMC18 MDAMB453 hours atar R : . . 1 6 6 24 24

YHZAX - $ 10a HDACI -- - * 753 *60 O * 75 DX Actin w $ 7 kba m 37 KOM men OMO weinionin OM SO * * * ENZ * * * ENZ

%HRAXIntensity *H2AXIntensity wewing * ornament Hours : MDAM3453 04 - 2 1 1 6 6 24 28 1 6 6 24 24

w 1 $ 408 w- 15 koskoa - 75 kDa -x 75S XDx2 - $ 7 koa w 37 ko 0359 *2 * 30 OHY 0 . 15 KisvanulXVZMLX XXH2AXIntensity 0 . 10 . Warning : seinen neurria o 10 15 20 25 Hours Hours Patent Application Publication Nov . 9 , 2017 Sheet 31 of 51 US 2017 /0319692 A1

FIG . 6E

MDAMB453 0 Gy3 GY • ADT DHT © ADT DHT

YDNAPKcs - 250 kDa mm 75 kDa

- 37 kDa Patent Application Publication Nov . 9 , 2017 Sheet 32 of 51 US 2017 /0319692 A1

FIG . 6F

. xenograft M

Inguinal mammary gland 200 mm

Mock Enzalutamide. . . 250 mm 250 mm Mock RT Mock RT Patent Application Publication Nov . 9 , 2017 Sheet 33 of 51 US 2017 /0319692 A1

FIG . 6G

. e VOXX . von ENZ ( 15 mg /kg ) wa ENZ (25 mg/ kg ) . W . YRT

. NZ ( 15 mg /kg ) + RT1 61 . ENZ (25 mg /kg ) + KT . TI

.

. AHHHHHH

. OWNjOnouni H

aunoJOLUL . * * *

.

.

.

. 3 12 15 18 21 365 12 15 18 21 . . Days Patent Application Publication Nov . 9 , 2017 Sheet 34 of 51 US 2017 /0319692 A1

FIG . 6H Control2000. ENZ XRT ENZ + RT

trimit

1 cm

FIG . 61

117 - MockANZ 15 mg/ kg osowna ENZ ( 15 mg/ kg ) * AT mm* * VOCKENZ ( 25 mg /kg ) MRT 10

*** ma * *** **** * *83 : * RelativeWeight .. 0 . 9 WWW

0. 85 Patent Application Publication Nov . 9 , 2017 Sheet 35 of 51 US 2017 /0319692 A1

FIG . 7

MCH211 VORCON * WOR1651 1 OVKAIE w xxxor w * Mi MUQXQX

8 1 **

8 . Control% %Control *

b

0 . 001muna o 0 2 4 6 8 10 endelhedhimbuwielbiamanten0 2 4 6 8 10 Patent Application Publication Nov . 9 , 2017 Sheet 36 of 51 US 2017 /0319692 A1

FIG . 8

Uterine

5 5

- 0.5

-

-

-

- ** 4 - 4 - 0.4 *

-

-

- 0.3

Survivalddose) - 3 3

-

density 0.2 SampleQuantiles *

2 * 0.1 * *

0. * 1 2 3 4 5 - 2 - 1 0 1 2 Survival de Theoretical Quantiles Colorectal

6 6

-

-

- **** *

-

-

5 - 5 F--

- * 4 4

Survivaldedose 3 3 Azisuap SZOOCOSTOOTOsoo000 SampleQuandles ** * 2 2 *

1 1 DT 0 1 2 3 4 5 6 - 7 - 1 0 1 2 val a dose Theoretical Quantiles Patent Application Publication Nov. 9, 2017 Sheet 37 of 51 US 20170319692 Al

FIG. 9A

3 { 00 FLNC 23Q015 ??? , 03 . 230 % %% % ResidueNo. ::: Q00

(33 ) H

5 { {} . :: :: : =

er

3 . w 0 1 2 3 4 5 6 0 1 2 3 4 5 6 \ $$ ? } { } Survival d [ dose )

* £0. ." 2032 { { } } , 35 } * } * *

ResidueNo. 1030 * 20 5 } } G

‘. 0 } 3 4 5 6 0 } 2 3 4 5 6 Survive?; 3( 348 __ $v??? 33258 Patent Application Publication Nov . 9 , 2017 Sheet 38 of 51 US 2017 /0319692 A1

FIG . 9B MLLT4 RANBP9 2000 2000 , ???????????????.???????? ResidueNo. 1500 3500 1000

500 ?.??????????????? 500 . CUTH

paanong nangangananananananananananang : - 2 3 4 5 2 3 4 5 Survival oídose ) Survival d [dose )

: FLNA HECW1 3000 d, 2000 1500 HECT ResidueNo. 2000 1000 .. 1000 ccococodococoote 500 ** Bo * 2 3 4 5 6 75 % Survival o [dose ) Patent Application Publication Nov . 9 , 2017 Sheet 39 of 51 US 2017 /0319692 A1

FIG . 9C 2500 , 4000

: :: :: ResidueNo 10000 ... . # 1000 WEMMANAMAN

Survival odoso ) Survival d {dose 3500 5000 TPS3BP1

ResidueNo. 1000

# 1000 #

inconianie hinnainningpinninginancimin Suviai didoso ) Survivalodoo ) Patent Application Publication Nov . 9 , 2017 Sheet 40 of 51 US 2017 /0319692 A1

FIG . 9D

4000 . 1000 SULFI ResidueNo.93000

#

. . . . 0001 * * #

w BS 6 Survivai doset Savival ddose

1200 . . STAG3. . CNTNI 800 . ResidueNo. wwwwww View

Survival dd069 ) Survival o doce ) Patent Application Publication Nov. 9 , 2017 Sheet 41 of 51 US 2017/ 0319692 A1

FIG . 9

3 4 Timininimin these MISS ???????????????????????????????????????????? UNPHAHAPHENHANELPHPWM ( ?? ) Patent Application Publication Nov . 9 , 2017 Sheet 42 of 51 US 2017 /0319692 A1

FIG . 10A

S * * Cavai FOR Gernssakin - 0 . 349 25xi o mis asme - 3. 3 7 .536303 Sama sem - 0 328 7 .58305 0 .0358 SNA S3' RNA deesdeks - 0 . 313 SAR 0195 { ???! ! - 0 . 30 % 6 . 0 * * 304 0136 Nansoxsa mediatos do - 0 . 298 0 .585 2256 Genes requiset A1 . - 3. - 0 . 328 35x105 OOSSS Coxasn ssceribly & maintenarse - 0 313 4 30x30 * 0 . 235 Csomatia ai 0313 « So 0 . 85 uweuw HOACA S - 0, 531 9. 257 0 : 35 5282 -82235SP ?CHER - 0 . 324 2 . 5x10 .0769 Seekere in yox XEA ? * XXXype mesmo - 0 , 310 5 .250 0 . 125 360 Seres coexgressed with 88061X CHRY - 0 . 7 6 . 7830 0130 C28ccio se ne R orma oblasti - 0 . 5 7 .540 0 . 190 /400 GBS e poi 0 .83 0 .03 . 20 0 . 196 G2/ V terpoint 0 . 277 0 . 00298 0 . 15 % Patent Application Publication Nov . 9 , 2017 Sheet 43 of 51 US 2017 /0319692 A1

FIG . 10B

eval FOR Tors 6 .322 75140 * ? 2 .385 2 .5X19 0 . 03si SAXASTAT 0 . 358 1 . 50xigt 0 .0251 ?{}??? AXT- FOXO3 Q . 49 1. 50xim 1 .012 GESI 3 2 .23315 " V0051 HER2 6 .34 3. 900 * 0 . 035 .. Bali828 0 .334 3 .72 0 . 0351 kex 0 .326 45X104 07351 WWWWWWWWWWWWWWWWW 239. 0 .324 . 4 . 30X0 0 Loxt netoxon 0 . 38 1 . 5010 * 8015 : Wawane Umsat 0 .391 39x10 03S . tama 2 . 390 1 .30xim 1 0231 yermeny 600 1 ,50x1 * 40151 PARK send moness 39 2 .30x10 0 .0357 Vorspeilolinens xatea etaxon 0359 .50x1n * 40151

W muasie 1 348 1. 50x 80191 WWWW Kdavation to hypoxis 2 .329 4 .56x104 0 .0031 USIN UV expose 6 . 327 4 . Sexi04 0 . 035 .1 Qidalive stress : 0 .328 .50x17 6061 xaxed atosp e rsosure 0925 . 4 . S0x144 0 .035i W Sexposure * 0 .32 % 4 .50X10 " 0 .0051 Logemene Qianwespe 0329 450xíny * 6151 Patent Application Publication Nov . 9 , 2017 Sheet 44 of 51 US 2017 /0319692 A1

FIG . 11

100 % KEAP1, NFE2L2w , CUL3 No alteranon LUAD P = 0 .028 Surviving winninn 20 %

* 0 40 80 120 160 200 240oooooooooooooooooooooooooooooooooooooooooooooo000000000000000000000000000000000000000000000000000000000 Months Survival total cases deceased median survival # with alteration 46 22 32. 70 # without alteration 1571 411 45 .31 Patent Application Publication Nov . 9 , 2017 Sheet 45 of 51 US 2017 /0319692 A1

FIG . 12

2 , 3 .09 O No CNA data ? pad 2. 5 O Gain 2 .04 o Shallow Deletion 0 Amplificadon 1 . 5 . { } { ?? ????????f 1 .03 0 . 5

Q 0 0 . 590 0 0

2 2 10 12 14 AR ,MRNA * * * * (log2 ) sont

** ** * * # # *** 8 # # # * # # # . # ** * * * * #

AR,MRNA(log2) # . ** # # :

* * ? 4 10 12 14 16 18 ESR1, mRNA (log2 ) Patent Application Publication Nov . 9 , 2017 Sheet 46 of 51 US 2017 /0319692 A1

FIG . 13

folosite i f 1 HMC18MCF7 BT474 ZR7530 CAMA1 HCC202 MDAMB453 ** LNCAP AR - - - - * 100 kDa Patent Application Publication Nov. 9 , 2017 Sheet 47 of 51 US 2017 /0319692 A1

FIG . 14

HMC18MDAMB453 tunity BT474 yErbB2 150 koa s 50 kDa Actin 37 KDa Patent Application Publication Nov . 9 , 2017 Sheet 48 of 51 US 2017 /0319692 A1

FIG . 15A

.

