USOO8440393B2

(12) United States Patent (10) Patent No.: US 8.440,393 B2 Birrer et al. (45) Date of Patent: May 14, 2013

(54) PRO-ANGIOGENIC IN OVARIAN OTHER PUBLICATIONS TUMORENDOTHELIAL CELL, SOLATES Boyd (The Basic Science of Oncology, 1992, McGraw-Hill, Inc., p. (75) Inventors: Michael J. Birrer, Mt. Airy, MD (US); 379). Tomas A. Bonome, Washington, DC Tockman et al. (Cancer Res., 1992, 52:2711s-2718s).* (US); Anil Sood, Pearland, TX (US); Pritzker (Clinical Chemistry, 2002, 48: 1147-1150).* Chunhua Lu, Missouri City, TX (US) Benedict et al. (J. Exp. Medicine, 2001, 193(1) 89-99).* Jiang et al. (J. Biol. Chem., 2003, 278(7) 4763-4769).* (73) Assignees: The United States of America as Matsushita et al. (FEBS Letters, 1999, vol. 443, pp. 348-352).* Represented by the Secretary of the Singh et al. (Glycobiology, 2001, vol. 11, pp. 587-592).* Department of Health and Human Abbosh et al. (Cancer Res. Jun. 1, 2006 66:5582-55.91 and Supple Services, Washington, DC (US); The mental Figs. S1-S7).* University of MD Anderson Cancer Zhai et al. (Chinese General Practice Aug. 2008, 11(8A): 1366 Center, Houston, TX (US) 1367).* Lu et al. (Cancer Res. Feb. 15, 2007, 64(4): 1757-1768).* (*) Notice: Subject to any disclaimer, the term of this Bagnato et al., “Activation of Mitogenic Signaling by Endothelin 1 in patent is extended or adjusted under 35 Ovarian Carcinoma Cells', Cancer Research, vol. 57, pp. 1306-1311, U.S.C. 154(b) by 194 days. 1997. Bouras et al., “Stanniocalcin 2 is an Estrogen-responsive (21) Appl. No.: 12/541,729 Coexpressed with the Estrogen Receptor in Human Breast Cancer', Cancer Research, vol. 62, pp. 1289-1295, 2002. (22) Filed: Aug. 14, 2009 Bryant et al., “EZH2 promotes proliferation and invasiveness of (Under 37 CFR 1.47) prostate cancer cells'. Prostate, vol. 67, No. 5, pp. 547-556, 2007. Bumgardner et al., "Chitosan: potential use as a bioactive coating for (65) Prior Publication Data orthopaedic and craniofacial f dental implants'. J. Biomater: Sci. US 2010/0286237 A1 Nov. 11, 2010 Polymer Edn, vol. 14, No. 5, pp. 423-438, 2003. Carson-Walter et al., “Cell surface tumor endothelial markers are conserved in mice and humans'. Cancer Res., vol. 61, No. 18, pp. Related U.S. Application Data 6649-6655, 2001. Chang et al., “The Murine Stanniocalcin 1 Gene Is Not Essential for (63) Continuation-in-part of application No. PCT/US2008/ Growth and Development'. Molecular and Cellular Biology, vol. 25. 054014, filed on Feb. 14, 2008. No. 23, pp. 10604-10610, 2005. Conejo-Garcia et al., “Vascular leukocytes contribute to tumor (60) Provisional application No. 60/901,455, filed on Feb. vascularization”. Blood, vol. 105, No. 2, pp. 679-681, 2005. 14, 2007. Donninger et al., “Whole genome expression profiling of advance stage papillary serous ovarian cancer reveals activated pathways'. (51) Int. Cl. Oncogene, vol. 23, No. 49, pp. 8065-8077, 2004. CI2O I/00 (2006.01) Dupraz et al., “Characterization of silane-treated hydroxyapatite CI2O I/68 (2006.01) powders for use as filler in biodegradable composites'. Journal of GOIN33/53 (2006.01) Biomedical Materials Research, vol. 30, pp. 231-238, 1996. GOIN33/567 (2006.01) GOIN33/574 (2006.01) (Continued) CI2N IS/II (2006.01) (52) U.S. Cl. USPC ...... 435/4; 435/6.1; 435/6.14; 435/7.1; Primary Examiner — Peter J Reddig 435/721: 435/723: 514/44. A (74) Attorney, Agent, or Firm — Klarduist Sparkman, LLP (58) Field of Classification Search ...... None See application file for complete search history. (57) ABSTRACT (56) References Cited A gene profiling signature for ovarian tumor endothelial cells is disclosed herein. The gene signature can be used to diag U.S. PATENT DOCUMENTS nosis or prognosis an ovarian tumor, identify agents to treat an 2003/0186909 A1 10/2003 McSwiggen ovariantumor, to predict the metastatic potential of an ovarian 2005/OO37389 A1 2/2005 Santin tumor and to determine the effectiveness of ovarian tumor 2005/0059682 A1* 3, 2005 Rubinfeld ...... 514,263.1 2006/0094.046 A1 5, 2006 Abo et al. treatments. Thus, methods are provided for identifying agents 2006/0.104981 A1* 5, 2006 Hikichi et al...... 424,155.1 that can be used to treat ovarian cancer, for determining the effectiveness of an ovarian tumor treatment, or to diagnose or FOREIGN PATENT DOCUMENTS prognose an ovarian tumor. Methods of treatment are also WO WO O2O83874 10, 2002 disclosed which include administering a composition that WO WOO3O80800 10, 2003 includes a specific binding agent that specifically binds to one WO WO 2004OO5883 1, 2004 of the disclosed ovarian endothelial cell tumor-associated WO WO 2004O15396 2, 2004 WO WO 2004 108896 12, 2004 molecules and inhibits ovarian tumor in the Subject. WO WO 2006085746 8, 2006 WO WO 2006138275 12/2006 WO WO 2007O16367 2, 2007 7 Claims, 20 Drawing Sheets US 8.440,393 B2 Page 2

OTHER PUBLICATIONS Narita et al., “Analysis of Heat Shock Related in Head-And-Neck Cancer Using cDNA Arrays'. Int. J. Radiation Garcia et al., “2-Methoxyestradiol Inhibits Prostate Tumor Develop Oncology Biol. Phys, vol. 53, No. 1, pp. 190-196, 2002. ment in Transgenic Adenocarcinoma of Mouse Prostate: Role of Tumor Necrosis Factor-O-Stimulated Gene 6', Clin Cancer Res, vol. Reedijk et al., “High-level Coexpression of JAG1 and NOTCH1 is 12, No. 3, pp.980-988, 2006. Observed in Human Breast Cancer and is Associated with Poor Halder et al., “Focal Adhesion Kinase Targeting Using In vivo Short Overall Survival”, Cancer Research, vol. 65, No. 18, pp. 8530-8537, Interfering RNA Delivery in Neutral Liposomes for Ovarian Carci 2005. noma Therapy”, Clin Cancer Res, vol. 12, No. 16, pp. 4916-4924, Voutilainen et al., “Versican in Epithelial Ovarian Cancer: Relation to 2006. Hyaluronan, Clinicopathologic Factors and Prognosis'. Int. J. Can Kleer et al., “EZH2 is a marker of aggressive breast cancer and cer, vol. 107, pp. 359-364, 2003. promotes neoplastic transformation of breast epithelial cells'. PNAS, Wang et al., “Identification of a novel function of TWIST, a bHLH vol. 100, No. 20, pp. 11606-11611, 2003. Lemaire et al., “Loss of HOP tumour suppressor expression in head , in the development of acquired taxol resistance in human and neck Squamous cell carcinoma'. British Journal of Cancer, vol. cancer cells'. Oncogene, vol. 23, No. 2, pp. 474-482, 2004. 91, pp. 258-261, 2004. Weiner et al., “Decreased Src Tyrosine Kinase Activity Inhibits Lu et al., “Selection of Potential Markers for Epithelial Ovarian Malignant Human Ovarian Cancer Tumor Growth in a Nude Mouse Cancer with Gene Expression Arrays and Recursive Descent Parti Model”, Clinical Cancer Research, vol. 5, pp. 2164-2170, 1999. tion Analysis”. Clinical Cancer Research, vol. 10, pp. 3291-3300. Yeung et al., "Cloning of a novel epidermal growth factor repeat 2004. Mok et al., “Biomarker Discovery in Epithelial Ovarian Cancer by containing gene EGFL6: expressed in tumor and fetal tissues'. Genomic approaches'. Advances in Cancer Research, vol. 96, pp. Genomics, vol. 62, No. 2, pp. 304-307, 1999. 1-22, 2006. Yeung et al., “Hypoxia-Inducible Factor-1-Mediated Activation of Nanda et al., “Identification of a Binding Partner for the Endothelial Stanniocalcin-1 in Human Cancer Cells'. Endocrinology, vol. 146, Cell Surface TEM7 and TEM7R, Cancer Research, vol. 64, No. 11, pp. 4951-4960, 2005. pp. 8507-851 1, 2004. Zhang et al., “Down-regulation of Jagged-1 induces cell growth Nanjundan et al., “Identification of a novel splice variant of AML1b inhibition and S phase arrest in prostate cancer cells'. Int. J. Cancer, in ovarian cancer patients conferring loss of wild-type tumor Sup vol. 119, pp. 2071-2077, 2006. pressive functions'. Oncogene, vol. 26, No. 18, pp. 2574-2584, E-published 2006. * cited by examiner U.S. Patent May 14, 2013 Sheet 1 of 20 US 8.440,393 B2

WoW

WN30

W . i gous valloo : x: 9Wg.TOO pop- zuwwav :::: ZsodSo

9W. N.

Xid

: on 1903

efuelo po affe env U.S. Patent May 14, 2013 Sheet 2 of 20 US 8.440,393 B2

Normal PTK2 Fyn MMP-9 32-arrestin Jagged 1 PLXDC1 2 4 6 Fold change (OD) U.S. Patent May 14, 2013 Sheet 3 of 20 US 8.440,393 B2

OOOO %

CN ais U.S. Patent May 14, 2013 Sheet 4 of 20 US 8.440,393 B2

000;,00800900€0 ?od?uolo40%

U.S. Patent May 14, 2013 Sheet 5 of 20 US 8.440,393 B2

CN N

s U.S. Patent May 14, 2013 Sheet 6 of 20 US 8.440,393 B2

Tumor Nodules

s G

CO a. D GS)

Tumor weight (g)

U.S. Patent May 14, 2013 Sheet 7 of 20 US 8.440,393 B2

Tumor weight (g) U.S. Patent May 14, 2013 Sheet 8 of 20 US 8.440,393 B2

–(+--

Tumor weight (g) Number of Nodules

9"SDI U.S. Patent May 14, 2013 Sheet 9 of 20 US 8.440,393 B2

0 5 10 15 20 25 Follow-up (years) tra 2 s sw *3. ow EAH2 s s *. Edo Wils * ... s * High Ezh2. 2 -X. Endo s 0 5 10 15 20 25 Follow-up (years)

5 s EZH2-Endo U.S. Patent May 14, 2013 Sheet 10 of 20 US 8.440,393 B2

F.G. 8

*p-0.05 **pz0.01 **pz0.001

p-value 0.0319 ©iNº. 10 EZH2 microarray expression (log2) U.S. Patent May 14, 2013 Sheet 11 of 20 US 8.440,393 B2

WASH1 Prooter C: control; W: VEGF (50 ng M-)

c is a g EZH2 is Histone 3 (ys 27) Actin S.d

WASH WASH1W

6 :

0.08

Contro VEGF - EZH2 siRNA VEGF SiRNA U.S. Patent May 14, 2013 Sheet 12 of 20 US 8.440,393 B2