MDAMB453 BT474 HMC18 0. 15 i ? 3 Patent Application Publication Nov . 9 , 2017 Sheet 49 of 51 US 2017 /0319692 A1

1.5 &3GY & AGY *** * * . .. VO Huistato 1.0

0005101520 *** MCF7 0.5 MDAMB453 0. 1.01 -0.05 0.5 FIG.15B log(ENZ)UM A2GY 2.0 GY3A

*** * * 1.5 HMC18 0.5101520 BT474 1.0 mm 0.5

0. 0. . ? $ 0.5 Patent Application Publication Nov . 9 , 2017 Sheet 50 of 51 US 2017 /0319692 A1

FIG . 16

Hours after iRi 0 .5 0. 5 2 3 3 Neutral ENZ : iehen . oblet - * - YH2AX ** 15 x08 * 7 $ Da 20 I HDACI Actin w 3 % TailMoment 23 — OMSO KysusuXVZH * * * ENZ ENZTRE ENZIIR Hours Patent Application Publication Nov . 9 , 2017 Sheet 51 of 51 US 2017 /0319692 A1

FIG . 17

Adjuvant Radiotherapy

sti

* Primary Sii *** * * * * * * * * * A** ** * surgery 37:11 : 7: 1** turintinin

** * ** ** *** * * * * * * * * * *** ** *

??? ? ? * ** printing 5

Neoadjuvant Surgery on Radiotherapy tingi. three ***

!** * 111

A **** * * * * * * * * * * * * ** ! ! ! ! ! ! !! 1 cincinin *** * * * *** ** * *

**

*** * * * US 2017 /0319692 A1 Nov. 9 , 2017

ANTI- AR AGENT AND RADIATION radiation therapy is performed at least 1 hour after the THERAPY FOR ANDROGEN RECEPTOR treating with the anti - androgen receptor agent. In further POSITIVE CANCER embodiments , the radiation therapy is performed at least 1 - 24 hours after the treating with the anti -androgen receptor [ 0001] The present application claims priority to U . S . agent ( e . g . , 1 . . . 4 . . . 8 . . . 12 . . . 16 . . . 20 . . . 24 . . Provisional application Ser . No . 62 /326 ,203 , filed Apr. 22 , . 28 hours ) . In certain embodiments , the treating with 2016 which is herein incorporated by reference in its anti- androgen receptor agent is at least 1 - 2 months before entirety . the radiation therapy . In other embodiments , when radiation therapy is administered , additional anti - androgen receptor FIELD agent is administered at the same or about the same time. In [ 0002] Provided herein are compositions, systems, kits , further embodiments , after such radiation therapy is con and methods for treating cancer in a subject with androgen cluded , and additional 1 day . . . 1 week . . . 1 month . . . 2 receptor positive cancer cells by sensitizing such cancer months of anti -androgen receptor agent treatment is pro cells with anti - androgen receptor therapy ( e . g . , Enzaluta vided . In particular embodiments , the radiation therapy is mide or anti- androgen antibody ) , and then treating with performed at least 1 . 1 - 3 days after the treating with the radiation therapy. anti- androgen receptor agent . In particular embodiments , the anti- androgen receptor agent is administered to the subject BACKGROUND in a dosage between 1 mg/ kg to 35 mg /kg ( e . g . , 1 . . . 5 . . . 10 . . . 23 . . . 30 . . . or 35 mg /kg ) . In other embodiments , [0003 ] Clinical radiotherapy has made significant the anti- androgen receptor agent is administered to the advances since its inception , growing into a tertiary spe subject in a dosage of about 10 - 15 mg/ kg . In certain embodi cialty with significant contributions to curative and palliative ments , the present disclosure is employed before and /or after treatments of cancer and health care cost '. A major limitation surgical removal of a subject ' s tumor . to its appropriate application , however , has been the lack of [ 0008 ] In further embodiments , the determining comprises measurable biological indicators , or biomarkers , that can receiving or reviewing a report that the subjecthas androgen reliably identify patients with cancers that are more or less positive cancer cells . In additional embodiments , the deter likely to respond to these treatments2 , 3. mining comprises performing an in vitro assay on a sample [ 0004 ] Advances in genomic technology have enabled a from the subject . In some embodiments , the anti- androgen cataloguing of cancer genes that has resulted in the identi receptor agent is an androgen receptor antagonists selected fication of genetic alterations that contribute to oncogenesis from the group consisting of: flutamide , nilutamide , bicalu and / or tumor progression and in some cases has led to tamide , enzalutamide, apalutamide , cyproterone acetate , significant therapeutic advances4 , 5 , 6 , 7 . In contrast , X - rays megestrol acetate , chlormadinone acetate , spironolactone , and DNA -damaging drugs are delivered based on the site of canrenone , drospirenone, ketoconazole , topilutamide ( fluri anatomical origin of disease and do not currently take into dil) , abiraterone, and cimetidine. In some embodiments , the account the genetic complexity that may regulate therapeutic anti- androgen receptor agent is a selective androgen receptor response . modulator (SARM ) selected from the group consisting of: Enobosarm ( Ostarine , MK - 2866 , GTx -024 ), BMS- 564 , 929 ; SUMMARY LGD -4033 (Ligandrol ) , agent in U . S . Pat . No. 7 ,605 , 152 [ 0005 ] Provided herein are compositions , systems, kits , SARM ( 5 - 3 / 5 - 6 ) ; AC - 262 ,356 ; JNJ- 28330835 ; LGD - 2226 ; and methods for treating cancer in a subject with androgen LGD -3303 ; S - 40503 ; and S -23 . In additional embodiments , receptor positive cancer cells by sensitizing such cancer the anti -androgen is an antibody, or fragment thereof, to the cells with anti- androgen receptor therapy ( e . g . , Enzaluta androgen receptor. mide or anti -androgen antibody ) , and then treating with [ 0009 ] In particular embodiments , the androgen receptor radiation therapy . positive cancer cells are a type of cancer selected from the [ 0006 ] In some embodiments , provided herein are meth group consisting of: breast, prostate, colon , leukemia , brain , ods of treating cancer comprising: a ) determining that a bone , skin , liver, pancreatic , stomach , and lung . In other subject has androgen receptor positive cancer cells ( e . g . , embodiments , the radiation therapy comprises subjecting the breast cancer cells ) ; b ) treating the subject with an anti subject to X - rays , gamma rays, and or charged particles . In androgen receptor agent; and c ) treating the subject with certain embodiments , the cells are triple negative breast radiation therapy , wherein the radiation therapy is performed cancer cells (estrogen receptor -negative , progesterone at least 30 minutes ( e . g . , at least 12 hours or 3 days ) after the receptor - negative and HER2- negative ) . treating with the anti- androgen receptor agent, and wherein [0010 ] In further embodiments , provided herein are sys the treating causes at least a portion of the androgen receptor tems comprising : a ) a radiation therapy device , and b ) an positive cancer cells to die . In certain embodiments , the anti- androgen receptor agent. In some embodiments , the treating with anti - androgen receptor agent is conduced for a radiation therapy device is configured to emit X - rays , day . . . a week . . . a month . . . or two months . In certain gamma rays , or charged particles for cancer treatment . In embodiments , the subject is a human subject ( e . g ., male or other embodiments , the anti -androgen receptor agent is an female ) . In other embodiments , the subject is an animal, androgen receptor antagonists selected from the group con such as a dog , cat, horse , cow , or other domesticated animal . sisting of: flutamide , nilutamide , bicalutamide , enzaluta [ 0007 ] In certain embodiments , the anti -androgen receptor mide , apalutamide , cyproterone acetate , megestrol acetate , agent sensitizes the androgen positive cancer cells to the chlormadinone acetate , spironolactone , canrenone , radiation therapy , such that a higher proportion of the cancer drospirenone , ketoconazole , topilutamide ( fluridil ) , and cells are killed by the radiation therapy than without the cimetidine . In particular embodiments , the anti - androgen anti - androgen receptor agent. In other embodiments , the receptor agent is a selective androgen receptor modulator US 2017 /0319692 A1 Nov. 9 , 2017