FIG. 10

* <0.01

s!!!eo 350-JA

EZ-2 levels EZH2 levels -C

U.S. Patent May 14, 2013 Sheet 13 of 20 US 8.440,393 B2

U.S. Patent May 14, 2013 Sheet 14 of 20 US 8.440,393 B2

c a i See E C an ass

GN

vo O Y O WNAS peye N

i

9HO-WNAS WNSpeyeN

O 53its 4 Z U.S. Patent May 14, 2013 Sheet 15 of 20 US 8.440,393 B2

O O. O O N D : N ve eyedngo %

U.S. Patent May 14, 2013 Sheet 16 of 20 US 8.440,393 B2

F.G. 13

1600 EZH2 2400 Actin 800 Arbitrary units

EAH2 Hs SiRNACH d Control EH2 His EZH2 Pus

siRNAICH siRNAICH siRNACH EZH2 Mm siRNA/CH

C31

2 Mean tumor weight Mean tumor weight 1 8 SKOW3ipf1

1 * p-0.05 ** PC0.001 U.S. Patent May 14, 2013 Sheet 17 of 20 US 8.440,393 B2

F.G. 14

HeyA8 MOEC

LLI |N nr. ÇN

Actin

HeyA8 MOEC

LLI N ~~ CN

Actin U.S. Patent May 14, 2013 Sheet 18 of 20 US 8.440,393 B2

e /o

: s

wr N ed O (6) ufew Jouin. U.S. Patent May 14, 2013 Sheet 19 of 20 US 8.440,393 B2