(SARM ) selected from the group consisting of: Enobosarm mutation . ( b ) Scatter plot of integral survival and amino acid ( Ostarine, MK -2866 , GTx -024 ) , BMS- 564, 929 ; LGD -4033 position for cell lines containing mutations in PIK3CA . (c ) ( Ligandrol) , agent in U . S . Pat. No. 7 ,605 , 152 SARM ( 5 - 3 / Association between radiation response and mutation in 5 - 6 ) ; AC - 262 , 356 ; JNJ- 28330835 ; LGD - 2226 ; LGD -3303 ; PIK3CA and PTEN in uterine carcinoma . ( d ) YAKT, AKT, S -40503 ; and S - 23 . and GAPDH levels in two uterine cancer cell lines with p85 [ 0011 ] In certain embodiments , the systems further com BD mutations . ( e ) Frequency of PIK3CA and PIK3CA p85 prise : c ) a report that a subject has androgen receptor cancer BD mutations as annotated by TCGA ; organized from left to cells , or c ) androgen receptor positive cancer cells . In some right by frequency of mutations in p85 BD . ( f ) , ( i ) Scatter embodiments , the anti -androgen is an antibody, or fragment plot of integral survival and amino acid position for cell lines thereof, to the androgen receptor . containing mutations in KEAP1 and NFE2L2 . ( g ) KEAP1 alteration frequency by lineage , and histology where appro DESCRIPTION OF THE FIGURES priate , as annotated by TCGA . Organized from left to right [ 0012 ] FIGS. 1A - 1E . Variation in cancer cell line survival by frequency of KEAP1 mutation . ( h ) Association between after radiation - induced damage. (a ) Distribution of cancer integral survival and genomic features in lung adenocarci types profiled by lineage . ( b ) The high - throughput platform noma . Red bar represents a copy - number change or muta accurately profiles cancer cell lines . Integral survival was tion in the corresponding gene . calculated for each cell line profiled by the high - throughput [0015 ] FIGS. 4A -4G .Gene expression changes regulating platform ( n = 1 ( top ) , n = 2 (bottom ) ] and by clonogenic sur oxidative stress response are associated with radiation resis vival measurements (n22 ). Scatter plots , linear regression , tance in several cancer lineages . ( a ) Correlation of NQO1 and R2 values were calculated comparing the integral sur and SQSTM1 expression with radiation resistance . Spear vivals of the high - throughput platform to clonogenic sur man correlation coefficient was calculated between gene vival. Data are expressed as the meansus. e . m . (c ) Integral expression and integral survival values . Correlation was survival is displayed by column scatter plot separated by then plotted relative to correlation rank . ( b ) Relationship lineage and histology where appropriate . HGG , high - grade between NQO1 and SQSTM1mRNA expression in CCLE . glioma. LUSC , lung squamous cancer, LULC , lung large ( c ) SSGSEA association between NFE2L2 signature score cell, SCLC , small - cell lung cancer, PNET, primitive neu and integral survival. ( d ) NFE2L2 is frequently activated in roectodermal tumors. ( d ) Histogram , probability density hepatocellular (HCC ) and biliary tumors . A column scatter function , and Normal Q - Q plots analyses of calculated plot of NFE2L2 signature score for 967 cell lines in the integral survival of 533 cell lines ( “ All” ) , 89 non - small cell CCLE organized by disease site and histology where appro lung cancer cell lines (“ NSCLC ” ), and 39 lung adenocarci priate . Solid bars represent the mean in each category . noma cell lines (“ LUAD ” ). (e ) Correlation of response Dashed line represents themedian across all CCLE lines . ( e ) between radiation and compounds . Spearman correlation NFE2L2 activity scores and ( f) SOSTM1mRNA levels from coefficient was calculated between integral survival values HCC and biliary cancer cell lines were plotted as a function after exposure to radiation or 481 compounds. Correlation of radiation integral survival. ( g ) Kaplan -Meier survival was then plotted relative to correlation rank . Some chemo analysis curve calculated from 122 hepatocellular cancer therapeutic agents in clinical use are shown . patients from TCGA ; cut- off = z1. 5 . z = + 0 . 8 or greater dem [0013 ] FIGS . 2A -2F . SCNA changes are associated with onstrated a statistically significant difference in overall sur survival after radiation - induced damage. ( a ) Plots of vival by log - rank test. fSCNA , integral survival, and number of mutations per [0016 ] FIGS. 5A -5F . Genes associated with survival after sample . ( b ) The top 50 probes that correlate with radiation radiation - induced damage in breast carcinoma. ( a ) The top resistance ( left) and sensitivity ( right) are shown . Radii 50 copy- number probes associated with radiation resistance ( single probe) or sectors (multiple probes ) correspond to in breast adenocarcinoma are shown . Radii ( single probe ) or chromosome positions . Each radius represents a distinct sectors (multiple probes ) correspond to chromosome posi probe that mapped to the designated chromosome position . tions . ( b ) ERBB2 amplification is associated with survival ( c ) Individual SCNA can regulate the response to radiation after radiation - induced damage . Three - dimensional scatter directly . We correlated radiation survival with the expression plot of integral survival, ERBB2 copy - number, and ERBB2 of genes within the altered segments and compared the mRNA expression . ( c ) SSGSEA identifies gene sets that means of the coefficients by pair -wise analysis, resistant correlate with resistance to radiation . Heatmap of ssGSEA versus sensitive . Spearman means for alterations depicted in scores (left = positive, right= negative ). A subset of the top 27 (b ) were analyzed by ANOVA and Tukey Contrasts . 95 % gene sets is shown. Genes sets depicted in red font are confidence level intervals for each pairwise comparison are associated with androgen signaling. ( d ) Scatter plot and shown . ( d ) Scatter plots , linear regression , and R2 values of linear regression of AR mRNA levels and radiation integral the integral survival and fSCNA by lineage . ( e ) Scatter plot survival in breast cancer. ( e ) AR is frequently expressed in and linear regression of integral survival, fSCNA , and the multiple cancer lineages . A box and whiskers plot of AR nunumber of mutations (MUT ) in uterine and colorectal car expression for 967 cell lines in the CCLE organized by cinoma. ( f ) Heatmap of integral survival ( left = resistant, lineage . Whiskers representminimum and maximum values . right = sensitive ) and gene mutations in uterine and colorectal ( f) (Left ) Relative AR , ER , and HER2 protein levels in carcinoma cells . Black bar represents a mutation in the whole cell extracts from breast cancer cell lines and LNCap , corresponding gene . a prostate cancer cell line . (Right ) Annotation of expression 10014 ) FIGS . 3A - 31. Mutations in genes associated with based on Western analysis include: - ( no observable expres distinct cellular functions correlate with survival after radia sion ) , + ( expression ) , or - / + (faint expression was observed tion - induced damage . ( a ) Top 19 genes that when mutated with longer exposure to film ). are associated with radiation sensitivity are organized by [0017 ] FIGS. 6A - 61. AR activity regulates the response to biological functions . The bars represent samples with a DNA damage in breast carcinoma. (a ) ( Top ) Cells were US 2017 /0319692 A1 Nov. 9 , 2017 cultured in steroid - deprived media - / + DHT for 24 hours throughput platform data are expressed as the means of at and then treated with IR : mock ( 0 ) , 4 Gy, or 6 Gy. Cells were least four replicates . Clonogenic survival data are expressed then supplemented with hormone -proficient media at 48 as the meansus . e . m of at least three experiments . hours post- IR . Cell number was determined on day 14 - 21 . [0019 ] FIG . 8 . Non -Gaussian variation in uterine and ( Bottom ) Cells were cultured in hormone -proficient media colorectal carcinoma cell line survival after radiation -in for 24 hours without or with enzalutamide ( ENZ ) and then duced damage . Box plot and whiskers plots , histogram , treated with IR ; mock ( 0 ) or 4 Gy . Error bars represent probability density function , and Normal Q - Q plots analyses normalized s . e . m of at least three experiments . ( b ) MCF7 , of calculated integral survival for cell lines derived from the BT474 , CAMA1, HCC202 , and MDAMB453 cells were uterine and colorectal carcinomas . Whiskers represent mini treated with ENZ , ENZ + IR , DHT, or DHT + IR as in ( a ) . mum and maximum integral survival values in each distri HMC18 cells were treated similarly but were profiled by bution . The horizontal line in the boxplot represents the clonogenic survival assays. Error bars represent normalized median . Histogram analysis of the uterine lineage suggests s . e . m of at least three experiments and the Student' s t - test a bimodal distribution . was used for statistical analysis . * , P < 0 .05 . (c ) Neutral [ 0020 ] FIGS. 9A - 9E . Mutations in a subset of genes Comet assay of MDAMB453 cell line , showing increased associated with radiation sensitivity demonstrate domain double - strand breaks when cells were irradiated in steroid selectivity in conferring sensitivity . Scatter plot of integral deprived conditions ( left ) or after 24 hours of treatment with survival and amino acid position for cell lines containing 20 uM of ENZ ( right) . Error bars represent s .e . m of at least mutations in the designated genes . Protein domains pre three experiments and the Student ' s t -test was used for dicted by InterPro and / or ExPasy are shown . statistical analysis . * , P < 0 .05 . ( d ) Cells were cultured in [0021 ] FIGS. 10A - 10B . SSGSEA identifies gene sets that hormone - proficient (FBS ) media for 24 hours - / + ENZ and correlate with radiation sensitivity ( a ) and resistance ( b ) . then treated with IR : mock ( o 3 Gy (HMC18 and Heat map of ssGSEA scores . Top gene sets, organized by MDAMB453 ) , or 2 Gy ( C4 - 2 ) . Cells were cultured in biological processes , are shown. steroid -deprived media for 48 hours , treated - / + DHT for 24 [ 0022 ] FIG . 11 Kaplan -Meier survival analysis curve cal hours, and then treated with IR : mock ( 0 ) , 3 Gy (HMC18 culated from lung adenocarcinoma patients from TCGA and MDAMB453 ) , or 2 Gy (C4 - 2 ) . YH2AX , HDAC1, and separated by NFE2L2 (mutation or CNA ) , KEAP1 (muta actin levels were measured at the indicated time points . tion or CNA ) , and CUL3 homozygous deletion versus none Relative intensity of yH2AX was calculated by ImageJ64 . of these alterations. There was a statistically significant ( e ) MDAMB453 cells were cultured in steroid -replete , ste difference in overall survival by log - rank test ( P = 0 .028 ) . roid -deficient , or steroid - deficient with 1 nM DHT for 24 [0023 ] FIG . 12 . Correlation of AR and ESR1 expression . hours , then treated with irradiation . Cells were harvested and (Panel a ) AR protein levels correlate with mRNA levels expression of yDNAPKcs was analyzed . Androgen depriva (Pearson r = 0 .615 ) . AR overexpression is mostly observed in tion therapy (ADT ) . ( f ) Schematic of treatment arms. ( g ) diploid tumor samples , suggesting that increased AR expres MDAMB453 cells were orthotopically injected into the sion is mainly regulated by increased gene transcription mammary gland of NSG mice and block randomized into rather than gene amplification . AR protein was measured by one of four treatment arms as shown . Tumor volume was reverse phase protein array (RPPA ) and mRNA measured by measured daily . Error bars represent normalized s . e . m of at RNA -Seq from 1098 breast cancer patient samples profiled least seven mice in each treatment arm . * , P < 0 . 05 compared by TCGA and analyzed by cbioportal. ( Panel b ) Scatter plot with all other treatment conditions based on ANOVA and and regression of AR and ESR1 mRNA expression are Tukey Contrasts . * * , P < 0 .05 for interaction based on two shown (Spearman r = 0 .51 ) . Breast tumors that express AR way ANOVA . ( h ) Pictorial depiction of representative mice and not ESR1 were determined using a cut- off of themedian from each arm of cohort 2 at the end of treatment. This for AR and ESR1 expression and delimited by a dashed cohort received ENZ at 25 mg kg - - . ( i ) Average weight for circle . RNA - Seq expression values were calculated by each arm in both cohorts was measured weekly . Data are expectation maximization (RSEM ) and scaled by log 2 expressed as the means = s. e . of at least seven mice in each transformation . treatment arm . [0024 ] FIG . 13 . Relative AR protein levels in whole cell [0018 ] FIG . 7 . High - throughput profiling of survival after extracts from breast cancer cell lines and LNCap , a prostate exposure to y - radiation . (Panel a ) Cells were plated in at cancer cell line . Increased time of film exposure reveals that least 7 wells in a 384 -well plate at cell densities ranging MCF7 cells express low levels of AR (compare to FIG . 11 ) . from 25 - 225 cells in 70 uL of media per well for 24 hours HMC18 cells do not express AR protein . and then treated with IR : 0 (mock ), 1 , 2 , 3 , 4 , 5 , 6 , 8 , or 10 0025 ]. FIG . 14 . ErbB2 is overexpressed but not activated Gy. Luminescence - based detection of cellular ATP , a surro in MDAMB453 cells . yErbB2 and actin levels were mea gate measure of cell number, was performed after 9 days sured in whole cell extracts from HMC18 , MDAMB453 , after radiation treatment. Proliferation fraction as a function and BT474 cells . of dose for NCIH211 and BT181 after exposure to y- rays is [0026 ] FIGS . 15A - 15B : Clonogenic survival measure shown . Integral survival (shaded area ) was estimated by the ments in breast cancer cells . ( a ) Clonogenic survival mea trapezoidal approximation of area under the curve . Data are surements of breast cancer cells HMC18 , MCF7, BT474 , expressed as the means = s . d of at least four replicates . ( Panel and MDAMB453 . Colony number was determined on days b ) Correlation between high - throughput platform ( n = 1 ) and 7 - 21 . Data are expressed as the meansus . e . m of at least three clonogenic survival measurements ( n > 2 ) across individual experiments . ( b ) Clonogenic survival measurement of doses . The proliferating fraction ( single - run ) and clonogenic enzalutamide -mediated radiosensitization in cell lines survival ( for each cell line, n > 2 ) were plotted as function of HMC18 , MCF7 , BT474 , and MDAMB453 . Cells were dose . The proliferation index closely approximated clono cultured in hormone - proficient (FBS ) media for 24 hours genic survival , especially within the G15 , range . High without or with enzalutamide (ENZ ) and then treated with US 2017 /0319692 A1 Nov. 9 , 2017