3. Control EZH2 His EZH2 Mm EZH2 His plus SiRNACH SRNACH siRNAICH Min SiRNACH

MVD

No, of positive blood vesses

CD3

Desmin

Pericyte coverage (%) Merge CO31 & Desrin

6 O

O U.S. Patent US 8.440,393 B2

d Seo WNO emSodjo on

US 8,440,393 B2 1. 2 PRO-ANGOGENC GENES IN OVARAN chemoresistance and die as a result of their cancer. Thus, a TUMORENDOTHELIAL CELL, SOLATES need exists to identify alternative treatments for ovarian can C. CROSS REFERENCE TO RELATED APPLICATION SUMMARY OF THE DISCLOSURE This is a continuation-in-part application of International A gene profiling signature is disclosed herein that can be Patent Application PCT/US2008/054014, filed Feb. 14, 2008, used to predict clinical outcome and develop therapeutics for designating the United States and published in English as WO treating ovarian cancer in a Subject. For example, the ovarian 10 endothelial cell tumor-associated molecules identified by the 2008/101118, which claims the benefit of U.S. Provisional gene profile signature can serve as prognostic indicators as Application No. 60/901,455, filed on Feb. 14, 2007. The well as targets for specific therapeutic molecules that can entire contents of these prior applications are incorporated reduce or eliminate ovarian cancer. Thus, methods of identi herein by reference. fying an agent for treating an ovarian tumor are provided. In 15 Some examples, the methods include contacting a cell. Such as ACKNOWLEDGMENT OF GOVERNMENT an ovarian tumor cell or an ovarian tumor endothelial cell, SUPPORT with one or more test agents under conditions sufficient for the one or more test agents to alter the activity of at least one This invention was made with government Support under ovarian endothelial cell tumor-associated molecule listed in contract CA083639 awarded by the National Institutes of any of Tables 1-5. The method includes detecting the activity Health. The government has certain rights in the invention. of the at least one ovarian endothelial cell tumor-associated molecule in the presence and absence of the one or more test FIELD OF THE DISCLOSURE agents. The activity of the at least one ovarian endothelial cell tumor-associated molecule in the presence of the one or more This disclosure relates to the field of ovarian cancer and in 25 test agents is then compared to the activity in the absence of particular, to methods for treating ovarian cancer by targeting such agents to determine if there is differential expression of ovarian endothelial cell tumor-associated molecules identi the at least one ovarian endothelial cell tumor associated fied by an ovarian tumor endothelial cell gene expression molecule. Differential expression of the ovarian endothelial profile and methods for identifying therapeutic agents. cell tumor-associated molecule in the presence of the test 30 agent(s) indicates that the one or more test agents can be used BACKGROUND to treat an ovarian tumor. Methods are also provided for treating an ovarian tumor. In Ovarian cancer is the fifth most common form of cancer in Some examples, the method includes administering to the women in the United States, accounting for three percent of subject a therapeutically effective treatment to inhibit ovarian the total number of cancer cases and twenty-six percent of 35 tumor growth. In an example, the treatment includes admin those occurring in the female genital tract. The American istering a therapeutically effective amount of a specific bind Cancer Society estimated that 15.310 deaths would becaused ing agent that binds with high specificity to one of the ovarian in women living in the United States in 2006. A large majority endothelial cell tumor-associated molecules listed in Tables of women who die of ovarian cancer will have had serous 1, 2, 4 or 5 and alters expression or activity of the molecules, 40 thereby treating the ovarian tumor in the Subject (for example, carcinoma of the ovarian epithelium, a condition which by decreasing tumor vascular growth, tumor growth or tumor occurs in sixty percent of all cases of ovarian cancer (Boring Volume). In particular examples, the specific binding agent et al., Cancer J. Clin. 44: 7-26, 1994). preferentially binds to and inhibits expression or activity of Women with ovarian cancer are typically asymptomatic one of the ovarian endothelial cell tumor-associated mol until the cancer has metastasized. As a result, most women 45 ecules that is upregulated in an ovariantumor endothelial cell, with ovarian cancer are not diagnosed until the cancer has such as Zeste homologue 2 (EZH2), EGF-like domain, mul progressed to an advanced and usually incurable stage tiple 6 (EGFL6), tumor necrosis factor, alpha-induced protein (Boente et al., Curr: Probl. Cancer 20: 83-137, 1996). Sur 6 (TNFAIP6), Twist homologue 1 (TWIST1), stanniocalcin 1 vival rates are much better in women diagnosed with early (STC1), homeodomain-only protein (HOP), chondroitin sul stage ovarian cancers, about ninety percent of these women 50 fate proteoglycan 2 (CSPG2), and plexin domain containing are still alive five years after diagnosis. 1 (PLXDC1). In other particular examples, ovarian tumor Treatment of ovarian cancer typically involves a variety of growth is inhibited by the specific binding agent preferen treatment modalities. Generally, Surgical intervention serves tially binding to and inhibiting expression of one of the ova as the basis for treatment (Dennis S Chi & William J. Hoskins, rian endothelial cell tumor-associated molecules listed in any Primary Surgical Management of Advanced Epithelial Ova 55 of Tables 1, 2, 4 or 5 which are involved in angiogenesis, such rian Cancer, in Ovarian Cancer 241, Stephen C. Rubin & as molecules involved in cell proliferation, tube formation or Gregory P. Sutton eds., 2d ed. 2001). Treatment of serous cell motility and are upregulated in ovarian tumor endothelial carcinoma often involves cytoreductive Surgery (hysterec cells. tomy, bilateral salpingo-oophorectomy, omentectomy, and Methods are also provided for determining the effective lymphadenectomy) followed by adjuvant chemotherapy with 60 ness of an agent for the treatment of an ovarian tumor in a paclitaxel and either cisplatin or carboplatin (Eltabbakh, G. Subject with the ovarian tumor. In an example, the method H. & Awtrey, C.S., Expert Op. Pharmacother: 2010): 109-24, includes detecting expression of an ovarian endothelial cell 2001). tumor-associated molecule in a sample from the Subject fol Despite a clinical response rate of 80% to primary treat lowing administration of the agent. The expression of the ment with Surgery and chemotherapy, most Subjects experi 65 ovarian endothelial cell tumor-associated molecule following ence tumor recurrence within two years of treatment. The administration is compared to a control. Such as specific overwhelming majority of subjects will eventually develop binding agents that bind to and inhibit one of the ovarian US 8,440,393 B2 3 4 endothelial cell tumor-associated molecules listed in any of time quantitative RT-PCR data of selected genes from the Tables 1, 2, 4 or 5 that is upregulated in ovarian endothelial pro-angiogenic gene signature provided in Table 1. tumor cells. An alteration in the expression of the ovarian FIG. 2 is a graph illustrating protein expression in ovarian endothelial cell tumor-associated molecule (such as a endothelial cells for a subset of the proteins detected follow decrease in expression of a molecule upregulated in ovarian ing staining of samples with immunofluorescently-labeled tumor endothelial cells or an increase in expression of a PTK2, Fyn, MMP-9, B2-arrestin, Jagged 1 and PLXDC1. molecule downregulated in Such cells) following treatment FIGS. 3A, 3B and 3C are digital images of siRNA-medi indicates that the agent is effective for the treatment of the ovarian cancer in the Subject. In a specific example, the ated silencing of (FIG. 3A) EZH2. (FIG. 3B) Jagged 1 and method includes detecting and comparing the protein expres (FIG.3C) protein tyrosine kinase 2 (PTK2), as assessed using sion levels of the ovarian endothelial cell tumor-associated 10 Western blot analyses. molecules. In other examples, the method includes detecting FIG. 3D is a graph illustrating the effect of EZH2, Jagged1 and comparing the mRNA expression levels of the ovarian or PTK2 silencing on human umbilical vein endothelial cell endothelial cell tumor-associated molecules. (HUVEC) tube formation. Methods of diagnosing and prognosing an ovarian tumor FIG.3E is a graph illustrating the effect of EZH2 silencing (such as a tumor that overexpresses at least one of the dis 15 on HUVEC migration. closed ovarian endothelial cell tumor-associated molecules) FIG.3F is a graph illustrating the effect of Jagged 1 silenc are provided. In some examples, such methods are performed ing on HUVEC migration. prior to the treatment methods described herein. However, FIG.3G is a graph illustrating the effect of PTK2 silencing such methods can also be used independently of the disclosed on HUVEC migration. treatment methods. In particular examples, the method FIG. 4 is a digital image illustrating down regulation of includes determining the metastatic potential of an ovarian EZH2 by mouse EZH2 siRNA in mouse ovarian endothelial tumor in a Subject by detecting expression of at least one cells. ovarian endothelial cell tumor-associated molecule in a FIGS. 5A, 5B and 6 provide graphs illustrating the thera sample obtained from a subject with an ovarian tumor. The at peutic effects of siRNA-mediated EZH2 down regulation on least one ovarian endothelial cell tumor-associated molecule 25 HeyA8 (FIG. 5A) and SKOV3ip 1 (FIGS. 5B and 6) ovarian is involved in promoting angiogenesis, such as cell prolifera tumors. tion, cell motility or tube formation, such as EZH2. The method further includes comparing expression of the at least FIGS. 7A-7Hillustrate EZH2 expression in human ovarian one ovarian endothelial cell tumor-associated molecule in the carcinoma. FIG. 7A provides digital images representative of sample obtained from the subject with the ovarian tumor to a human tumors with low and high EZH2 expression based on control. An alteration in the expression of the at least one 30 immunohistochemical staining. FIG. 7B provides Kaplan ovarian endothelial cell tumor-associated molecule involved Meier curves of disease-specific mortality for patients whose in promoting angiogenesis indicates that the Subject has an ovarian tumors expressed high and low levels of EZH2 ovarian tumor with increased metastatic potential. (EZH2-T). The log-rank test (two-sided) was used to com In additional examples, methods are disclosed for predict pare differences between the two groups. Increased EZH2-T ing a clinical outcome in a subject with an ovarian tumor, Such 35 was significantly associated with decreased overall Survival as advanced stage epithelial ovarian cancer. In an example, (p<0.001). FIG.7C provides digital images representative of the methods include detecting expression of at least one ova human ovariantumor vasculature (arrowheads point to endot rian endothelial cell tumor-associated molecules listed in helial cells) with low and high immunohistochemical staining Tables 1-5 or combinations thereof (such as at least 1, at least for EZH2. FIG.7D provides Kaplan-Meier curves of disease 3, at least 5 or at least 10 of such molecules) in a sample 40 specific mortality of patients whose ovarian vasculature obtained from the subject with the ovarian tumor. The meth expressed low versus high EZH2 (EZH2-Endo) EZH2-Endo ods also can include comparing expression of the at least one was predictive of poor overall survival. FIG. 7E provides ovarian endothelial cell tumor-associated molecule in the digital images representative of human ovarian tumors with sample obtained from the subject with the ovarian tumor to a low or high immunohistochemical staining for VEGF. FIG. control (Such as a normal sample or value representing Such 45 7F provides a bar graphVEGF expression was strongly asso expression expected in a normal sample), wherein an alter ciated with high EZH2-Endo (p<0.01). FIG. 7G provides ation in the expression of the at least one ovarian endothelial digital images representative of human ovarian tumors with cell tumor-associated molecule indicates that the Subject has a decreased chance of Survival. For example, an alteration in low or high immunohistochemical staining for microVessel the expression, such as an increase in the expression of EZH2 density (MVD). FIG. 7H provides a bar graph illustrating indicates a poor prognosis, such as a decreased chance of 50 high MVD counts in a tumor were significantly associated Survival. In one example, a decreased chance of Survival with high EZH2-Endo expression (p<0.001). Images in pan includes a Survival time of equal to or less than a year. Alter els A, C, and E were taken at original magnification x200, and ations in the expression can be measured using methods in panelg at original magnification X200. known in the art, and this disclosure is not limited to particular FIGS. 8A-8C are graphs illustrating VEGF-increased methods. For example, expression can be measured at the 55 EZH2 expression in endothelial cells. Results in FIGS. 8A nucleic acid level (Such as by real time quantitative poly and 8B are in response to 6-hour treatments with EGF (25 merase chain reaction or microarray analysis) or at the protein ng/uI), VEGF (50 ng/ull), conditioned medium (CM) from level (such as by Western blot analysis). the non-cancerous ovarian epithelial cell line IOSE120, two The foregoing and other features of the disclosure will ovarian cancer cell lines OVCA420 and SKOV3, and com become more apparent from the following detailed descrip 60 plete medium with either 10% serum (A) or 2% serum (B). tion of several embodiments which proceeds with reference Percent fold changes represent the mean+/-S.d. of triplicate to the accompanying figures. experiments compared to untreated control cells. *p-0.05; **p<0.01: ***p<0.001. FIG. 8A illustrates that EZH2 pro BRIEF DESCRIPTION OF THE FIGURES moter activity is increased in an endothelial cell line in 65 response to EGF, VEGF, and conditioned media from ovarian FIG. 1 is a graph illustrating the comparative fold change in cancer cell lines. EAhy926 hybridoma endothelial cell line relative expression levels between microarray data and real was cotransfected with the Renilla luciferase plasmid and US 8,440,393 B2 5 6 firefly luciferase plasmid either with or without the EZH2 (CH:TPP). FIG. 11C is a graph illustrating that Zeta potential promoter construct followed by treatment with EGF, VEGF of siRNA/CH nanoparticles showed slight positive charge. and conditioned medium and promoter