IR ; mock ( 0 ) or radiation . Colony number was determined that correlate with radiation survival , revealing new insights on days 7 -21 . Data are expressed as the means = s . e . m of at into the genetic basis of tumor cellular response to DNA least three experiments . damage . These results demonstrate the diversity of tumor [0027 ] FIG . 16 : AR regulates DNA damage and repair in cellular response to ionizing radiation and establish multiple AR , ER positive cells . ( a ) BT474 cells were cultured in lines of evidence that new genetic features regulating cel hormone -proficient ( FBS ) media for 24 hours - / + ENZ ( 20 lular response after DNA damage can be identified . uM ) and then treated with IR : mock ( 0 ) or 4 Gy. YH2AX , HDAC1, and Actin levels were measured at the indicated Methods time points . Relative intensity of yH2AX was calculated by [0033 ] Cell Line Validation . ImageJ64 as described ( ref) . ( b ) Neutral Comet assay of 100341 Cell lines from the Broad Biological Samples Plat BT474 cells , showing increased double - strand breaks when form were thawed and tested for survival after irradiation cells were irradiated (10 Gy) after 24 hours of treatment with between January 2012 and February 2013 . Cells were grown ENZ ( 20 uM ) . Data are expressed as the meansus . e . m of at in media ( Supplementary Data 1 ) supplemented with 10 % least three experiments . fetal bovine serum ( ThermoFisher, MA ) and 100 U mL - 1 [0028 ] FIG . 17 : Treatment schematic . Schematic repre Penicillin , 100 ug mL - 1 of Streptomycin , and 292 ug mL - 1 senting embodiments for treatment delivering radiotherapy L -Glutamine (Corning , N . Y . ) . When a reference SNP geno after surgery (Panel a ) , and embodiments for delivering type was available for a cancer cell line through the CCLE radiotherapy with androgen blockade in the neoadjuvant or project, SNP genotyping was conducted by Fluidigm . “ pre -operative setting (Panel b ) . The surgery may comprise , 87 . 8 % of the 533 cancer cell lines analyzed were positively for example , breast conservation (keeping the breast intact matched in this Example to their reference genotype. For ( top schematic with nipple preserved ) ) or mastectomy ( the cellular validation studies, C4 - 2 cells were from the labo removal of all or most breast tissue (bottom chest sche ratory of Karen E . Knudsen ( Thomas Jefferson University ) matic ) . and HEC59 cells were from the laboratory of Thomas Kunkel (NIES ) . The cell lines were cross referenced with the DETAILED DESCRIPTION database of cross -contaminated or misidentified cell lines [0029 ] Provided herein are compositions, systems, kits , curated by the International Cell Line Authentication Com and methods for treating cancer in a subject with androgen mittee and NCBI BioSample and identified six cell lines that receptor positive cancer cells by sensitizing such cancer could have been contaminated or misidentified : BT20 , J82 , cells with anti -androgen receptor therapy (e . g. , Enzaluta JHH1,MDAMB435S , MKN7, and RT4 . All six of these cell mide or anti -androgen antibody ), and then treating with lines were SNP gentoyped and confirmed to match the radiation therapy . references genotype . [0030 ] In certain embodiments , androgen receptor ( AR ) is tested in a subject' s cells to determine if and / or what level Cell Culture and Irradiation . of AR is expressed prior to any anti- AR treatment and [0035 ] All cultures were maintained at 37° C . in a humidi radiation therapy . The present disclosure is not limited by fied 5 % CO , atmosphere and tested to ensure absence of the type of assay that is employed for such AR detection . In Mycoplasma. Plates were treated with y - radiation delivered certain embodiments , the AQUA approach is used for mea at 0 .91 Gymin -1 with a 137Cs source using a GammaCell 40 suring AR receptor levels in the nucleus of breast cancer test Exactor (Best Theratronics ; Ontario , Canada ). spots on a tissue microarray . AR clone Dako AR441 was used may be used as the antibody. In other embodiments , the Antibody and Reagents . MARS assay is employed ( e . g ., Dennis et al. , Cytometry [0036 ] Anti- AKT ( clone C67E7 , # 4691P, 1 : 1000 ), anti Part A , 73A : 390 - 399 , 2008 , herein incorporated by refer phospo - S473 - AKT ( clone D9E , # 4060P , 1 :1000 ), anti -AR ence for such assays ) . In some embodiments , an ELISA type ( clone D6F11 , # 5153 , 1 : 2000 ) , anti- HDAC1 ( clone 10E2 , assay is employed to detected Androgen receptor level ( e . g . , # 5356 , 1 : 2000 ) , anti -yH2AX ( clone 20E3 , # 9718 , 1 :2000 ) , Androgen Receptor ELISA Kit ( ab128498 ) from ABCAM ) . anti - actin (clone 8H10D10 , # 3700 , 1 : 4000 ) , anti - yErbB2 (clone Tyr1248 , # 2247 , 1 : 1000 ) , and anti -GAPDH (clone EXAMPLES D16H11, # 5174 , 1: 4000 -7500 ) were from Cell Signaling Technology (Beverly , Mass . ) . Anti -HER2 (clone e2 - 4001 , Example 1 # MS730PO , 1: 2000 ) and anti -ER ( clone AB - 17 , [0031 ] A Genetic Basis for the Variation in the Vulner # RB1521PO , 1 : 1500 ) were from ThermoFisher (Waltham , ability of Cancer to DNA Damage Mass . ) . Anti - yDNP - PKcs ( clone S2056 , # 18192 , 1 : 1000 was [0032 ] Radiotherapy is not currently informed by the from Abcam (Cambridge , Mass . ) . Enzalutamide was from genetic composition of an individual patient' s tumor. To Selleck (Houston , Tex . ) . DHT was from Steraloids (New identify genetic features regulating survival after DNA port , R . I . ) . damage , a large -scale profiling of cellular survival after exposure to radiation in a diverse collection of 533 geneti High - Throughput Proliferation Assay. cally annotated human tumor cell lines was conducted in this [0037 ] Cells were plated using a Multidrop Combi liquid Example . It was shown that sensitivity to radiation is char handler ( Thermo Fisher) in at least quadruplicates for each acterized by significant variation across and within lineages . time point at three cell densities ( range 25 -225 cells per Results from this were combined with genomic features to well ) in a white 384 -well plate (Corning , N . Y . ) . Plates were identify parameters that predict radiation sensitivity . Iden irradiated and at 9 days post -irradiation , media was aspirated tified were somatic copy number alterations, gene mutations, and 40 ul of CELLTITER -GLO reagent ( 50 % solution in and the basal expression of individual genes and gene sets PBS ) ( Promega, WI) was added to each well. Relative US 2017 /0319692 A1 Nov. 9 , 2017 luminescence units were measured using an Envision mul typing calls , and appropriate segmentation of the copy tilabel plate reader ( Perkin Elmer ) with a measurement time number profiles . Finally , the Genomic Identification of Sig of 0 . 1 seconds . Luminescence signal is proportional to the nificant Targets in Cancer (GISTIC ) algorithm was used to amount of ATP present. For chemical radiosensitization identify focal regions of copy number alterations in indi measurements , drug was added 24 hours prior to irradiation . vidual samples . A gene - level copy number was also gener The luminescence signal was plotted as a function of cell ated , defined as the maximum absolute segmented value density and a cell density within the linear range for lumi between the gene ' s genomic coordinates , and calculated for nescence (or growth ) was selected to generate integral all genes using the hg18 coordinates provided by the refFlat survival for each cell line. and wgRna databases from UCSC Genome Browser. Sepa rate binary variables representing amplifications ( above 0 . 7 ) Integral Survival. and deletions (below - 0 . 7 ) were generated based on the GISTIC gene- level copy number output described above . [ 0038 ] The area under the curve was estimated by trap These binary amplification /deletion variables for each gene ezoidal approximation . First, X -axis values representing were used as input to compute the IC against the drug radiation doses 1 , 2 , 3 , 4 , 5 , 6 , 8 , and 10 Gy were log 2 sensitivity phenotype . mRNA gene expression was mea transformed . The survival values for each trapezoid were sured by the GeneChip Human Genome U133 Plus 2 . 0 multiplied by the dose interval , [ f (X1 ) + F ( X )/ 2 ] * AX , Array . Raw Affymetrix CEL files were converted to a single summed and re - scaled by multiplying by ( 7 - log2 10 ) so that value for each probe set using Robust Multi -array Average integral survival is defined from 0 ( completely sensitive ) to (RMA ) and normalized using quantile normalization . Either 7 ( completely resistant) . the original Affymetrix U133 + 2 CDF file or a redefined Clonogenic Survival . custom CDF file ( ENTREZG -v15 ) was used for the sum marization . SsGSEA enrichment scores were calculated [0039 ] Cells were plated at appropriate dilutions , irradi based on the weighted difference of the Empirical Cumula ated , and incubated for 7 - 21 days for colony formation . For tive Distribution Functions of the genes in the set relative to chemical radiosensitization measurements, drug was added the genes not included in an individual set34 . The result is a 24 hours prior to irradiation . Colonies were fixed in a single score per cell line per gene set, transforming the solution of acetic acid and methanol 1 : 3 ( v / v ) and stained original dataset into a more interpretable higher - level with 0 .5 % (w / v ) crystal violet as previously described “ l. A description . Gene sets were obtained from the C2 sub colony was defined to consist of 50 cells or greater . Colonies collection of the Molecular Signatures database (MSigDB ) were counted digitally using ImageJ software as 70 , an additional collection of oncogenic signatures , and describedø2. Integration of survival as a function of dose , or other cancer- related gene sets curated from the literature , area under the curve , was calculated using Prism , GraphPad resulting in a dataset that has 4 ,628 pathway profiles for each Software (La Jolla , Calif .) . sample . SsGSEA values were used as input to compute the IC The nominal p - values for the information based associa Information -Based Association Score . tion metric scores between the genetic parameters (altera [ 0040 ] The association between genomic alterations ( e . g . tions or ssGSEA scores ) and radiation response scores were mutations or SCNA ) or ssGSEA profiles for each gene set estimated using an empirical permutation test . and the radiation response profile was determined using the Information Coefficient ( IC ) 2 , 63 , 64. NFE2L2 Pathway Signatures. 10042 ] For the gene transcription signature of pathway Genetic Data . NFE2L2 ( or NRF2 ) , the expression values from the CCLE [0041 ] Cancer cell lines were profiled at the genomic dataset were extracted . For each gene , expression valued level . Briefly , mutation information was obtained both by were normalized to standard deviations from the median using massively parallel sequencing of > 1 , 600 genes and by across cell lines . The average normalized expression of the mass spectrometric genotyping (OncoMap 3 . 0 ) , which inter signature genes was computed within each cell line in which rogated 381 specific mutations in 33 known oncogenes and data was available . Across the cell lines , the gene signature tumor suppressors . Genotypes were transformed to categori scores were normalized to standard deviations from the cal values (mutation = 1 , no mutation = 0 ) and were used as median across CCLE , and a “ summary score ” for each input to compute the IC . Genotyping/ copy- number analysis pathway was computed as the average of the individual was performed using Affymetrix Genome- Wide Human normalized signature scores .' SNP Array 6 . 0 . Raw Affymetrix CEL files were converted to a single value for each probe set representing a SNP allele Comet Assays or a copy number probe using a GenePattern pipeline and hg18 Affymetrix probe annotations. Copy numbers were [0043 ] Single -cell gel electrophoresis was conducted in then inferred based upon estimating probe set specific linear alkaline or neutral buffer according to the manufacturer' s calibration curves, followed by normalization by the most instructions, Trevigen (Gaithersburg , Md. ) . Slides were similar HapMap normal samples . Segmentation of normal blinded and enumerated by a single user. ized log 2 ratios ( specifically , log 2 (CN / 2 ) ) was performed using the circular binary segmentation (CBS ) algorithmº , Western Blot Analysis . followed by median centering of the segment values to a [ 0044 ] Whole cell lysates were prepared using M -PER value of zero in each sample . Next, quality checking of each lysis buffer and clarified by centrifugation . Proteins were array was performed , including visual inspection of the array separated by SDS -PAGE and transferred onto 0 .45 uM pseudo - images , probe - to - probe noise variation between nitrocellulose membranes (Maine Manufacturing ; Sanford , copy -number values, confidence levels of Birdseeds geno - Me. ). After primary antibody incubation for 1 -2 hours at US 2017 /0319692 A1 Nov. 9 , 2017 room temperature , washings, and incubation with secondary To assess differences in the distribution of response across antibodies , blots were developed with a chemoluminescence all profiled cancer types, in a category of cancers derived system ( Amersham /GE Healthcare ) . For yH2AX measure from single organ , and in a single lineage , the histogram and ments, proteins were transferred onto 0 . 2 uM nitrocellulose probability density distribution of integral survival was (Bio - Rad ) . plotted for all profiled cell lines , those derived from non small cell lung cancers , and those from lung adenocarci Mouse Xenograft Studies . noma ( FIG . 1d ) . All three demonstrated a normal distribu tion . In fact, the majority of lineages ( cut -off > 25 cell lines [0045 ] Female NSG mice , 6 -8 weeks of age , were profiled ) , including lung adenocarcinoma, breast, glioma , obtained from the Cleveland Clinic Biological Resources ovary , pancreas, and melanoma, were normally distributed Unit facility. All mouse studies were conducted under a ( D ' Agostino -Pearson : K - < 0 . 65 , P > 0 . 5 ) . Only two lineages , protocol approved by the Cleveland Clinic Institutional colorectal and uterine , demonstrated non -Gaussian distribu Animal Care and Use Committee MDAMB453 cells were tions, mostly due to a higher proportion of resistant cell lines resuspended in serum free media and injected into the than predicted by a normal distribution ( FIG . 8 ). This is inguinal mammary gland . Once tumors reached 200 mm », attributed to cell lines with large values of copy number. mice were block randomized and assigned to vehicle , Taken together, the high -throughput platform accurately enzalutamide , vehicle plus radiotherapy, or enzalutamide profiles cell lines from multiple lineages for radiation plus radiotherapy. Two cohorts consisting of these four arms response and reveals a mostly Gaussian distribution of underwent treatment. Vehicle consisted of a volume of 5 mL per kg of PEG -400 solution containing 1 . 5 % / 0 of DMSO for radiation response within lineages . cohort 1 and 2 .5 % DMSO for cohort 2 via oral gavage daily . Cohort I received enzalutamide at 15 mg kg- and cohort 2 X -Rays and DNA Damaging Drugs received enzaluatmide at 25 mg kg - 7. Radiotherapy was [0049 ] To assess the accuracy of the platform and probe delivered to a dose of 1 . 5 Gy in three fractions once tumor similarities between radiation and drug therapy , the corre size reached 250 mm " . Treatment was not blinded to the lation of responses between radiation and 481 compound investigator. Tumor volume was measured daily . Mice were probes profiled by CTD2 was calculated ( FIG . le ) . The sacrificed once their tumors reached an approximate size of most positively correlated compound , VX -680 , is a pan 1 ,000 mm or at treatment days 21 - 28 . The significance of inhibitor of the Aurora kinases (Spearman r = 0 .60 ) ?º . Aurora the difference between treatment groups was assessed by kinases are essential for the regulation of chromosome one -way and the interaction between drug and radiation was segregation and cytokinesis during mitosis and their inhibi measured by two -way ANOVA . tion leads to cell death by mitotic catastrophe , a major mode of cell death after irradiation . In addition to VX - 680 , geno Results toxic chemotherapeutics such as etoposide , paclitaxel, doxo rubicin , bleomycin , and others were likely to be correlated [ 0046 ] Variation in Survival after Irradiation with radiation sensitivity . These correlations suggest similar 10047 ) Radiation survival of 533 cancer cell lines com genetic dependencies between genotoxic compounds and prising 26 cancer types was profiled using a recently devel radiation . oped high - throughput profiling platform (FIG . 1a ) . This SCNA Regulate Survival after Irradiation platform was previously benchmarked against the clono [0050 ] Somatic copy -number alteration ( s ) (SCNA ) are genic survival assay in lung squamous cancer cell lines . It common in cancerll , have a critical role in promoting was previously demonstrated that the high - throughput mea oncogenesis ! ?, and an understanding of their phenotypic surements closely approximated clonogenic survival by effects has led to advances in cancer diagnostics and thera most radiation response parameters , with the greatest level peutics7, 13, 14 . The interaction between the SCNA landscape of correlation observed with a longer time to readout, at and the response to radiation remains poorly defined . The doses within the Giso range of most cell lines profiled , and fraction of the genome that contains a SCNA or (fSCNA ) when comparingmean integral survival values . To assess the was measured by measuring the length of segments with log platform ' s validity beyond lung squamous lineages , clono 2 SCNA values larger than 0 . 2 from the GISTIC output, genic survival was measured in cell lines derived from divided by the length of all segments measured . Therefore , multiple lineages and exhibiting a wide range of responses the fSCNA represents a surrogate measure of genomic to radiation . Survival as a function of dose was integrated instability based on relative SCNA . A positive correlation and this generated values for each cell line ( FIG . 7 Panel a ). (Pearson r = 0 .27 ) was observed between fSCNA and integral Integral survival ( single experiment) or mean integral sur survival ( FIG . 2a ) . The log 2 of the number of mutations was vival values ( average of duplicates) for 15 cell lines were plotted per sample and integral survival and observed a calculated and compared to values from the clonogenic modest negative correlation ( Pearson r = - 0 . 19 ) . assay (for each cell line , n > 2 ) ( FIG . 1b ) . High -throughput [0051 ] It was reasoned that the overall positive correlation and colony integral survival values were significantly cor of fSCNA with radiation survival could reflect an increased related , with Pearson r = 0 .85 , R2 = 0 .73 for single experimen capability of tumor cells to repair DNA double -strand breaks tal profiling and Pearson r = 0 . 89 , R2= 0 . 80 with the average after radiation , utilizing mechanisms that are also used in the of two profiling experiments . It was shown that the prolif creation of SCNA such as non -homologous or micro -ho eration index approximated clonogenic survival across indi mology mediated end - joining or other error - prone repair vidual doses , with the best approximations occurring within mechanisms ( e . g . non -allelic homologous recombination )15 . the G150 range (FIG . 7 Panel b ). Alternatively , individual SCNA could regulate survival after [0048 ] A column scatter plot of integral survival demon radiation by changing the expression of specific genes strated significant variation in survival across and within within the structurally altered chromosomal segments . The lineages (FIG . 1c ), the latter being on the order of 5 -7 fold . former is predicted to create a stochastic order of individual US 2017 /0319692 A1 Nov. 9 , 2017

SCNA correlated with survival , the latter would identify logical function ( FIG . 3a ) . Seven of these genes have discrete SCNA on both sides of the survival spectrum . To previously been implicated in the DNA damage response : assess the association of individual SCNA with radiation TPR18 , FLNA19, TP53BP120 , SMG121 , RANBP922 , response , alterations were correlated with radiation survival SMARCA423 , and STAG324. A subset of the 19 genes using the Information Coefficient ( IC ). The top 50 gene level demonstrated domain - selectivity in conferring sensitivity SCNA correlating with resistance and sensitivity were orga (FIG . 9 ) . Other top genes that correlated with radiation nized by chromosome position and the results were depicted sensitivity have not previously been implicated in radiation using a wheel- plot ( FIG . 2b ) . The relative enrichment for induced damage response . discrete chromosomal segments that correlated with resis [0054 ] It was sought to determine whether identified genes tance or sensitivity suggested that individual SCNA events were regulators of radiation response or merely associated were not randomly distributed across the radiation response with sensitivity . An example of the latter is PIK3CA . The range . To assess whether SCNA can contribute to radiation phosphatidylinosoitol 3 - kinase (PI3K ) /protein kinase b response directly , radiation survival was correlated with the (AKT ) pathway is frequently deregulated in human cancer expression of genes within the altered segments and com and is an important tumor cell survival pathway25 . Muta pared the means of the coefficients by pair -wise analysis , tions in PIK3CA , which typically result in activation of resistant versus sensitive (FIG . 2c ) . The amplified regions PI3K /AKT , were associated with radiation sensitivity in the that correlated with radiation resistance had a significantly dataset ( FIG . 3a ) . However , activation of PI3K / AKT has higher mean correlation coefficient than amplifications that previously been associated with radiation resistance26 , 27. A correlated with radiation sensitivity . The inverse was closer look at the domains within PIK3CA and their role in observed for deleted regions that correlated with radiation radiation sensitivity indicates that cell lines with mutations resistance . In some cases , changes in the expression of in the p85 binding domain ( p85 BD ) are mostly sensitive to individual genes have previously been implicated in radiation , driving the overall association (FIG . 3b ) . Cell response to cytotoxic stress . For example , genes overex lines with mutations in the p85 BD of PIK3CA were largely pressed as a consequence of focal amplicon 20q11 . 2 include from the uterine lineage ( 38 % of the cell lines profiled in functionally important genes in cell cycle regulation (E2F1 ) , FIG . 3b ) , had a co -occurring mutation in PTEN ( FIG . 3c ) , chromosome maintenance (KIF3B ) , glutathione synthesis and two representative cell lines were yAKT replete ( FIG . (GSS ) , and apoptosis (BCL2L1 ) . All of these genes were 3d ), indicating p85 BD mutations retain PI3K enzymatic positively correlated with radiation resistance . Collectively , activity . Mutations in the PIK3CA p85 BD were also fre these results indicate that SCNA regulate the response of quently identified ( > 20 % ) in uterine tumor samples profiled cells to radiation - induced damage in part through direct by the The Cancer Genome Atlas ( TCGA ) network (FIG . changes in gene expression . 3e ) , paralleling the cell line data ? . These results indicate [0052 ] The frequency and distribution of SCNA vary that PIK3CA p85 BD mutations reflect markers of a radio across tumor lineages 2 . A scatterplot of fSCNA and integral sensitive lineage ( uterine carcinoma ). survival revealed differences in the degree and direction of [0055 ] Mutations were analyzed that conferred radiation association across lineages. Colorectal, uterine and ovarian resistance and identified the key regulator of oxidative stress carcinomas showed a positive correlation between survival response , KEAP1 , which ranked ninth ( IC = 0 . 112 ; P = 0 . and fSCNA values (FIG . 2d ) . Colorectal and uterine carci 0513 , calculated using the empirical permutation test ) from nomas have been previously shown to contain tumors with a list of > 1600 genes (FIG . 38) . It has been shown that extensive SCNA and low mutation frequency and a subset of mutations in KEAP1 result in the stabilization and activity tumors with low SCNA and high mutation frequency . The of the master transcriptional regulator of oxidative damage latter is attributed to either MLH1 silencing and / or DNA response , Nrf2 (encoded by NFE2L2 ) , thereby conferring polymerase € (POLE ) mutations16 , 17 . Consistent with these radiation resistance8, 28 , 29 , 30 . Recently , the spectrum of findings , a correlation was observed between radiation sur KEAP1 mutations was analyzed , revealing distinct func vival and SCNA and an anti- correlation between radiation tional categories including passenger , loss - of- function , survival and mutation frequency in both lineages (FIG . 2e ) . hypomorphic , or “ super- binders ” 31 . It was reasoned that the Mutations were identified in individual genes that correlated likelihood of passenger or hypomorphic mutationsmasking with radiation sensitivity in both lineages using the IC ( FIG . association is less likely to occur in a lineage with frequent 2 /) . Of the top 20 genes that correlated with radiation KEAP1 mutations. To test this , the IC in adenocarcinoma of sensitivity in uterine and colorectal cancers , six and eight, the lung was analyzed , which has the highest frequency of respectively , have previously been associated with DNA KEAP1 mutations of any lineage profiled by the TCGA repair . These results suggest an association between low network to date (FIG . 39 ) 32 . Consistent with TCGA network SCNA , high mutation frequency, DNA repair gene disrup data , the strongest association between KEAP1 mutation tion , and radiation sensitivity in these lineages . and radiation resistance was in adenocarcinoma of the lung Gene Mutations Regulate Cellular Survival after Irradiation (IC = 0 .352 ; P = 0 .0224 , calculated using the empirical permu [0053 ] Recent studies have identified recurrent gene muta tation test ) (FIG . 3h ). CUL3 , encoding the ubiquitin ligase tions that are correlated with the likelihood of response to adapter that binds to Keap1 and degrades Nrf2 , was also specific agents in cancer , 5 . Identifying gene mutations that associated with radiation resistance in adenocarcinoma of correlate with radiation response have the potential to simi the lung (FIG . 3h ) . larly inform clinical management. Gene mutations were [0056 ] To assess the impact ofmutation position on the IC , identified that correlated with radiation sensitivity across all the relative importance of residue position on survival was lineages using the IC . Higher IC values were observed for assessed in the binding partner of Keap1, Nrf2 (FIG . 3i) . In genes with mutations and radiation sensitivity compared to human cancer, somatic mutations in NFE2L2 frequently resistance . The top 19 genes that were associated with occur within the two KEAP1 binding sites (D2LG and radiation sensitivity when mutated were organized by bio E79TGE ) 33 . The IC value for NFE2L2 mutation across all US 2017 /0319692 A1 Nov. 9 , 2017 lineages (IC = - 0 .0697 ; P = 0 .329 , calculated using the empiri tant role for SQSTM1 in Nrf2 activation in HCC39 . To test cal permutation test ) was significantly higher when muta the association between Nrf2 and Sqstml activity and radia tions restricted to the two KEAP1 binding sites were con tion survival in HCC , integral survival values were plotted sidered ( IC = 0 . 245 ; P = 0 .033 , calculated using the empirical with NFE2L2 activity ( FIG . 4e ) and SQSTM1 expression permutation test ) . (FIG . 48) . It was found that HCC had the strongest associa [0057 ] Taken together, these results describe gene muta tion between radiation survival and Nrf2 activity in any tion determinants of radiation - induced cellular damage lineage profiled (R2 = 0 .41 in HCC v . R2 = 0 . 22 in LUSC ). response and reveal distinct functional consequences for SQSTM1 expression was also correlated with radiation categories of mutations within individual genes. survival, albeit at a lower level than that obtained with the Gene Expression Profiles Regulate Survivalafter Irradiation Nrf2 score . This was attributed to noise associated with [0058 ] SSGSEA ( single - sample Gene Set Enrichment single gene expression measurements , compared to a com Analysis ) projections28 , 34 was used as a gene set identifi posite Nrf2 score that includes 565 genes. HCC is com cation tool to find genetic pathways that are differentially monly managed by genotoxic therapies (chemo - and / or correlated with radiation response . The profiles of each gene radio -embolization , external beam radiotherapy ) and / or sur set /pathway were compared with the radiation response gery , suggesting that patients who resist genotoxic stress scores ( integral survival) across cancer types. The ssGSEA may have poorer clinical outcomes . HCC patients with scores are displayed in a heatmap with the top gene sets that elevated SQSTM1 expression have a significantly lower correlate and anti- correlate with radiation survival organized overall survival, indicating a poor overall prognosis for by cellular pathways ( FIG . 10 ) . The top gene sets that patients with active Nrf2 in HCC (FIG . 4g ) ( TCGA net correlated with radiation sensitivity revealed pathways work ). This is analogous to the poor prognosis of NSCLC including DNA damage response , cell cycle , chromatin patients with active Nrf2 [ 40 and FIG . 11 ]. organization , and RNA metabolism . The top gene set that correlated with radiation resistance revealed pathways Radiogenomic Profiling of Breast Cancer including cellular signaling, lipid metabolism and transport, [0061 ] For many women with breast cancer , breast - con stem -cell state , cellular stress, and inflammation . The mul serving surgery or mastectomy can result in the removal of titude of pathways associated with radiation response indi detectable macroscopic disease . However, tumor foci might cates that cellular processes well beyond DNA repair regu remain in local and regional tissue such as the intact breast late cellular survival after radiation . A closer look at active or chest wall and /or regional lymph nodes. Tumor recur signaling pathways that correlate with radiation resistance rence can cause considerable morbidity , dissemination of reveals several targetable cellular receptors including TGFB , disease , and an increased probability of breast- cancer mor EGFR , Estrogen Receptor (ER ) , NFKB , JAK /STAT , AKT, tality41, 42 . Radiotherapy significantly decreases the risk of FGFR , HER2, RAF , MEK , and Wnt? / - catenin . Some of local and regional recurrence and breast cancer mortality 43 , these receptors have previously been shown to confer resis 44 . Despite the demonstrated efficacy of breast radiotherapy, tance to radiation in select cell lines . These results indicate there remains an important need for identifying patients who the broad role these signaling pathways serve in regulating are more likely to fail therapy and improving radiation radiation survival across several tumor lineages and impli treatments in those patients . cate new targets for radiosensitization . [0062 ] To identify genetic determinants of radiation sur [ 0059 ] To assess the importance of the expression of vival in breast carcinomas, an unbiased query of gene individual genes on radiation survival, correlation coeffi mutations or copy number changes was begun that corre cients were calculated between 18 , 988 genes and integral lated with radiation survival in 28 breast cancer cell lines. survival values ( FIG . 4a ) . NQO1 and SOSTM1, both tran The top 50 segments correlating with resistance were orga scriptionally activated by Nrf236 , were the ninth and 11th nized by chromosomal position and the results were genes identified as strongly associated with resistance , cor depicted using a wheel- plot ( FIG . 5a ) . Amplification of roborating a role for oxidative stress response in radiation 17q12 -22 , which contains ERBB2, was associated with resistance already implicated by the gene mutation data . radiation resistance . Cell lines containing 17912 - 22 ampli NQO1 encodes the NAD ( P ) H - quinone oxidoreductase , an fication had elevated ERBB2 copy number and mRNA , and enzyme that detoxifies cells from reactive oxygen species were generally derived from tumors clinically annotated as induced quinone -containing compounds37 . The ubiquitin having ERBB2 amplifications. ERBB2 copy number (Pear binding protein , Sqstml (or p62 ) plays a role in oxidative son r = 0 .43 ) and gene expression (Pearson r = 0 .45 ) were stress , cellular signaling, and autophagy38 . Sqstmlhas been correlated with radiation survival (FIG . 5b ) . These results previously shown to interact with Keap1 and accumulation suggest that ERBB2 is the likely mediator of radiation of Sqstml can lead to an increase in Nrf2 activity 39 . resistance in the 17q12 - 22 amplicon . Consistent with these [0060 ] Consistent with these results , NQO1 and SQSTM1 results , overexpression of ERBB2 has previously been gene expressions are strongly correlated across 979 Cancer shown to confer therapeutic resistance in breast cancer cells Cell Line Encyclopedia ( CCLE) cell lines (FIG . 4b ) and and Trastuzumab , a monoclonal antibody that interferes with NFE2L2 transcriptional activity is associated with radiation ErbB2 , sensitizes ErbB2 expressing cells to radiation “ . survival across all lineages (FIG . 4c ). NFE2L2 transcrip AR Regulates Survival after Irradiation in Breast Cancer tional activity plotted by lineage revealed the highest overall [0063 ] To find additional genetic pathways that are differ activity in hepatocellular carcinoma (HCC ) and biliary can entially correlated with radiation response, ssGSEA projec cer (FIG . 4d ). It was shown that Nrf2 is mainly activated by tions were applied ( FIG . 50 ). Multiple pathways were cor mutations in NFE2L2 and /or KEAP1 and /or deletions in related with radiation resistance ( ER , ERBB2, JAK /STAT3 , CUL3 in lung squamous cancers (LUSC ) 28 . Similar gene and PI3K ). However, one of the most correlated and intrigu alterations have not been identified in HCC or biliary cancer ing gene sets identified was associated with androgen sig ( TCGA network ) . Instead , recent reports suggest an impor- naling . Androgen receptor ( AR ) expression is most fre US 2017 /0319692 A1 Nov. 9 , 2017 quently observed in prostate cancer, where it has been shown same growth survival assay was employed across several to promote resistance to radiation46. Several clinical studies cell lines and showed protection and sensitization with DHT have demonstrated a benefit in overall survival with the and ENZ , respectively, and only in cell lines that expressed combination of androgen - deprivation therapy and radiation AR and independent of ER and ErbB2 activity ( FIG . 6b ) . compared to radiation alone in patients with intermediate For cells capable of forming colonies, these results were and high - risk prostate cancer47 , 48 , 49 . Although its role in confirmed using a clonogenic survival assay (FIG . 15 ) . breast oncogenesis remains poorly defined , AR is detected in a majority ofbreast carcinomas at levels greater than normal AR Protects Breast Cancer Cells from DNA Damage breast ' . A subset of AR -positive , triple -negative breast [ 0067 ] Using the neutral comet assay , it was next deter carcinomas , which lack ER and progesterone receptor (PR ) mined whether the reduction in cellular survival was asso expression and ErbB2 overexpression , appear to be depen ciated with increased DNA damage MDAMB453 cells were dent on AR signaling for growth51 , 52 . A phase II study that pretreated for 24 hours with either DHT ( FIG . 6c , left ) or explored bicalutamide in AR -positive , ER /PR - negative ENZ (FIG . 6c , right) , prior to 15 Gy of irradiation . These metastatic breast cancer showed a modest clinical benefit3. results indicate that DHT decreased and ENZ increased A single - arm phase II is currently assessing the more potent DNA damage in MDAMB453 cells . To assess the role of AR antagonist , enzalutamide ( ENZ ) , in women with DNA repair , the kinetics of yH2AX formation and resolution advanced , AR -positive , ER / PR -negative breast cancer . was measured , as surrogate marker of DNA DSBs and Therefore , the clinical effectiveness of anti- androgens as subsequent repair , under the same conditions . YH2AX kinet single agent therapy in the management of breast cancer ics were consistent with the results of the comet assay : levels patients remains unknown . Based on the initial results and of yH2AX in MDAMB453 cells were increased shortly after these observations, it was sought to examine the role of AR irradiation by ENZ increased and decreased by DHT ( FIG . and test the rationale for combining androgen blockade and 6d ) , an effect that persisted at 24 hours . The kinetics of radiation in breast cancer , as is the standard of care in locally yH2AX formation and resolution in AR - negative HMC18 advanced prostate cancer47 , 48. cells was unaffected by ENZ treatment, while DHT hastened [0064 ] It was shown that AR mRNA levels correlated with yH2AX resolution in AR - positive C4 - 2 prostate cancer cells . radiation survival (Pearson r = 0 .48 ) (FIG . 5d ) . Data was These and similar results in the AR - positive , ER - positive , analyzed from the CCLE to determine the relative mRNA ErbB2 active cell line , BT474 (FIG . 16 ) , suggest that levels of AR across lineages. Prostate cancers had the suppression of androgen signaling results in increased DNA highest mean value among 28 tumor types followed by damage and / or decreased repair in breast cancer cells that osteosarcoma and breast cancer (FIG . 5e ) . An association express AR , independent of ER and ErbB2 activity . between AR mRNA and protein levels (Pearson r = 0 .615 ) [0068 ] Previous work on AR regulated radiation resistance was determined , and AR and ESR1 (or ERa ) mRNA (Pear in prostate cancer suggested a decrease in the activity of son r = 0 . 34 ) in TCGA patient samples (FIG . 12 ) . A subset of PRKDC (or DNAPKcs ) , a key signaling molecule that TCGA samples expressed AR and not ESR1. Consistent initiates early stages of NHEJ, after androgen deprivation . with this data , it was shown that some of the breast carci The phosphorylation of DNAPKcs on Ser2056 was exam noma cell lines that were profiled expressed AR with vary ined , which is indicative of activated DNAPKcs , in ing levels of ER (and ErbB2) (FIG V . 5fJ and FigFIG . 13 )). MDAMB453 cells . Cells cultured in steroid - deprived media [0065 ] MDAMB453 cells expressed AR but not ER , and followed by irradiation had greater yDNAPKcs levels com although ErbB2 was overexpressed in these cells , the level pared to steroid -replete media (FIG . 6e ) . Supplementation of of expression was significantly lower than that observed in steroid - deprived media with DHT maintained yDNAPKcs at other AR positive cell lines (BT474 , ZR7530 , or HCC202 ) similar levels to those cells grown in the steroid -replete ( FIG . 57) . To examine the activity of ErbB2 in this cell line , media following irradiation . The greater level of yDNAPKcs levels of yErbB2 levels at tyrosine 124854 were measured . are consistent with increased DNA damage after androgen yErbB2 was not elevated in MDAMB453 cells , indicating deprivation in MDAMB453 cells and suggests cell - to -cell or that ErbB2 is not activated in this cell line (FIG . 14 ) . These lineage - to - lineage variability in the mechanism by which results indicate that ERBB2 overexpression is not sufficient androgen signaling mediates survival after irradiation . These to activate ErbB2 in MDAMB453 cells . Taken together , results demonstrate that androgen activity plays a pivotal MDAMB453 cells are AR -positive , ER - negative and ErbB2 role in the response of AR expressing breast cancer cells to inactive, consistent with its gene expression -based classifi ionizing radiation . cation into the luminal AR expressing subtype of triple [0069 ] To assess this concept in vivo , MDAMB453 ortho negative breast cancers. topic xenografts were randomized into one of four treatment [0066 ] MDAMB453 cells were used to test the interaction groups :mock , ENZ , ionizing radiation , or ENZ and ionizing between androgen signaling and survival after radiation . radiation ( FIG . 6f- i ) . The combination of ENZ ( at 15 mg Dihydrotestosterone (DHT ) re - supplementation of kg - or 25 mg kg ) and ionizing radiation resulted in more MDAMB453 cells cultured in steroid -deprived media effective suppression of tumor growth than either modality before radiation showed dose dependent rescue of cell alone . However , despite no effects on tumor growth when growth ( FIG . 6a , top ) . Conversely , ENZ , which competes used alone , ENZ at a dose 15 mg kg - showed interactive with DHT for binding to the receptor and prevents its tumor suppression with radiotherapy . Suppression of tumor nuclear translocation , decreased cell number compared to growth was observed with ENZ at a dose of 25 mg kg - - . vehicle alone ( FIG . 6a , bottom ) Of note , this experimental However, mice that received both ENZ at a dose of 25 mg system models single fraction radiation treatment. Patients kg - - and radiation lost ~ 10 % of their body weight during the typically receive 16 -30 total fractions of treatment, which is course of treatment , suggesting that selection of ENZ dose predicted to compound the magnitude of the observed is likely to be important in the optimization of the thera interaction between androgen signaling and radiation . The peutic index . Taken together , these data show that androgen US 2017 /0319692 A1 Nov. 9 , 2017 ablation cooperates with ionizing radiation to decrease not comprehensive . Finally , levels of non - coding RNA, tumor cell growth and survival in AR -positive breast adeno metabolites , and proteins and their post - translational modi carcinomas . fications are also likely to impact the intrinsic cellular [ 0070 ] This example , which is exemplary , represents the response to radiation . This data enables correlations of largest analysis to date of cancer cell line survival after ionizing radiation sensitivity with levels of these important exposure to ionizing radiation . 533 genetically annotated biomolecules as additional genomic and cellular datasets tumor cell lines were profiled using a single , validated emerge . experimental platform and genetic determinants of tumor [ 0074 ] In summary, the results reveal a genetic basis of the cellular response to radiation were identified . The distribu variation in the vulnerability of cancer to DNA damage . This tions of survival after exposures to radiation across and information can help guide the transition of the use of X -rays within lineages were mostly Gaussian distributions , demon and DNA -damaging drugs from the current generic strating significant underlying biological diversity . The approach to one in which these therapies are guided by clinical responses of some cancer types to radiotherapy vary genetic alterations in individual patient' s tumor. in a manner not fully explained by clinical or histopatho logical features . These results suggest that intrinsic cellular REFERENCES determinants are likely to contribute to this variance . [0075 ] 1 . Alhassani A , Chandra A , Chernew M E . The [0071 ] Radiation sensitivity and genomic parameters were sources of the SGR " hole " . N Engl J Med 366 , 289 - 291 correlated using a statistical methodology that is sensitive to ( 2012 ) . non - linear relationships and with better resolution at the [0076 ] 2 . Ree A H , Redalen K R . Personalized radio high end of the matching range . It was shown that individual therapy : concepts , biomarkers and trial design . Br J SCNA , gene mutations, and the basal expression of indi Radiol 88 , 20150009 ( 2015 ). vidual genes and gene sets correlate with radiation survival. [0077 ] 3 . Yard B , Chie E K , Adams D J , Peacock C , In lineage - specific analysis of breast cancer cell lines it was Abazeed M E . Radiotherapy in the Era of Precision demonstrated , for the first time, a role for AR expression in Medicine . Semin Radiat Oncol 25 , 227 -236 ( 2015 ) . the response of breast cancer cells to ionizing radiation , [ 0078 ] 4 . Chapman P B , et al. Improved survival with mainly by preventing DNA damage. By studying a large vemurafenib in melanoma with BRAF V600E mutation . number of cancer types , it was found that genetic correlates N Engl J Med 364 , 2507 - 2516 ( 2011 ) . in any single cancer type can be found in other cancer types [0079 ] 5 . Lynch T J , et al. Activating mutations in the as well ( e . g . Nrf2 activation in LUSC , LUAD , and hepato epidermal growth factor receptor underlying responsive biliary cancer and AR expression in prostate and breast ness of non -small -cell lung cancer to gefitinib . N Engl J adenocarcinomas ) . This supports the view that although Med 350 , 2129 - 2139 (2004 ). diverse, cancer genomes reflect combinations of a limited [0080 ] 6 . Rosell R , et al . Erlotinib versus standard che number of functionally relevant events that can confer motherapy as first- line treatment for European patients therapeutic resistance across cancer types. Importantly , the with advanced EGFR mutation - positive non - small - cell positive correlation between cellular response to ionizing lung cancer (EURTAC ) : a multicentre , open - label , ran radiation and genotoxic compounds suggests common domised phase 3 trial. Lancet Oncol 13, 239 -246 ( 2012 ) . genetic dependencies between the twomost common cancer [0081 ] 7 . Slamon D J, et al . Use of chemotherapy plus a therapies in use , DNA - damaging chemotherapy and X -rays . monoclonal antibody against HER2 for metastatic breast [ 0072 ] Several new genetic determinants of response to cancer that overexpresses HER2. N Engl J Med 344 , DNA damage were identified . These genetic alterations can have predictive capacity by identifying the likelihood of 783 -792 (2001 ) . response to therapy and , consequently , prognosis . Diagnos 100821 8 . Abazeed M E , et al. Integrative radiogenomic tics that measure genetic changes can assist in the selection profiling of squamous cell lung cancer . Cancer Res 73 , of patients likely to respond to anti -cancer agents4 , 5 , , 7 , 5 6289 -6298 (2013 ) . The potential for stratification of patients from heteroge 100831 9 . Seashore -Ludlow B , et al. Harnessing Connec neous populations to genetically similar subgroups can help tivity in a Large - Scale Small -Molecule Sensitivity Data guide the transition of DNA -damaging chemotherapy and set . Cancer Discov 5 , 1210 - 1223 ( 2015 ) . X - rays from a generic population - based approach to one that [ 0084 ] 10 . Harrington E A , et al . VX - 680 , a potent and is more personalized . A subset of the alterations that were selective small -molecule inhibitor of the Aurora kinases , identified can also guide combinatorial therapeutic strategies suppresses tumor growth in vivo . Nature medicine 10 , since these alterations conferred resistance and are targets of 262 - 267 ( 2004 ) . current FDA approved drugs, creating an opportunity for the [0085 ] 11 . Beroukhim R , et al . The landscape of somatic precision targeting of therapeutic resistance ( e . g . AR expres copy - number alteration across human cancers . Nature sion and anti- androgens in breast cancer and NFE2L2 acti 463 , 899 - 905 ( 2010 ) . vation and anti - PI3K therapies in NSCLC28, 58 , 59) . [0086 ] 12 . Zack T I , et al . Pan -cancer patterns of somatic [0073 ] Although several genetic determinants were iden copy number alteration . Nat Genet 45 , 1134 - 1140 (2013 ) . tified that regulate the survival of cells after exposure to 100871 13 . Engelman JA , et al. MET amplification leads radiation , there are likely substantial additional parameters . to gefitinib resistance in lung cancer by activating ERBB3 Many of the cancer types that were profiled were repre signaling . Science 316 , 1039 - 1043 (2007 ) . sented by relatively few samples and others were not rep [0088 ] 14 . Maris J M . Recent advances in neuroblastoma. resented . Some SCNA were not measured due to the reso N Engl J Med 362 , 2202 - 2211 ( 2010 ) . lution limit of the array platform and although the input of [0089 ] 15 . Hastings P J , Lupski JR , Rosenberg S M , Ira > 1600 cancer relevant genes profiled for mutations provides G . Mechanisms of change in gene copy number. Nature a powerful initial assessment of likely relevant genes, it is reviews Genetics 10 , 551- 564 ( 2009 ) . US 2017 /0319692 A1 Nov. 9 , 2017