activity was deter FIG. 11D is a graph illustrating incorporation efficiency of mined. FIG. 8B illustrates that EZH2 mRNA levels are siRNA into CH nanoparticles with 3:1 ratio of CH:TPP increased in HUVEC in response to EGF, VEGF, and condi resulting in >75% incorporation efficiency. FIG. 11E is a tioned media from ovarian cancer cell lines. Cells were digital image following atomic force microscopy (AFM) treated as indicated and purified RNA was used in real-time demonstrating that siRNA/CH nanoparticles were spherical quantitative RT-PCR. Control values were normalized using and <150 nm in size. 3 housekeeping genes. FIG. 8C Pearsons analysis shows FIGS. 12A-12E illustrate incorporation, stability and intra significant correlation between EZH2 and VEGF expression 10 cellular uptake of siRNA/CH nanoparticles. FIG. 12A is a values (Log) from 29 microdissected high-grade serous pap digital image illustrating electrophoretic migration of naked illary ovarian adenocarcinomas. siRNA and siRNA/CH nanoparticles. SiRNA/CH nanopar FIGS.9A-9E show EZH2 gene silencing increases VASH1 ticles (open arrow) remained at top of the gel compared to mRNA expression in endothelial cells. FIG. 9A is a digital naked siRNA (solid arrow), which migrated downward. FIG. image of a polyacrylamide gel illustrating PCR products gen 15 12B is a digital image illustrating electrophoretic migration erated by as ChIP assay of EZH2 binding to human VASH1 of siRNA/CH nanoparticles in the presence of 50% serum. promoter in response to VEGF in HUVEC. Cross-linked SiRNA/CH nanoparticles were collected at different time chromatin from HUVEC was treated with (+) or without (-) points of incubation at 37°C. (Lane 1; naked siRNA, Lanes 2 VEGF and immunoprecipitated (IP) using EZH2 or mouse to 5; siRNA/CH nanoparticles). Naked siRNA (solid arrow) IgG antibodies. The input and immunoprecipitated DNA was degraded over 12 to 24 hours in serum containing media; were subjected to PCR using primers corresponding to the whereas CH nanoparticles (open arrow) protected the siRNA 3800 to 3584 base pairs upstream of VASH1 transcription from degradation in serum. Increased binding efficiency of start site. PCR products were examined on ethidium bromide siRNA/CH nanoparticles was noted compared to naked stained agarose gel. FIG.9B is a bar graph illustrating EZH2 siRNA. FIG. 12C is a fluorescence microscopy digital image mRNA levels in cells transfected with control or mouse EZH2 25 of HeyA8 cells after incubating either with siRNA alone or siRNA and harvested after 72 hours. RNA was isolated and with siRNA/CH nanoparticles at 4°C. for 20 minutes in PBS. subjected to real-time quantitative RT-PCR. The fold differ FIG. 12D is a series of tracings from a flow cytometry analy ence in levels of EZH2 mRNA represents the mean of tripli sis demonstrating that uptake efficiency of nanoparticles into cate experiments compared to control siRNA treated cells. cells was increased by 72-fold after incubating cells in PBS at Error bars represents.e.m. *p-0.05. FIG. 9C is a bar graph 30 4°C. for 20 minutes. FIG. 12E is graphical representation of illustrating the fold difference in levels of VASH1 mRNA as percentage of uptake of Alexa-555 siRNA by cells by flow compared to control siRNA treated cells. Error bars represent cytometry analysis. s.e.m. *p-0.01. FIG.9D illustrates the effect of EZH2 gene FIGS. 13 A-13E illustrate in vivo siRNA delivery using CH silencing on methylation status of VASH1 in VEGF-treated nanoparticles and the distribution of siRNA following single MOECs as detected by methylation specific PCR. The inhibi 35 intravenous injection of Alexa-555 siRNA/CH nanoparticles tory units of methylated VASH1 were normalized by that of in orthotopic HeyA8 tumor bearing nude mice. FIG. 13A the un-methylated VASH1 and represent the mean of tripli provides a pair of digital images illustrating fluorescent cate experiments. FIG.9E is a digital image of a Western blot siRNA distribution in tumor tissue of hematoxylin and eosin, of lysate collected 48 hours after transfection of MOEC with original magnification x200 (left); stained with anti-CD31 control, VEGF treated and mouse EZH2 siRNA treated cells. 40 (green) antibody to detect endothelial cells (right). FIG. 13B FIGS. 10A-10C show E2F transcription factors increases provides a pair of digital images of 50-um sections stained upon VEGF treatment in MOEC. FIG. 10A is a bar graph with CytoX Green and examined with confocal microscopy illustrating expression levels of E2F transcription factors in (original magnification x400) (left); lateral view (right) MOEC. Cells were treated with VEGF for 6 hours and Sub Images taken every 1 um were stacked and examined from the jected to Q-RT-PCR. FIG. 10B provides a pair of bar graphs 45 lateral view. Nuclei were labeled green and fluorescent illustrating silencing of E2F1, E2F3 and E2F5 transcription siRNA (red) was seen throughout the section. At all time factors by targeted siRNA in MOEC. Cells were transfected points, punctated emissions of the siRNA were noted in the with corresponding siRNAs. After 24 hours and 48 hours, perinuclear regions of individual cells, and siRNA was seen cells were collected; RNA was isolated and was subjected to in >80% of fields examined. (c) Western blot of lysates from real-time Q-RT-PCR. E2F3 and E2F5 gene silencing 50 orthotopic tumors collected 24, 48, 72 and 96 hours after a decreases EZH2 expression levels. EZH2 expression levels single injection of control siRNA/CH or human (EZH2 Hs were analyzed in E2F1, E2F3 and E2F5 silenced samples siRNA/CH). FIG.13D provides multiple digital images illus using Q-RT-PCR. The fold difference in levels of mRNA trating EZH2 gene silencing in HeyA8 tumor as well as tumor expression represents the mean of triplicate experiments endothelial cells. Tumors were collected after 48 hours of compared to cells (A) and VEGF treated cells (B). Error bars 55 single injection of control siRNA/CH, or EZH2 Hs siRNA/ represents.e.m. *p-0.01. FIG. 10C is a bar graph illustrating CH, or EZH2 Mm siRNA/CH and stained for EZH2 (green) the effect of VASH1 gene silencing on tube formation in and CD31 (red). Images were taken at original magnification, endothelial cells. HUVECs were plated on Matrigel after x200. FIG. 13E is a pair of graphs illustrating the effects of transfecting the cells with either control or human VASH1 EZH2 HS SiRNA/CH or EZH2 Mm SiRNA/CH on tumor siRNA. Vascular tube formation was evaluated by micro 60 weight in mouse orthotopic tumor models. Nude mice were scopic observation. injected with HeyA8 or SKOV3ip 1 ovarian cancer cells and 1 FIGS. 11A-11E illustrate the physical characteristics of week later, were randomly assigned (10 mice per group) to siRNA/CH nanoparticles. FIG. 11A is a table providing the receive therapy: (1) control siRNA/CH, (2) EZH2 Hs siRNA/ composition of CH/TPP/siRNAs. FIG. 11B is a graph illus CH, (3) EZH2 Mm siRNA/CH, and (4) combination of EZH2 trating the mean particle size of siRNA/CH particles as mea 65 Hs siRNA/CH plus EZH2 Mm siRNA/CH. Mice were sacri Sured using light scattering with a particle analyzer, showing ficed when any animals in control or a treatment group that nanoparticles maintained 100-200 nm size up to 7:1 ratio became moribund (after 3 to 4 weeks of therapy) and mouse US 8,440,393 B2 7 8 weight, tumor weight and tumor location were recorded. (Folkman, J. Nat. Canc. Inst. 82: 4-6, 1990). Targeting tumor Error bars represents.e.m. *p-0.05: **p<0.001. angiogenesis by inhibiting endothelial cells that Support FIG. 14 provides a series of digital images of Western blots tumor growth is particularly promising because of their pre of lysate collected 72 hours after transfection of HeyA8 cells Sumed genetic stability. The recent Success of a humanized or MOEC with control, human EZH2, or mouse EZH2 5 monoclonal antibody bevacizumab (trade name AvastinR) siRNA. against vascular endothelial growth factor in prolonging the FIG. 15 is a pair of graphs illustrating the weight distribu lives of patients with advanced colon and breast carcinoma tion of HeyA8 and SKOV3ip1 tumors. Seven days following demonstrates the promise of such approaches (Hurwitz et al., tumor cell injection, mice were randomly divided into 4 N. Engl. J. Med. 350: 2335-2342, 2004 and Jain et al., Nat. groups (10 mice per group) to receive therapy: (1) control 10 Clin. Pract. Oncol. 3: 24-40, 2006). However, the full spec siRNA/CH, (2) EZH2 Hs siRNA/CH, (3) EZH2 Mm siRNA/ trum of differences in the tumor vasculature compared to its CH, and (4) combination of EZH2 Hs siRNA/CH plus EZH2 normal counterpart is not known. Identification of additional Mm siRNA/CH. Mice were sacrificed when any animals in targets on tumor endothelium may allow opportunities for control or a treatment group became moribund (after 3 to 4 developing new therapeutic approaches to inhibit angiogen weeks of therapy) and tumor weight was recorded. 15 esis in a tumor-specific manner. FIGS. 16A-16B(A) Effect of tumor (EZH2 Hs siRNA/CH) In recent years, whole genome expression profiling of can or endothelial (EZH2Mm siRNA/CH) targeted EZH2 siRNA cer using methods such as microarray and serial analysis of on MVD and pericyte coverage. Tumors harvested following gene expression (SAGE) have provided insight into the 3 to 4 weeks of therapy were stained for CD31 (MVD; red) molecular pathways involved in cancer onset and progres and desmin (pericyte coverage; green). All pictures were 20 Sion. While selected genes in ovarian cancer vasculature have taken at original magnification X200. The bars in the graphs been characterized, there is little information regarding glo correspond sequentially to the labeled columns of images at bal gene expression alterations in ovarian cancer endothe left. Error bars represent s.e.m. *p-0.05; **p<0.001. (B) lium. Effects of VASH1 gene silencing on tumor growth in vivo. Disclosed herein is a gene expression signature identifying Nude mice were injected with SKOV3ip 1 ovarian cancer 25 endothelial cell tumor-associated molecules in ovarian tumor cells and 1 week later, were randomly divided into 5 groups endothelial cell isolates. Endothelial cells were purified from (10 mice per group): (1) control siRNA/CH, (2) EZH2 Mm human ovarian tissues and invasive ovarian epithelial cancers, siRNA1/CH, (3) EZH2 Mm siRNA2/CH (4) EZH2 Mm and a gene expression profile was established for ovarian siRNA3/CH (5) VASH1 Mm siRNA1/CH and (6) combina tumor endothelial cells using microarray analyses. The gene tion of EZH2 Mm siRNA1/CH plus VASH1 Mm siRNA/CH. 30 expression profile disclosed herein identifies genes whose FIG.16B is a bar graph illustrating the number of cells that expression is differentially regulated in tumor versus normal migrated in the presence and absence of VASH1 siRNA. Mice endothelial cells. This profile reveals distinct expression pro were sacrificed when any animals in control or a treatment files for tumor endothelial cell isolates as compared to non group became moribund (after 3 to 4 weeks of therapy) and tumor endothelial isolates. mouse weight, tumor weight and tumor location were 35 The disclosed gene expression profile also reveals genes recorded. Error bars represents.e.m. *p-0.05. and collections or sets of genes that serve as effective molecu FIG. 17 illustrates the effects of EZH2 His siRNA/CH or lar markers for angiogenesis in ovarian cancer, predict clini EZH2 Mm siRNA/CH on proliferation. Tumors were har cal outcome as well as Such genes or gene sets that can vested following 3-4 weeks of therapy and then stained for provide clinically effective therapeutic targets for ovarian proliferating cell nuclear antigen (PCNA). All images were 40 cancer. This has significant implications for the treatment of taken at original magnification x 100. The bars in the graphs ovarian cancer. For example, methods are disclosed for treat correspond sequentially to the labeled columns of images at ing ovarian cancer (for example, reducing or inhibiting ova left. Error bars represents.e.m. *p-0.05. rian cancer growth by targeting ovarian endothelial cell tumor-associated molecules, such as molecules believed to be SEQUENCE LISTING 45 involved in angiogenesis). For example, molecules involved in cell motility, tube formation or cell proliferation can be The nucleic and amino acid sequences listed in the accom identified by the gene profile signature. In an example, a panying sequence listing are shown using standard letter therapeutically effective amount of a specific binding agent is abbreviations for nucleotide bases, and three letter code for administered to a Subject. For example, the specific binding amino acids, as defined in 37 C.F.R. 1.822. Only one strand of 50 agent preferentially binds to one or more of the identified each nucleic acid sequence is shown, but the complementary ovarian endothelial cell tumor-associated molecules listed in strand is understood as included by any reference to the any of Tables 1, 2, 4, 5 or a combination thereof to alter the displayed strand. In the accompanying sequence listing: expression or activity of such molecule (e.g., increase expres SEQ ID NOS: 1, 6-27, and 44-47 are nucleic acid sion or activity of a molecule that is downregulated in ovarian sequences of exemplary primers. 55 endothelial tumor cells or decrease expression or activity of a SEQID NOS: 2-5 and 28-43 are nucleic acid sequences of molecule that is upregulated in Such cells). In one example, exemplary siRNAs. the specific binding agent preferentially binds to one or more of the identified ovarian endothelial cell tumor-associated DETAILED DESCRIPTION OF SEVERAL molecules listed in any of Tables 1, 2, 4, 5 or a combination EMBODIMENTS 60 thereofthat are upregulated in ovarian endothelial tumor cells (as indicated by a positive fold change in Table 1) to decrease Despite improvements in Surgery and chemotherapy, mor expression or activity of the one or more molecules. As a tality rates in women with advanced ovarian carcinoma have result, ovarian cancer in the subject is thereby reduced or remained largely unchanged (Cannistra, N. Engl.J.Med. 329: eliminated. In a particular example, the specific binding agent 1550-1559, 1993). Therefore, novel therapeutic strategies are 65 is an inhibitor, such as a siRNA, of one or more of the dis needed. Growth of tumors, both at the primary and metastatic closed ovarian endothelial cell tumor-associated molecules sites, requires a blood Supply for expansion beyond 1-2 mm described in any of Table 1, 2, 4 or 5 whose expression is US 8,440,393 B2 10 upregulated in ovarian endothelial tumor cells, such as EZH2. under conditions that allow for hybridization of the primers to In some examples, the specific binding agent preferentially a nucleic acid molecule in the sample. The primers are binds to one or more of the identified ovarian endothelial cell extended under suitable conditions, dissociated from the tem tumor-associated molecules listed in any of Tables 1,2,4, 5 or plate, and then re-annealed, extended, and dissociated to a combination thereof that are downregulated in ovarian amplify the number of copies of the nucleic acid molecule. endothelial tumor cells (as indicated by a negative fold Other examples of in vitro amplification techniques include change in Table 1) to increase expression or activity of the one quantitative real-time PCR, strand displacement amplifica or more molecules. As a result, ovarian cancer in the Subject tion (see U.S. Pat. No. 5,744,311); transcription-free isother is thereby reduced or eliminated. mal amplification (see U.S. Pat. No. 6,033,881); repair chain 10 reaction amplification (see WO 90/01069); ligase chain reac Terms tion amplification (see EP-A-320 308); gap filling ligase chain reaction amplification (see U.S. Pat. No. 5,427.930); The following explanations of terms and methods are pro coupled ligase detection and PCR (see U.S. Pat. No. 6,027, vided to better describe the present disclosure and to guide 889); and NASBATM RNA transcription-free amplification those of ordinary skill in the art in the practice of the present 15 (see U.S. Pat. No. 6,025,134). disclosure. The singular forms “a” “an and “the refer to A commonly used method for real-time quantitative poly one or more than one, unless the context clearly dictates merase chain reaction involves the use of a double stranded otherwise. For example, the term “comprising a nucleic acid DNA dye (such as SYBR Green I dye). For example, as the molecule' includes single or plural nucleic acid molecules amount of PCR product increases, more SYBR Green I dye and is considered equivalent to the phrase “comprising at binds to DNA, resulting in a steady increase in fluorescence. least one nucleic acid molecule.” The term 'or' refers to a Another commonly used method is real-time quantitative single element of stated alternative elements or a combination TaqMan PCR (Applied Biosystems). The 5' nuclease assay of two or more elements, unless the context clearly indicates provides a real-time method for detecting only specific ampli otherwise. As used herein, “comprises' means “includes.” fication products. The use of fluorogenic probes makes it Thus, "comprising A or B.’ means “including A, B, or A and 25 possible to eliminate post-PCR processing for the analysis of B. without excluding additional elements. probe degradation. The probe is an oligonucleotide with both Unless explained otherwise, all technical and scientific a reporterfluorescent dye and a quencher dye attached. While terms used herein have the same meaning as commonly the probe is intact, the proximity of the quencher greatly understood to one of ordinary skill in the art to which this reduces the fluorescence emitted by the reporter dye by För disclosure belongs. Although methods and materials similar 30 ster resonance energy transfer (FRET) through space. Probe or equivalent to those described herein can be used in the design and synthesis has been simplified by the finding that practice or testing of the present disclosure, suitable methods adequate quenching is observed for probes with the reporter and materials are described below. The materials, methods, at the 5' end and the quencher at the 3' end. and examples are illustrative only and not intended to be Angiogenesis: limiting. 35 A physiological process involving the growth of new blood Administration: vessels from pre-existing vessels. Angiogenesis can occur To provide or give a subject an agent, such as a chemo under normal physiological conditions such as during growth therapeutic agent, by any effective route. Exemplary routes of and development or wound healing (known as physiological administration include, but are not limited to, injection (Such angiogenesis) as well as pathological conditions such as in as Subcutaneous, intramuscular, intradermal, intraperitoneal, 40 the transition of tumors from a dormant state to a malignant and intravenous), oral, Sublingual, rectal, transdermal, intra state (known as pathological angiogenesis). As used herein, nasal, vaginal and inhalation routes. pro-angiogenic genes are genes that facilitate angiogenesis, Agent: Such as angiogenesis in an ovarian tumor. Any protein, nucleic acid molecule, compound, Small mol The complex phenomenon of angiogenesis begins with ecule, organic compound, inorganic compound, or other mol 45 degradation of the basement membrane by cellular proteases. ecule of interest. Agent can include a therapeutic agent, a This allows endothelial cells to penetrate and migrate (pro diagnostic agent or a pharmaceutical agent. A therapeutic or cess known as cell motility) into the extracellular matrix and pharmaceutical agent is one that alone or together with an then proliferate. In the final stages of this process, the endot additional compound induces the desired response (such as helial cells align themselves to form capillary or tubelike inducing a therapeutic or prophylactic effect when adminis 50 structures (process known as tube formation). These new tered to a Subject). In a particular example, a pharmaceutical structures then form a network that undergoes significant agent (Such as a siRNA to any of the genes listed in Tables 2 remodeling and rearrangement before fully functioning cap and Table 4) significantly reduces angiogenesis. A test agent illaries exist. Therefore, angiogenesis can be studied or iden is any Substance, including, but not limited to, a protein (Such tified by monitoring tube formation, cell motility, and/or cell as an antibody), nucleic acid molecule (such as a siRNA), 55 proliferation. organic compound, inorganic compound, or other molecule Antibody: of interest. In particular examples, a test agent can permeate a A polypeptide ligand including at least a light chain or cell membrane (alone or in the presence of a carrier). heavy chain immunoglobulin variable region which specifi Amplifying a Nucleic Acid Molecule: cally recognizes and binds an epitope of an antigen, such as an To increase the number of copies of a nucleic acid mol 60 ovarian endothelial cell tumor-associated molecule or a frag ecule. Such as a gene or fragment of a gene, for example a ment thereof. Antibodies are composed of a heavy and a light region of an ovarian endothelial cell tumor-associated gene. chain, each of which has a variable region, termed the variable The resulting products are called amplification products. heavy (V) region and the variable light (V) region. An example of in vitro amplification is the polymerase Together, the V region and the V, region are responsible for chain reaction (PCR), in which a biological sample obtained 65 binding the antigen recognized by the antibody. In one from a subject (Such as a sample containing ovarian cancer example, an antibody specifically binds to one of the proteins cells) is contacted with a pair of oligonucleotide primers, listed in Tables 8 and 9. US 8,440,393 B2 11 12 This includes intact immunoglobulins and the variants and with immune spleen cells. Monoclonal antibodies include portions of them well known in the art, such as Fab' frag humanized monoclonal antibodies. ments, F(ab)' fragments, single chain Fv proteins (“scFv'). A "polyclonal antibody is an antibody that is derived from and disulfide stabilized Fv proteins (“dsEv'). A scFv protein different B-cell lines. Polyclonal antibodies are a mixture of is a fusion protein in which a light chain variable region of an immunoglobulin molecules secreted against a specific anti immunoglobulin and a heavy chain variable region of an gen, each recognizing a different epitope. These antibodies immunoglobulin are bound by a linker, while in dsEvs, the are produced by methods known to those of skill in the art, for chains have been mutated to introduce a disulfide bond to instance, by injection of an antigen into a Suitable mammal stabilize the association of the chains. The term also includes (such as a mouse, rabbit or goat) that induces the B-lympho 10 cytes to produce IgG immunoglobulins specific for the anti genetically engineered forms such as chimericantibodies (for gen which are then purified from the mammals serum. example, humanized murine antibodies), heteroconjugate A “chimeric antibody' has framework residues from one antibodies (such as, bispecific antibodies). See also, Pierce species, such as human, and CDRS (which generally confer Catalog and Handbook, 1994-1995 (Pierce Chemical Co., antigen binding) from another species, such as a murine anti Rockford, Ill.); Kuby, J., Immunology, 3" Ed., W.H. Freeman 15 body that specifically binds an ovarian endothelial cell tumor & Co., New York, 1997. associated molecule. Typically, a naturally occurring immunoglobulin has A "humanized' immunoglobulin is an immunoglobulin heavy (H) chains and light (L) chains interconnected by dis including a human framework region and one or more CDRS ulfide bonds. There are two types of light chain, lambda and from a non-human (for example a mouse, rat, or synthetic) kappa. There are five main heavy chain classes (or isotypes) immunoglobulin. The non-human immunoglobulin provid which determine the functional activity of an antibody mol ing the CDRs is termed a “donor” and the human immuno ecule: IgM, Ig|D, IgG, IgA and IgE. globulin providing the framework is termed an “acceptor.” In Each heavy and light chain contains a constant region and one embodiment, all the CDRs are from the donor immuno a variable region, (the regions are also known as “domains'). globulin in a humanized immunoglobulin. Constant regions In combination, the heavy and the light chain variable regions 25 need not be present, but if they are, they must be substantially specifically bind the antigen. Light and heavy chain variable identical to human immunoglobulin constant regions, e.g., at regions contain a “framework” region interrupted by three least about 85-90%, such as about 95% or more identical. hyperVariable regions, also called “complementarity-deter Hence, all parts of a humanized immunoglobulin, except mining regions” or “CDRs'. The extent of the framework possibly the CDRs, are substantially identical to correspond region and CDRs have been defined (see, Kabat et al., 30 ing parts of natural human immunoglobulin sequences. Sequences of Proteins of Immunological Interest, U.S. Humanized immunoglobulins can be constructed by means Department of Health and Human Services, 1991, which is of genetic engineering (see for example, U.S. Pat. No. 5,585, hereby incorporated by reference). The Kabat database is 089). now maintained online. The sequences of the framework Array: regions of different light or heavy chains are relatively con 35 An arrangement of molecules, such as biological macro served within a species. The framework region of an antibody, molecules (such as peptides or nucleic acid molecules) or that is the combined framework regions of the constituent biological samples (such as tissue sections), in addressable light and heavy chains, serves to position and align the CDRS locations on or in a Substrate. A "microarray' is an array that in three-dimensional space. is miniaturized so as to require or be aided by microscopic The CDRs are primarily responsible for binding to an 40 examination for evaluation or analysis. epitope of an antigen. The CDRS of each chain are typically The array of molecules (“features') makes it possible to referred to as CDR1, CDR2, and CDR3, numbered sequen carry out a very large number of analyses on a sample at one tially starting from the N-terminus, and are also typically time. In certain example arrays, one or more molecules (such identified by the chain in which the particular CDR is located. as an oligonucleotide probe) will occur on the array a plurality Thus, a V. CDR3 is located in the variable domain of the 45 of times (such as twice), for instance to provide internal heavy chain of the antibody in which it is found, whereas a V, controls. The number of addressable locations on the array CDR1 is the CDR1 from the variable domain of the light can vary, for example from at least one, to at least 2, to at least chain of the antibody in which it is found. An antibody that 5, to at least 10, at least 20, at least 30, at least 50, at least 75, binds RET will have a specific V region and the V, region at least 100, at least 150, at least 200, at least 300, at least 500, sequence, and thus specific CDR sequences. Antibodies with 50 least 550, at least 600, at least 800, at least 1000, at least different specificities (such as different combining sites for 10,000, or more. In particular examples, an array includes different antigens) have different CDRs. Although it is the nucleic acid molecules, such as oligonucleotide sequences CDRs that vary from antibody to antibody, only a limited that are at least 15 nucleotides in length, such as about 15-40 number of amino acid positions within the CDRs are directly nucleotides in length. In particular examples, an array involved in antigenbinding. These positions within the CDRs 55 includes oligonucleotide probes or primers which can be used are called specificity determining residues (SDRs). to detect sensitive to ovarian endothelial cell tumor-associ References to “V” or “VH’ refer to the variable region of ated molecule sequences, such as at least one of those listed in an immunoglobulin heavy chain, including that of an Fv, Table 1, such as at least 5, at least 7, at least 10, at least 20, at scEv, dsFv or Fab. References to “V,” or “VL refer to the least 30, at least 40, at least 50, at least 60, at least 70, at least variable region of an immunoglobulin light chain, including 60 80, at least 90, at least 100, at least 150, at least 200, at least that of an Fv, scFv, dsFv or Fab. 250, at least 300, at least 400, at least 500, at least 600, at least A "monoclonal antibody' is an antibody produced by a 700, at least 800, at least 900, at least 1000, or at least 1100 single clone of B-lymphocytes or by a cell into which the light sequences listed in Table 1 (for example, 2, 6, 8, 9, 10, 11, 12. and heavy chain genes of a single antibody have been trans 13, 14, 15, 16, 17, 18, 19, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, fected. Monoclonal antibodies are produced by methods 65 70, 75, 80, 85,90, 95, 100,105,110, 120, 130, 140, 150, 175, known to those of skill in the art, for instance by making 225, 275,325, 350,375, 450, 550, 650, 750, 850, 950, 1050 hybrid antibody-forming cells from a fusion of myeloma cells or 1149 