[0090 ] 16 . Cancer Genome Atlas N . Comprehensive promote malignancy . Proceedings of the National Acad molecular characterization of human colon and rectal emy of Sciences of the United States of America 105 , cancer. Nature 487 , 330 -337 (2012 ). 13568 - 13573 ( 2008 ) . [0091 ] 17 . Cancer Genome Atlas Research N , et al. Inte [0108 ] 34 . Barbie DA , et al. Systematic RNA interference grated genomic characterization of endometrial carci reveals that oncogenic KRAS -driven cancers require noma. Nature 497 , 67 - 73 ( 2013 ) . TBK1. Nature 462 , 108 - 112 ( 2009 ) . [0092 ] 18 . Galy V , Olivo -Marin JC , Scherthan H , Doye V , [0109 ] 35 . Zaidi SH , Huddart R A , Harrington K J . Novel Rascalou N , Nehrbass U . Nuclear pore complexes in the targeted radiosensitisers in cancer treatment . Current drug organization of silent telomeric chromatin . Nature 403 , discovery technologies 6 , 103 - 134 ( 2009 ). 108 - 112 ( 2000 ) . 0110j 36 . Malhotra D , et al. Global mapping of binding [0093 ] 19 . Velkova A , Carvalho MA, Johnson J O , Tav sites for Nrf2 identifies novel targets in cell survival tigian S V , Monteiro A N . Identification of Filamin A as response through ChIP -Seq profiling and network analy a BRCA1 - interacting protein required for efficient DNA sis . Nucleic Acids Res 38, 5718 -5734 (2010 ). repair. Cell cycle 9 , 1421- 1433 ( 2010 ) . [0111 ] 37 . Dinkova - Kostova A T , Talalay P . NAD ( P ) H : 10094 ] 20 . Schultz L B , Chehab N H , Malikzay A , Hala quinone acceptor oxidoreductase 1 (NQ01 ) , a multifunc zonetis T D . p53 binding protein 1 (53BP1 ) is an early tional antioxidant enzyme and exceptionally versatile participant in the cellular response to DNA double - strand cytoprotector. Arch Biochem Biophys 501, 116 - 123 breaks. J Cell Biol 151 , 1381 - 1390 ( 2000 ) . [0095 ] 21. Brumbaugh K M , et al. The mRNA surveillance ( 2010 ) . protein hSMG - 1 functions in genotoxic stress response [01121 38 . Mathew R , et al . Autophagy suppresses tum pathways in mammalian cells. Molecular cell 14 , 585 -598 origenesis through elimination of p62. Cell 137 , 1062 ( 2004) . 1075 ( 2009 ). [0096 ] 22 . Bao J, et al . RAN -binding protein 9 is involved [0113 ] 39. Komatsu M , et al. The selective autophagy in alternative splicing and is critical for male germ cell substrate p62 activates the stress responsive transcription development and male fertility . PLoS genetics 10 , factor Nrf2 through inactivation of Keap1 . Nature cell e1004825 (2014 ) . biology 12 , 213 - 223 ( 2010 ) . [0097 ] 23 . Gong F , Fahy D , Liu H , Wang W , Smerdon M [0114 ] 40 . Solis L M , et al. Nrf2 and Keap1 abnormalities J . Role of the mammalian SWI/ SNF chromatin remodel in non - small cell lung carcinoma and association with ing complex in the cellular response to UV damage . Cell clinicopathologic features . Clin Cancer Res 16 , 3743 cycle 7 , 1067- 1074 (2008 ) . 3753 ( 2010 ) . [0098 ] 24 . Lee J , Hirano T . RAD21L , a novel cohesin [0115 ] 41. Buchanan C L , et al. Locoregional recurrence subunit implicated in linking homologous chromosomes after mastectomy: incidence and outcomes . Journal of the in mammalian meiosis . J Cell Biol 192 , 263 - 276 ( 2011 ) . American College of Surgeons 203 , 469 - 474 ( 2006 ) . [0099 ] 25. Martini M , De Santis MC, Braccini L , Gulluni [0116 ] 42 . Huston T L , Simmons R M . Locally recurrent F , Hirsch E . PI3K / AKT signaling pathway and cancer : an breast cancer after conservation therapy . American jour updated review . Annals of medicine 46 , 372 - 383 ( 2014 ). nal of surgery 189 , 229 - 235 ( 2005 ) . [0100 ] 26 . Krasilnikov M , et al . Contribution of phospha [0117 ] 43. Early Breast Cancer Trialists ' Collaborative G , tidylinositol 3 -kinase to radiation resistance in human et al. Effect of radiotherapy after breast- conserving sur melanoma cells . Molecular carcinogenesis 24 , 64 -69 gery on 10 - year recurrence and 15 - year breast cancer ( 1999 ). death : meta - analysis of individual patient data for 10 , 801 [0101 ] 27 . Jiang Z , et al . Phosphatase and tensin homo women in 17 randomised trials . Lancet 378 , 1707 - 1716 logue deficiency in glioblastoma confers resistance to (2011 ) . radiation and temozolomide that is reversed by the pro [0118 ] 44 . Ebctcg , et al. Effect of radiotherapy after mas tease inhibitor nelfinavir . Cancer Res 67 , 4467 - 4473 tectomy and axillary surgery on 10 - year recurrence and (2007 ) . 20 - year breast cancer mortality : meta - analysis of indi [0102 ] 28 . Abazeed M E , et al . Integrative radiogenomic vidual patient data for 8135 women in 22 randomised profiling of squamous cell lung cancer. Cancer Res, trials . Lancet 383 , 2127 - 2135 (2014 ) . (2013 ). [0119 ] 45 . Liang K , Lu Y , Jin W , Ang K K , Milas L , Fan [0103 ] 29 . Taguchi K , Motohashi H , Yamamoto M . Z . Sensitization of breast cancer cells to radiation by Molecular mechanisms of the Keap1- Nrf2 pathway in trastuzumab . Mol Cancer Ther 2 , 1113 - 1120 ( 2003 ) . stress response and cancer evolution . Genes Cells 16 , [0120 ] 46 . Joon DL, et al. Supraadditive apoptotic 123 - 140 ( 2011 ) . response of R3327 - G rat prostate tumors to androgen [ 0104 ] 30 . Zhang P , et al. Loss of Kelch - like ECH -asso ablation and radiation . Int J Radiat Oncol Biol Phys 38 , ciated protein 1 function in prostate cancer cells causes 1071 - 1077 ( 1997 ). chemoresistance and radioresistance and promotes tumor [0121 ] 47 . Bolla M , et al. Long -term results with imme growth . Mol Cancer Ther 9 , 336 - 346 ( 2010 ) . diate androgen suppression and external irradiation in [0105 ] 31 . Hast B E , et al. Cancer -derived mutations in patients with locally advanced prostate cancer ( an KEAP1 impair NRF2 degradation but not ubiquitination . EORTC study ) : a phase III randomised trial . Lancet 360 , Cancer Res 74 , 808 - 817 (2014 ) . 103 - 106 (2002 ). [0106 ] 32 . Cancer Genome Atlas Research N . Compre [0122 ] 48 . D ' Amico AV, Manola J , Loffredo M , Renshaw hensive molecular profiling of lung adenocarcinoma . A A , DellaCroce A , Kantoff P W . 6 -month androgen Nature 511, 543 -550 ( 2014 ) . suppression plus radiation therapy vs radiation therapy 0107 ] 33 . Shibata T , et al. Cancer related mutations in alone for patients with clinically localized prostate cancer : NRF2 impair its recognition by Keap1 -Cul3 E3 ligase and a randomized controlled trial . Jama 292 , 821 - 827 ( 2004 ) . US 2017 /0319692 A1 Nov. 9 , 2017