of those listed). In an example, the array is a com US 8,440,393 B2 13 14 mercially available such as a U133 Plus 2.0 oligonucleotide analog has bound to its target, there is a Sudden increase in array from Affymetrix. (Affymetrix, Santa Clara, Calif.). absorption at a characteristic temperature as the oligonucle Within an array, each arrayed sample is addressable, in that otide (or analog) and target disassociate from each other, or its location can be reliably and consistently determined melt. In another example, the method involves detecting a within at least two dimensions of the array. The feature appli signal. Such as a detectable label, present on one or both cation location on an array can assume different shapes. For nucleic acid molecules (or antibody or protein as appropri example, the array can be regular (such as arranged in uni ate). form rows and columns) or irregular. Thus, in ordered arrays The binding between an oligomer and its target nucleic the location of each sample is assigned to the sample at the acid is frequently characterized by the temperature (T) at time when it is applied to the array, and a key may be provided 10 which 50% of the oligomer is melted from its target. A higher in order to correlate each location with the appropriate target (T) means a stronger or more stable complex relative to a or feature position. Often, ordered arrays are arranged in a complex with a lower (T). symmetrical grid pattern, but samples could be arranged in Biological Activity: other patterns (such as in radially distributed lines, spiral An expression describing the beneficial or adverse effects lines, or ordered clusters). Addressable arrays usually are 15 of an agent on living matter. When the agent is a complex computer readable, in that a computer can be programmed to chemical mixture, this activity is exerted by the substances correlate a particular address on the array with information active ingredient or pharmacophore, but can be modified by about the sample at that position (such as hybridization or the other constituents. Activity is generally dosage-depen binding data, including for instance signal intensity). In some dent and it is not uncommon to have effects ranging from examples of computer readable formats, the individual fea beneficial to adverse for one substance when going from low tures in the array are arranged regularly, for instance in a to high doses. In one example, a specific binding agent sig Cartesian grid pattern, which can be correlated to address nificantly reduces the biological activity of the one or more information by a computer. ovarian endothelial cell tumor-associated molecules that is Protein-based arrays include probe molecules that are or upregulated in ovarian endothelial tumor cells (such as those include proteins, or where the target molecules are or include 25 listed in Tables 2 and 4) which reduces or eliminates ovarian proteins, and arrays including nucleic acids to which proteins cancer. Such as by reducing or inhibiting angiogenesis. In are bound, or vice versa. In some examples, an array contains Some examples, a specific binding agent significantly antibodies to ovarian endothelial cell tumor-associated pro increases the biological activity of one or more ovarian endot teins, such as any combination of those listed in Table 1. Such helial cell tumor-associated molecules that is downregulated as at least 2, least 5, at least 7, at least 10, at least 20, at least 30 in ovarian endothelial tumor cells (such as those listed in 30, at least 40, at least 50, at least 60, at least 70, at least 80, Table 3). at least 90, at least 100, at least 150, at least 200, at least 250, Cancer: at least 300, at least 400, at least 500, at least 600, at least 700, The “pathology' of cancer includes all phenomena that at least 800, at least 900, at least 1000, or at least 1100 compromise the well-being of the subject. This includes, sequences listed in Table 1 (for example, 2, 6, 8, 9, 10, 11, 12. 35 without limitation, abnormal or uncontrollable cell growth, 13, 14, 15, 16, 17, 18, 19, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, metastasis, interference with the normal functioning of 70, 75, 80, 85,90, 95, 100,105,110, 120, 130, 140, 150, 175, neighboring cells, release of cytokines or other secretory 225, 275,325, 350,375,450, 550, 650, 750, 850, 950, 1050 products at abnormal levels, Suppression or aggravation of or 1149 of those listed). inflammatory or immunological response, neoplasia, prema Binding or Stable Binding: 40 lignancy, malignancy, invasion of Surrounding or distant tis An association between two Substances or molecules, such Sues or organs, such as lymph nodes, etc. "Metastatic disease' as the hybridization of one nucleic acid molecule to another refers to cancer cells that have left the original tumor site and (or itself), the association of an antibody with a peptide, or the migrate to other parts of the body for example via the blood association of a protein with another protein or nucleic acid stream or lymph system. molecule. An oligonucleotide molecule binds or stably binds 45 Chemotherapeutic Agent or Chemotherapy: to a target nucleic acid molecule if a Sufficient amount of the Any chemical agent with therapeutic usefulness in the oligonucleotide molecule forms base pairs or is hybridized to treatment of diseases characterized by abnormal cell growth. its target nucleic acid molecule, to permit detection of that Such diseases include tumors, neoplasms, and cancer as well binding. “Preferentially binds’ indicates that one molecule as diseases characterized by hyperplastic growth Such as pso binds to another with high affinity, and binds to heterologous 50 riasis. In one embodiment, a chemotherapeutic agent is an molecules at a low affinity. agent of use in treating ovarian cancer, such as papillary Binding can be detected by any procedure known to one serous ovarian cancer. In one example, a chemotherapeutic skilled in the art, Such as by physical or functional properties agent is a radioactive compound. One of skill in the art can of a target: oligonucleotide complex or a protein:antibody readily identify a chemotherapeutic agent of use (see for complex. For example, binding can be detected functionally 55 example, Slapak and Kufe, Principles of Cancer Therapy, by determining whether binding has an observable effect Chapter 86 in Harrison's Principles of Internal Medicine, upon a biosynthetic process such as expression of a gene, 14th edition: Perry et al., Chemotherapy, Ch. 17 in Abeloff, DNA replication, transcription, translation, and the like. Clinical Oncology 2nd ed., 2000 Churchill Livingstone, Inc: Physical methods of detecting the binding of complemen Baltzer and Berkery. (eds): Oncology Pocket Guide to Che tary strands of nucleic acid molecules, include but are not 60 motherapy, 2nd ed. St. Louis, Mosby-Year Book, 1995; Fis limited to, such methods as DNase I or chemical footprinting, cher Knobf, and Durivage (eds): The Cancer Chemotherapy gel shift and affinity cleavage assays, Northern blotting, dot Handbook, 4th ed. St. Louis, Mosby-Year Book, 1993). Che blotting and light absorption detection procedures. For motherapeutic agents used for treating ovarian cancer example, one method involves observing a change in light include, but are not limited to, carboplatin, cisplatin, pacli absorption of a solution containing an oligonucleotide (or an 65 taxel, docetaxel, doxorubicin, epirubicin, topotecan, irinote analog) and a target nucleic acid at 220 to 300 nm as the can, gemcitabine, iazofurine, gemcitabine, etoposide, temperature is slowly increased. If the oligonucleotide or Vinorelbine, tamoxifen, Valspodar, cyclophosphamide, meth US 8,440,393 B2 15 16 otrexate, fluorouracil, mitoxantrone and vinorelbine. Combi nucleotides for use under the desired conditions is provided nation chemotherapy is the administration of more than one by Beltz et al. Methods Enzymol. 100:266-285, 1983, and by agent (Such as more than one chemotherapeutic agent) to treat Sambrook et al. (ed.), Molecular Cloning: A Laboratory CaCC. Manual, 2nd ed., vol. 1-3, Cold Spring Harbor Laboratory Chrondroitin Sulfate Proteoglycan 2 (CSPG2): Press, Cold Spring Harbor, N.Y., 1989. An extracellular matrix component of the vitreous gel that Contacting: has been reported to be an anti-cell adhesive. In particular Placement in direct physical association, including both a examples, expression of CSPG2 is increased in ovarian can Solid and liquid form. Contacting can occur in vitro, for cer endothelial cells. The term CSPG2 includes any CSPG2 example, with isolated cells or in Vivo by administering to a gene, cDNA, mRNA, or protein from any organism and that is 10 Subject. CSPG2 and is expressed and in some examples overex Decrease: pressed in ovarian cancer endothelial cells. To reduce the quality, amount, or strength of something. In Nucleic acid and protein sequences for CSPG2 are publicly one example, a therapy decreases a tumor (such as the size of available. For example, GenBank Accession Nos.: a tumor, the growth of a tumor, number of tumors, the NM 004385 and BCO96495 disclose CSPG2 nucleic acid 15 metastasis of a tumor, or combinations thereof), or one or sequences, and GenBank Accession Nos.: AAH50524, more symptoms associated with a tumor, for example as NP 004376, and AAH96495 disclose CSPG2 protein compared to the response in the absence of the therapy (Such sequences, all of which are incorporated by reference as pro as a therapy administered to affect tumor size by inhibiting vided by GenBank on Feb. 14, 2007. angiogenesis via administration of a binding agent capable of In one example, CSPG2 includes a full-length wild-type binding to one or more of the ovarian endothelial cell tumor (or native) sequence, as well as CSPG2 allelic variants that associated markers listed in Tables 1 through 5 that is retain the ability to be expressed in ovarian tumor endothelial involved in promoting angiogenesis, Such as by inhibiting an cells and/or modulate ovarian tumor endothelial cells, such as ovarian endothelial cell tumor-associated marker that is increase vascular growth. In certain examples, CSPG2 has at upregulated in ovarian endothelial tumor cells or by increas least 80% sequence identity, for example at least 85%, 90%, 25 ing activity of an ovarian endothelial cell tumor-associated 95%, or 98% sequence identity to GenBank Accession No. marker that is downregulated in ovarian endothelial tumor AAH50524, NP 004376, or AAH96495. In other examples, cells). In a particular example, a therapy decreases the size of CSPG2 has a sequence that hybridizes to AFFYMETRIX(R) a tumor, the growth of a tumor, the number of tumors, the Probe ID No. 204619 s at and 221731 Xa and retains metastasis of a tumor, or combinations thereof. Subsequent to CSPG2 activity (such as the capability to be expressed in 30 the therapy, such as a decrease of at least 10%, at least 20%, ovariantumor endothelial cells and/or modulate tumor and/or at least 50%, or even at least 90%. Such decreases can be vascular growth). measured using the methods disclosed herein. Complementarity and Percentage Complementarity: Determining Expression of a Gene Product: Molecules with complementary nucleic acids form a stable Detection of a level expression in either a qualitative or duplex or triplex when the strands bind, (hybridize), to each 35 quantitative manner. other by forming Watson-Crick, Hoogsteen or reverse Hoogs Diagnosis: teen base pairs. Stable binding occurs when an oligonucle The process of identifying a disease by its signs, symptoms otide molecule remains detectably bound to a target nucleic and results of various tests. The conclusion reached through acid sequence (such as an ovarian endothelial cell tumor that process is also called “a diagnosis. Forms of testing associated molecule) under the required conditions. 40 commonly performed include blood tests, medical imaging, Complementarity is the degree to which bases in one urinalysis, and biopsy. nucleic acid strand with the bases in a second DNA (Deoxyribonucleic Acid): nucleic acid strand. Complementarity is conveniently A long chain polymer which includes the genetic material described by percentage, that is, the proportion of nucleotides of most living organisms (some viruses have genes including that form base pairs between two strands or within a specific 45 ribonucleic acid, RNA). The repeating units in DNA poly region or domain of two strands. For example, if 10 nucle mers are four different nucleotides, each of which includes otides of a 15-nucleotide oligonucleotide form base pairs one of the four bases, adenine, guanine, cytosine and thymine with a targeted region of a DNA molecule, that oligonucle bound to a deoxyribose Sugar to which a phosphate group is otide is said to have 66.67% complementarity to the region of attached. Triplets of nucleotides, referred to as codons, in DNA targeted. 50 DNA molecules code for amino acid in a polypeptide. The In the present disclosure, “sufficient complementarity” term codon is also used for the corresponding (and comple means that a Sufficient number of base pairs exist between an mentary) sequences of three nucleotides in the mRNA into oligonucleotide molecule and a target nucleic acid sequence which the DNA sequence is transcribed. (such as a ovarian endothelial cell tumor-associated mol Differential Expression: ecule, for example any of the genes listed in Table 1) to 55 A difference. Such as an increase or decrease, in the con achieve detectable binding. When expressed or measured by version of the information encoded in a gene (Such as an percentage of base pairs formed, the percentage complemen ovarian endothelial cell tumor-associated molecule) into tarity that fulfills this goal can range from as little as about messenger RNA, the conversion of mRNA to a protein, or 50% complementarity to full (100%) complementary. Ingen both. In some examples, the difference is relative to a control eral, sufficient complementarity is at least about 50%, for 60 or reference value. Such as an amount of gene expression that example at least about 75% complementarity, at least about is expected in a subject who does not have ovarian cancer or 90% complementarity, at least about 95% complementarity, in a normal (non-cancerous) endothelial cell sample. Detect at least about 98% complementarity, or even at least about ing differential expression can include measuring a change in 100% complementarity. gene expression. A thorough treatment of the qualitative and quantitative 65 Downregulated or Inactivation: considerations involved in establishing binding conditions When used in reference to the expression of a nucleic acid that allow one skilled in the art to design appropriate oligo molecule. Such as a gene, refers to any process which results US 8,440,393 B2 17 18 in a decrease in production of a gene product. A gene product part of a cell. Such as the synthesis of a protein. Gene expres can be RNA (such as mRNA, rRNA, tRNA, and structural sion can be influenced by external signals. For instance, expo RNA) or protein. Therefore, gene downregulation or deacti Sure of a cell to a hormone may stimulate expression of a Vation includes processes that decrease transcription of a gene hormone-induced gene. Different types of cells can respond or translation of mRNA. Examples of genes whose expres 5 differently to an identical signal. Expression of a gene also sion is downregulated in ovarian tumor endothelial cells can can be regulated anywhere in the pathway from DNA to RNA be found in Table 1 (indicated by a negative fold change, such to protein. Regulation can include controls on transcription, as TLOC1 and HS6ST2) and Table 3 (such as PLN, SELE, translation, RNA transport and processing, degradation of GREB1, OGN and LCXD3). intermediary molecules such as mRNA, or through activa Examples of processes that decrease transcription include 10 tion, inactivation, compartmentalization or degradation of those that facilitate degradation of a transcription initiation specific protein molecules after they are produced. complex, those that decrease transcription initiation rate, The expression of a nucleic acid molecule can be altered those that decrease transcription elongation rate, those that relative to a normal (wild type) nucleic acid molecule. Alter decrease processivity of transcription and those that increase ations in gene expression, Such as differential expression, transcriptional repression. Gene downregulation can include 15 include but are not limited to: (1) overexpression; (2) under reduction of expression above an existing level. Examples of expression; or (3) Suppression of expression. Alternations in processes that decrease translation include those that the expression of a nucleic acid molecule can be associated decrease translational initiation, those that decrease transla with, and in fact cause, a change in expression of the corre tional elongation and those that decrease mRNA stability. sponding protein. Specific examples of ovarian endothelial Gene downregulation includes any detectable decrease in cell tumor-associated molecules that are up-regulated in ova the production of a gene product. In certain examples, pro rian tumor endothelial cells are provided in Tables 2 and 4. duction of a gene product decreases by at least 2-fold, for Specific examples of ovarian endothelial cell tumor-associ example at least 3-fold or at least 4-fold, as compared to a ated molecules that are down-regulated in ovarian tumor control (such an amount of gene expression in a normal endothelial cells are listed in Table 3. For example, EZH2, endothelial cell). In one example, a control is a relative 25 EGFL6, TNFAIP6, TWIST1, STC1, HOP, CSPG2, and amount of gene expression or protein expression in a biologi PLXDC1 are upregulated or increased in expression in ova cal sample taken from a Subject who does not have ovarian rian tumor endothelial cells, while TLOC1 and HS6ST2 are CaCC. downregulated or decreased in expression in Such cells. Endothelial Cell: Protein expression can also be altered in Some manner to be Cells that line the interior surface of blood vessels, forming 30 different from the expression of the protein in a normal (wild an interface between circulating blood in the lumen and the type) situation. This includes but is not necessarily limited to: rest of the vessel wall. For example, endothelial cells line the (1) a in the protein such that one or more of the entire circulatory system. Further, both blood and lymphatic amino acid residues is different; (2) a short deletion or addi capillaries are composed of a single layer of endothelial cells. tion of one or a few (such as no more than 10-20) amino acid Epidermal Growth Factor-like Domain Multiple 6 35 residues to the sequence of the protein; (3) alonger deletion or (EGFL6): addition of amino acid residues (such as at least 20 residues), A member of the epidermal growth factor (EGF) repeat Such that an entire protein domain or Sub-domain is removed Superfamily of genes known to encode proteins that govern or added; (4) expression of an increased amount of the protein cellular proliferative responses. EGFL6 has been identified as compared to a control or standard amount; (5) expression of a possible regulator of cell cycle and oncogenesis. 40 a decreased amount of the protein compared to a control or In particular examples, expression of EGFL6 is increased standard amount; (6) alteration of the subcellular localization in ovarian cancer endothelial cells. The term EGFL6 includes or targeting of the protein; (7) alteration of the temporally any EGFL6 gene, cDNA, mRNA, or protein from any organ regulated expression of the protein (such that the protein is ism and that is EGFL6 and is expressed and in some examples expressed when it normally would not be, or alternatively is overexpressed in ovarian cancer endothelial cells. 45 not expressed when it normally would be); (8) alteration in Nucleic acid and protein sequences for EGFL6 are publicly stability of a protein through increased longevity in the time available. For example, GenBank Accession Nos.: that the protein remains localized in a cell; and (9) alteration NM O15507, NM 019397 and BCO38587 disclose EGFL6 of the localized (such as organ or tissue specific or Subcellular nucleic acid sequences, and GenBank Accession Nos.: localization) expression of the protein (such that the protein is AAQ88699, CAM23572, and AAF27812 disclose EGFL6 50 not expressed where it would normally be expressed or is protein sequences, all of which are incorporated by reference expressed where it normally would not be expressed), each as provided by GenBank on Feb. 14, 2007. compared to a control or standard. In one example, EGFL6 includes a full-length wild-type Controls or standards for comparison to a sample, for the (or native) sequence, as well as EGFL6 allelic variants that determination of differential expression, include samples retain the ability to be expressed in ovarian tumor endothelial 55 believed to be normal (in that they are not altered for the cells and/or modulate ovarian tumor endothelial cells, such as desired characteristic, for example a sample from a subject increase vascular growth. In certain examples, EGFL6 has at who does not have cancer, Such as ovarian cancer) as well as least 80% sequence identity, for example at least 85%, 90%, laboratory values, even though possibly arbitrarily set, keep 95%, or 98% sequence identity to GenBank Accession No. ing in mind that Such values can vary from laboratory to AAQ88699, CAM23572, or AAF27812. In other examples, 60 laboratory. EGFL6 has a sequence that hybridizes to Affymetrix Probe Laboratory standards and values may be set based on a ID No. 219454 at and retains EGFL6 activity (such as the known or determined population value and can be Supplied in capability to be expressed in ovarian tumor endothelial cells the format of a graph or table that permits comparison of and/or modulate tumor and/or vascular growth). measured, experimentally determined values. Expression: 65 Gene Expression Profile (or Fingerprint): The process by which the coded information of a gene is Differential or altered gene expression can be detected by converted into an operational, non-operational, or structural changes in the detectable amount of gene expression (such as US 8,440,393 B2 19 20 cDNA or mRNA) or by changes in the detectable amount of (1989) Molecular Cloning, second edition, Cold Spring Har proteins expressed by those genes. A distinct or identifiable bor Laboratory, Plainview, N.Y. (chapters 9 and 11). The pattern of gene expression, for instance a pattern of high and following is an exemplary set of hybridization conditions and low expression of a defined set of genes or gene-indicative is not limiting: nucleic acids such as ESTs; in Some examples, as few as one 5 Very High Stringency (Detects Sequences that Share at or two genes provides a profile, but more genes can be used in Least 90% Identity) a profile, for example at least 3, at least 4, at least 5, at least 6, Hybridization: 5xSSC at 65° C. for 16 hours at least 10, at least 20, at least 25, at least 30, at least 50, at least Wash twice: 2xSSC at room temperature (RT) for 15 min 80, at least 100, at least 190, at least 200, at least 300, at least utes each 400, at least 500, at least 550, at least 600, at least 700, at least 10 Wash twice: 0.5xSSC at 65° C. for 20 minutes each 800, at least 900, at least 1000, at least 1100 or more of those High Stringency (Detects Sequences that Share at Least listed in any of Tables 1-5. A gene expression profile (also 80% Identity) referred to as a fingerprint) can be linked to a tissue or cell Hybridization: 5x-6xSSC at 65° C.-70° C. for 16-20 hours type (such as ovarian cancer cell), to a particular stage of Wash twice: 2XSSC at RT for 5-20 minutes each normal tissue growth or disease progression (such as 15 Wash twice: 1XSSC at 55° C.-70° C. for 30 minutes each advanced ovarian cancer), or to any other distinct or identifi Low Stringency (Detects Sequences that Share Greater able condition that influences gene expression in a predict than 50% Identity) able way. Gene expression profiles can include relative as Hybridization: 6xSSC at RT to 55° C. for 16-20 hours well as absolute expression levels of specific genes, and can Wash at least twice: 2x-3XSSC at RT to 55° C. for 20-30 be viewed in the context of a test sample compared to a minutes each. baseline or control sample profile (such as a sample from a Inhibitor: Subject who does not have ovarian cancer or normal endot Any chemical compound, nucleic acid molecule, peptide helial cells). In one example, a gene expression profile in a or polypeptide such as an antibody or RNAi that can reduce Subject is read on an array (such as a nucleic acid or protein activity of a gene product or interfere with expression of a array). For example, a gene expression profile is performed 25 gene, respectively. In some examples, an inhibitor can reduce using a commercially available array Such as a Human or inhibit the activity of a protein that is encoded by a gene Genome U133 2.0 Plus Microarray from AFFYMETRIX(R) either directly or indirectly. Direct inhibition can be accom (AFFYMETRIX(R), Santa Clara, Calif.). plished, for example, by binding to a protein and thereby Homeodomain-Only Protein, Transcript Variant 2 (HOP): preventing the protein from binding an intended target, Such A transcriptional repressor that modulates serum response 30 as a receptor. Indirect inhibition can be accomplished, for factor-dependent cardiac-specific gene expression and car example, by binding to a protein's intended target, such as a diac development. In particular examples, expression of HOP receptor or binding partner, thereby blocking or reducing is increased in ovarian cancer endothelial cells. The term activity of the protein. In some examples, an inhibitor of the HOP includes any HOP gene, cDNA, mRNA, or protein from disclosure can inhibit agene by reducing or inhibiting expres any organism and that is HOP and is expressed and in some 35 sion of the gene, interalia by interfering with gene expression examples overexpressed in ovarian cancer endothelial cells. (transcription, processing, translation, post-translational Nucleic acid and protein sequences for HOP are publicly modification), for example, by interfering with the gene's available. For example, GenBank Accession Nos.: mRNA and blocking translation of the gene product or by NM 13921 1, XM 001083738, and XM 001137349 dis post-translational modification of a gene product, or by caus close HOP nucleic acid sequences, and GenBank Accession 40 ing changes in intracellular localization. Nos.: AAH14225, NP 631958, and NP 631957 disclose Isolated: HOP protein sequences, all of which are incorporated by An "isolated biological component (such as a nucleic acid reference as provided by GenBank on Feb. 14, 2007. molecule, protein, or cell) has been Substantially separated or In one example, HOP includes a full-length wild-type (or purified away from other biological components in the cell of native) sequence, as well as HOP allelic variants that retain 45 the organism, or the organism itself, in which the component the ability to be expressed in ovarian tumor endothelial cells naturally occurs, such as other chromosomal and extra-chro and/or modulate ovarian tumor endothelial cells, such as mosomal DNA and RNA, proteins and cells. Nucleic acid increase vascular growth. In certain examples, HOP has at molecules and proteins that have been "isolated include least 80% sequence identity, for example at least 85%, 90%, nucleic acid molecules and proteins purified by standard puri 95%, or 98% sequence identity to GenBank Accession No.: 50 fication methods. The term also embraces nucleic acid mol AAH14225, NP 631958, or NP 631957. In other ecules and proteins prepared by recombinant expression in a examples, HOP has a sequence that hybridizes to AFFYME host cell as well as chemically synthesized nucleic acid mol TRIX(R) Probe ID No. 211597 s at and retains HOP activity ecules and proteins. For example, an isolated serous papillary (such as the capability to be expressed in ovarian tumor endot ovarian cancer cell is one that is Substantially separated from helial cells and/or modulate tumor and/or vascular growth). 