[0123 ] 49. Pilepich M V , et al. Phase III radiation therapy [0138 ] 64 . Joe H . Relative Entropy Measures of Multi oncology group (RTOG ) trial 86 - 10 of androgen depri variate Dependence . Journal of the American Statistical vation adjuvant to definitive radiotherapy in locally Association 84 , 157 - 164 ( 1989 ) . advanced carcinoma of the prostate . Int J Radiat Oncol [0139 ] 65 . Barretina J, et al. The Cancer Cell Line Ency Biol Phys 50 , 1243 - 1252 (2001 ) . clopedia enables predictive modelling of anticancer drug [0124 ] 50 . Brys M , Wojcik M , Romanowicz -Makowska sensitivity. Nature 483 , 603 -607 (2012 ) . H , Krajewska W M . Androgen receptor status in female [0140 ] 66 . Cancer Genome Atlas Research N . Compre breast cancer: RT- PCR and Western blot studies. Journal hensive genomic characterization defines human glioblas of cancer research and clinical oncology 128 , 85 - 90 toma genes and core pathways . Nature 455 , 1061 - 1068 ( 2002 ) . ( 2008 ) . [0125 ] 51 . Barton V N , et al. Multiple molecular subtypes 10141 ) 67 . Olshen AB , Venkatraman ES, Lucito R , Wigler of triple- negative breast cancer critically rely on androgen M . Circular binary segmentation for the analysis of array receptor and respond to enzalutamide in vivo . Mol Cancer based DNA copy number data . Biostatistics 5 , 557 - 572 Ther 14 , 769 - 778 (2015 ) . ( 2004 ) . [0126 ] 52 . Cochrane D R , et al. Role of the androgen [ 0142 ] 68 . Korn JM , et al. Integrated genotype calling and receptor in breast cancer and preclinical analysis of association analysis of SNPs, common copy number enzalutamide . Breast cancer research : BCR 16 , R7 polymorphisms and rare CNVs . Nat Genet 40 , 1253 - 1260 (2014 ) . ( 2008 ) . (0127 ] 53 . Gucalp A , et al. Phase II trial of bicalutamide (0143 ] 69 . Mermel C H , Schumacher S E , Hill B , Mey in patients with androgen receptor -positive , estrogen erson ML, Beroukhim R , Getz G . GISTIC2. 0 facilitates receptor- negative metastatic Breast Cancer. Clin Cancer sensitive and confident localization of the targets of focal Res 19 , 5505 -5512 (2013 ) . somatic copy -number alteration in human cancers . [0128 ] 54 . Akiyama T , Matsuda S , Namba Y , Saito T , Genome biology 12 , R41 ( 2011 ) . Toyoshima K , Yamamoto T. The transforming potential of [0144 ] 70 . Li B , Dewey C N . RSEM : accurate transcript the c - erbB - 2 protein is regulated by its autophosphory quantification from RNA - Seq data with or without a lation at the carboxyl- terminal domain . Mol Cell Biol 11 , reference genome. BMC Bioinformatics 12 , 323 (2011 ) . 833 - 842 ( 1991 ) . [0145 ] 71 . Cancer Genome Atlas Research N . Compre [0129 ] 55 . Lehmann B D , et al. Identification of human hensive genomic characterization of squamous cell lung triple -negative breast cancer subtypes and preclinical cancers . Nature 489 , 519 -525 ( 2012 ) . models for selection of targeted therapies. The Journal of [0146 ] All publications and patents mentioned in the clinical investigation 121, 2750 -2767 (2011 ) . specification and / or listed below are herein incorporated by reference . Various modifications and variations of the [0130 ] 56 . Goodwin J F , et al. A hormone -DNA repair described method and system of the invention will be circuit governs the response to genotoxic insult . Cancer apparent to those skilled in the art without departing from Discov 3 , 1254 - 1271 ( 2013 ) . the scope and spirit of the invention . Although the invention [0131 ] 57 . Druker B J, et al. Efficacy and safety of a has been described in connection with specific embodi specific inhibitor of the BCR - ABL tyrosine kinase in ments , it should be understood that the invention as claimed chronic myeloid leukemia . N Engl J Med 344 , 1031 - 1037 should not be unduly limited to such specific embodiments . (2001 ). Indeed , various modifications of the described modes for [0132 ] 58 . Lee J M , Hanson J M , Chu W A , Johnson JA . carrying out the invention that are obvious to those skilled Phosphatidylinositol 3 -kinase , not extracellular signal in the relevant fields are intended to be within the scope regulated kinase , regulates activation of the antioxidant described herein responsive element in IMR - 32 human neuroblastoma We claim : cells . The Journal of biological chemistry 276 , 20011 1 . A method of treating cancer comprising : 20016 (2001 ) . a ) determining that a subject has androgen receptor posi [0133 ] 59 . Mitsuishi Y , et al. Nrf2 redirects glucose and tive cancer cells ; glutamine into anabolic pathways in metabolic repro b ) treating said subject with an anti - androgen receptor gramming. Cancer Cell 22 , 66 -79 ( 2012 ) . agent under conditions such that at least some or most [0134 ] 60 . Cowley G S , et al. Parallel genome- scale loss or all of said androgen receptor positive cancer cells are of function screens in 216 cancer cell lines for the sensitized to radiation therapy; and identification of context- specific genetic dependencies . c ) treating said subject with radiation therapy , wherein Scientific data 1 , 140035 (2014 ). said radiation therapy is performed at least 1 hour after [0135 ] 61. Franken N A , Rodermond H M , Stap J , Have said treating with said anti - androgen receptor agent, man J, van Bree C . Clonogenic assay of cells in vitro . Nat and wherein said treating causes at least a portion of Protoc 1 , 2315 - 2319 ( 2006 ) . said androgen receptor positive cancer cells to die . [0136 ] 62 . Cai Z , Chattopadhyay N , Liu W J , Chan C , 2 . The method of claim 1 , wherein said anti - androgen Pignol J P , Reilly R M . Optimized digital counting receptor agent sensitizes said androgen positive cancer cells colonies of clonogenic assays using ImageJ software and to said radiation therapy , such that a higher proportion of customized macros: comparison with manual counting . said cancer cells are killed by said radiation therapy than International journal of radiation biology 87 , 1135 - 1146 without said anti -androgen receptor agent. ( 2011) . 3 . The method of claim 1 , wherein said radiation therapy [ 0137 ] 63 . Linfoot E H . An informational measure of is performed at least 6 hours after said treating with said correlation . Information and Control 1 , 85 - 89 (1957 ) . anti- androgen receptor agent. US 2017 /0319692 A1 Nov. 9 , 2017

4 . The method of claim 1 , wherein said radiation therapy 13 . The method of claim 1 , wherein said androgen recep is performed at least 1 - 24 hours after said treating with said tor positive cancer cells are a type of cancer selected from anti -androgen receptor agent. the group consisting of: breast , prostate , colon , leukemia , 5 . The method of claim 1 , wherein said radiation therapy brain , bone , skin , liver , pancreatic , stomach , and lung. is performed at least 1 - 2 months after said treating with said 14 . The method of claim 1 , wherein said radiation therapy anti - androgen receptor agent. comprises subjecting said subject to X -rays , gamma rays , 6 . The method of claim 1 , wherein said anti- androgen and or charged particles . receptor agent is administered to said subject in a dosage 15 . A system comprising : between 1 mg/ kg to 35 mg/ kg . a ) a radiation therapy device , and 7 . The method of claim 1 , wherein said anti - androgen b ) an anti - androgen receptor agent. receptor agent is administered to said subject in a dosage of 16 . The system of claim 15 , wherein said radiation about 10 - 15 mg/kg . therapy device is configured to emit X - rays , gamma rays, or 8 . The method of claim 1 , wherein said determining charged particles for cancer treatment. comprises receiving or reviewing a report that said subject 17 . The system of claim 15 , wherein said anti- androgen has androgen positive cancer cells . receptor agent is an androgen receptor antagonists selected 9 . The method of claim 1 , wherein said determining from the group consisting of: flutamide , nilutamide , bicalu comprises performing an in vitro assay on a sample from tamide, enzalutamide, apalutamide , cyproterone acetate , said subject . megestrol acetate , chlormadinone acetate , spironolactone , 10 . The method of claim 1 , wherein said anti -androgen receptor agent is an androgen receptor antagonists selected canrenone , drospirenone , ketoconazole , topilutamide (fluri from the group consisting of: flutamide, nilutamide, bicalu dil ) , and cimetidine . tamide , enzalutamide , apalutamide, cyproterone acetate , 18 . The system of claim 15 , wherein said anti- androgen megestrol acetate , chlormadinone acetate , spironolactone , receptor agent is a selective androgen receptor modulator canrenone , drospirenone, ketoconazole , topilutamide ( fluri (SARM ) selected from the group consisting of: Enobosarm dil ), and cimetidine. (Ostarine , MK - 2866 , GTX - 024 ), BMS - 564 , 929 ; LGD - 4033 11 . The method of claim 1 , wherein said anti - androgen (Ligandrol ) , agent in U . S . Pat . No . 7 ,605 , 152 SARM ( 5 - 3 / receptor agent is a selective androgen receptor modulator 5 -6 ) ; AC - 262, 356 ; JNJ- 28330835 ; LGD - 2226 ; LGD - 3303 ; (SARM ) selected from the group consisting of: Enobosarm S - 40503 ; and S - 23 . ( Ostarine , MK - 2866 , GTX -024 ) , BMS - 564 , 929 ; LGD -4033 19 . The system of claim 15 , further comprising : c ) a report (Ligandrol ), agent in U . S. Pat. No . 7 ,605 , 152 SARM (5 -3 / that a subject has androgen receptor cancer cells, or c ) AR 5 -6 ) ; AC - 262, 356 ; JNJ- 28330835 ; LGD - 2226 ; LGD - 3303 ; positive cancer cells from said subject. S - 40503 ; and S -23 . 20 . The system of claim 15 , wherein said anti- androgen is 12 . The method of claim 1 , wherein said anti - androgen is an antibody, or fragment thereof, to the androgen receptor. an antibody, or fragment thereof, to the androgen receptor. * * * * *