55 other ovarian cell Subtypes, such as endometrioid, clear cell Hybridization: or mucinous Subtypes. To form base pairs between complementary regions of two Label: strands of DNA, RNA, or between DNA and RNA, thereby An agent capable of detection, for example by ELISA, forming a duplex molecule. Hybridization conditions result spectrophotometry, flow cytometry, or microscopy. For ing in particular degrees of Stringency will vary depending 60 example, a label can be attached to a nucleic acid molecule or upon the nature of the hybridization method and the compo protein, thereby permitting detection of the nucleic acid mol sition and length of the hybridizing nucleic acid sequences. ecule or protein. Examples of labels include, but are not Generally, the temperature of hybridization and the ionic limited to, radioactive isotopes, enzyme Substrates, co-fac strength (such as the Na" concentration) of the hybridization tors, ligands, chemiluminescent agents, fluorophores, hap buffer will determine the stringency of hybridization. Calcu 65 tens, enzymes, and combinations thereof. Methods for label lations regarding hybridization conditions for attaining par ing and guidance in the choice of labels appropriate for ticular degrees of stringency are discussed in Sambrook et al., various purposes are discussed for example in Sambrooketal. US 8,440,393 B2 21 22 (Molecular Cloning: A Laboratory Manual, Cold Spring Har Oligonucleotide Probe: bor, N.Y., 1989) and Ausubel et al. (In Current Protocols in A short sequence of nucleotides, such as at least 8, at least Molecular Biology, John Wiley & Sons, New York, 1998). In 10, at least 15, at least 20, at least 21, at least 25, or at least 30 a particular example, a label is conjugated to a binding agent nucleotides in length, used to detect the presence of a comple that specifically binds to one or more of the ovarian endothe mentary sequence by molecular hybridization. In particular lial cell tumor-associated molecules disclosed in Tables 1 examples, oligonucleotide probes include a label that permits through 5 to allow for the detection/screening for angiogen detection of oligonucleotide probe: target sequence hybrid esis and/or the presence of a tumor in a Subject. ization complexes. Such as with an ovarian endothelial cell Malignant: tumor-associated molecule listed in Tables 1-5. 10 Ovarian Cancer: A Cells that have the properties of anaplasia invasion and malignant ovarian neoplasm (an abnormal growth located metastasis. on the ovaries). Cancer of the ovaries includes ovarian carci Mammal: noma, papillary serous cystadenocarcinoma, mucinous cys This term includes both human and non-human mammals. tadenocarcinoma, endometrioid tumors, celioblastoma, clear Examples of mammals include, but are not limited to: humans 15 cell carcinoma, unclassified carcinoma, granulosa-thecal cell and Veterinary and laboratory animals, such as pigs, cows, tumors, Sertoli-Leydig cell tumors, dysgerminoma, and goats, cats, dogs, rabbits and mice. malignant teratoma. The most common type of ovarian can Neoplasm: cer is papillary serous carcinoma. Abnormal growth of cells. Surgery is an exemplary treatment for ovarian cancer and Normal Cell: can be necessary for diagnosis. The type of Surgery depends Non-tumor cell, non-malignant, uninfected cell. upon how widespread the cancer is when diagnosed (the Nucleic Acid Array: cancer stage), as well as the type and grade of cancer. The An arrangement of nucleic acids (such as DNA or RNA) in Surgeon may remove one (unilateral oophorectomy) or both assigned locations on a matrix. Such as that found in cDNA ovaries (bilateral oophorectomy), the fallopian tubes (salp arrays, or oligonucleotide arrays, such as those listed in 25 ingectomy), and the uterus (hysterectomy). For Some very Tables 1-5. early tumors (stage 1, low grade or low-risk disease), only the Nucleic Acid Molecules Representing Genes: involved ovary and fallopian tube will be removed (called a Any nucleic acid, for example DNA (intron or exon or “unilateral salpingo-oophorectomy,” USO), especially in both), cDNA, or RNA (such as mRNA), of any length suitable young females who wish to preserve their fertility. In 30 advanced disease as much tumor as possible is removed (de for use as a probe or other indicator molecule, and that is bulking Surgery). In cases where this type of Surgery is suc informative about the corresponding gene. cessful, the prognosis is improved compared to subjects Nucleic Acid Molecules: where large tumor masses (more than 1 cm in diameter) are A deoxyribonucleotide or ribonucleotide polymer includ left behind. ing, without limitation, cDNA, mRNA, genomic DNA, and 35 Chemotherapy is often used after Surgery to treat any synthetic (such as chemically synthesized) DNA. The nucleic residual disease. At present systemic chemotherapy often acid molecule can be double-stranded or single-stranded. includes a platinum derivative with a taxane as a method of Where single-stranded, the nucleic acid molecule can be the treating advanced ovarian cancer. Chemotherapy is also used sense Strand or the antisense Strand. In addition, nucleic acid to treat Subjects who have a recurrence. molecule can be circular or linear. 40 Ovarian Endothelial Cell Tumor-Associated (or Related) The disclosure includes isolated nucleic acid molecules Molecule: that include specified lengths of an ovarian endothelial cell A molecule whose expression is altered in ovarian tumor tumor-associated molecule nucleotide sequence, for endothelial cells. Such molecules include, for instance, sequences for genes listed in Tables 1 through 4. Such mol nucleic acid sequences (such as DNA, cDNA, or mRNAs) ecules can include at least 10, at least 15, at least 20, at least 45 and proteins. Specific genes include those listed in Tables 1 25, at least 30, at least 35, at least 40, at least 45 or at least 50 through 5. Thus, the presence of the respective ovarian endot consecutive nucleotides of these sequences or more, and can helial cell tumor-associated molecules can be used to diag be obtained from any region of a ovarian endothelial cell nose, or determine the prognosis of an ovarian tumor in a tumor-associated molecule. Subject. Oligonucleotide: 50 In an example, an ovarian endothelial cell tumor-associ A plurality of joined nucleotides joined by native phos ated molecule is any molecule listed in Tables 1 through 5. phodiester bonds, between about 6 and about 300 nucleotides Specific examples of ovarian endothelial cell tumor-associ in length. An oligonucleotide analog refers to moieties that ated molecules that are up-regulated in ovarian tumor endot function similarly to oligonucleotides but have non-naturally helial cells are provided in Tables 2 and 4. Specific examples occurring portions. For example, oligonucleotide analogs can 55 of ovarian endothelial cell tumor-associated molecules that contain non-naturally occurring portions, such as altered are down-regulated in ovarian tumor endothelial cells are Sugar moieties or inter-Sugar linkages, such as a phospho listed in Table 3. As illustrated in Table 4, a number of the rothioate oligodeoxynucleotide. identified ovarian cell tumor-associated molecules are related Particular oligonucleotides and oligonucleotide analogs to cell proliferation, tube formation and cell motility. can include linear sequences up to about 200 nucleotides in 60 Ovarian endothelial cell tumor-associated molecules can length, for example a sequence (such as DNA or RNA) that is be involved in or influenced by cancer in different ways, at least 6 nucleotides, for example at least 8, at least 10, at including causative (in that a change in a ovarian endothelial least 15, at least 20, at least 21, at least 25, at least 30, at least cell tumor-associated molecule leads to development of or 35, at least 40, at least 45, at least 50, at least 100 or even at progression of ovarian cancer) or resultive (in that develop least 200 nucleotides long, or from about 6 to about 50 nucle 65 ment of or progression of ovarian cancer causes or results in otides, for example about 10-25 nucleotides, such as 12, 15 or a change in the ovarian endothelial cell tumor-associated 20 nucleotides. molecule). US 8,440,393 B2 23 24 Pharmaceutically Acceptable Carriers: from the template, and then re-annealed, extended, and dis The pharmaceutically acceptable carriers (vehicles) useful Sociated to amplify the number of copies of the nucleic acid. in this disclosure are conventional. Remington’s Pharmaceu The product of a PCR can be characterized by methods tical Sciences, by E.W. Martin, Mack Publishing Co., Easton, known in the art Such as electrophoresis, restriction endonu Pa., 15th Edition (1975), describes compositions and formu clease cleavage patterns, oligonucleotide hybridization or lations suitable for pharmaceutical delivery of one or more ligation, and/or nucleic acid sequencing, using standard tech therapeutic agents, such as one or more compositions that niques. include a binding agent that specifically binds to at least one Primers: of the disclosed ovarian endothelial cell tumor-associated Short nucleic acid molecules, for instance DNA oligo molecules. 10 nucleotides 10-100 nucleotides in length, such as about 15, In general, the nature of the carrier will depend on the 20, 25, 30 or 50 nucleotides or more in length. Primers can be particular mode of administration being employed. For annealed to a complementary target DNA strand (such as a instance, parenteral formulations can include injectable fluids gene listed in Table 1) by nucleic acid hybridization to form a that include pharmaceutically and physiologically acceptable hybrid between the primer and the target DNA strand. Primer fluids such as water, physiological saline, balanced salt Solu 15 pairs can be used for amplification of a nucleic acid sequence, tions, aqueous dextrose, glycerol or the like as a vehicle. In such as by PCR or other nucleic acid amplification methods addition to biologically-neutral carriers, pharmaceutical known in the art. compositions to be administered can contain minor amounts Methods for preparing and using nucleic acid primers are of non-toxic auxiliary Substances, such as wetting or emulsi described, for example, in Sambrook et al. (In Molecular fying agents, preservatives, and pH buffering agents and the Cloning: A Laboratory Manual. CSHL, New York, 1989), like, for example sodium acetate or Sorbitan monolaurate, Ausubel et al. (ed.) (In Current Protocols in Molecular Biol Sodium lactate, potassium chloride, calcium chloride, and ogy, John Wiley & Sons, New York, 1998), and Innis et al. triethanolamine oleate. (PCR Protocols, A Guide to Methods and Applications, Aca Plexin Domain Containing 1 (PLXDC1): demic Press, Inc., San Diego, Calif., 1990). PCR primer pairs A large transmembrane receptor. In vitro, plexin-C1 has 25 can be derived from a known sequence, for example, by using been shown to bind the GPI-anchored semaphorin Sema7A computer programs intended for that purpose Such as Primer and the soluble viral semaphorins SemaVA (A39R) and (Version 0.5, (C) 1991, Whitehead Institute for Biomedical SemaVB (AHV). Plexin C1 engagement by SemaVA inhibits Research, Cambridge, Mass.). One of ordinary skill in the art integrin-mediated dendritic cell adhesion and chemotaxis in will appreciate that the specificity of a particular primer vitro, Suggesting a role for plexin C1 in dendritic cell migra 30 increases with its length. Thus, for example, a primer includ tion. ing 30 consecutive nucleotides of an ovarian endothelial cell In an example, expression of PLXDL1 is increased in tumor-associated molecule will anneal to a target sequence, ovarian tumor endothelial cells. The term PLXDC1 includes Such as another homolog of the designated endothelial cell any plexin C1 gene, cDNA, mRNA, or protein from any tumor-associated protein, with a higher specificity than a organism and that is a PLXDC1 and is expressed and in some 35 corresponding primer of only 15 nucleotides. Thus, in order examples overexpressed in ovarian tumor endothelial cells. to obtain greater specificity, primers can be selected that Exemplary nucleic acid and protein sequences for include at least 20, at least 25, at least 30, at least 35, at least PLXDC1 are publicly available. For example, GenBank 40, at least 45, at least 50 or more consecutive nucleotides of Accession Nos.: NM 018797, XM 622776, AB208934, an ovarian endothelial cell tumor-associated nucleotide and NM 005761 disclose PLXDC1 nucleic acid sequences 40 Sequence. and GenBank Accession Nos.: NP 061267, XP 622776, Prognosis: BAD921.71, and NP 005752 disclose PLXDC1 protein A prediction of the course of a disease, such as ovarian sequences, all of which are incorporated by reference as pro cancer. The prediction can include determining the likelihood vided by GenBank on Feb. 14, 2007. of a Subject to develop aggressive, recurrent disease, to Sur In one example, a PLXDC1 sequence includes a full-length 45 Vive a particular amount of time (e.g., determine the likeli wild-type (or native) sequence, as well as PLXDC1 allelic hood that a subject will survive 1, 2, 3 or 5 years), to respond variants, fragments, homologs or fusion sequences that retain to a particular therapy (e.g., chemotherapy), or combinations the ability to be expressed in ovarian tumor endothelial cells thereof. and/or modulate ovarian tumor endothelial cells, such as Purified: increase vascular growth. In certain examples, PLXDC1 has 50 The term “purified” does not require absolute purity: at least 80% sequence identity, for example at least 85%, rather, it is intended as a relative term. Thus, for example, a 90%. 95%, or 98% sequence identity to GenBank Accession purified protein preparation is one in which the protein No.: NP 061267, XP 622776, BAD92171, or referred to is more pure than the protein in its natural envi NP 005752. In other examples, a PLXDC1 has a sequence ronment within a cell. For example, a preparation of a protein that hybridizes to AFFYMETRIX(R) Probe ID No. 214081 at 55 is purified such that the protein represents at least 50% of the and retains PLXDC1 activity (such as the capability to be total protein content of the preparation. Similarly, a purified expressed in ovarian tumor endothelial cells and/or modulate oligonucleotide preparation is one in which the oligonucle tumor and/or vascular growth). otide is more pure than in an environment including a com Polymerase Chain Reaction (PCR): plex mixture of oligonucleotides. An in vitro amplification technique that increases the num 60 Recombinant: ber of copies of a nucleic acid molecule (for example, a A recombinant nucleic acid molecule is one that has a nucleic acid molecule in a sample or specimen). In an sequence that is not naturally occurring or has a sequence that example, a biological sample collected from a subject (e.g., is made by an artificial combination of two otherwise sepa with ovarian cancer) is contacted with a pair of oligonucle rated segments of sequence. This artificial combination can otide primers, under conditions that allow for the hybridiza 65 be accomplished for example, by chemical synthesis or by the tion of the primers to nucleic acid template in the sample. The artificial manipulation of isolated segments of nucleic acid primers are extended under Suitable conditions, dissociated molecules, such as by genetic engineering techniques. US 8,440,393 B2 25 26 Sample (or Biological Sample): programs blastp, blastin, blastX, thlastnandthlastX. Additional A biological specimen containing genomic DNA, RNA information can be found at the NCBI web site. (including mRNA), protein, or combinations thereof, BLASTN is used to compare nucleic acid sequences, while obtained from a subject. Examples include, but are not limited BLASTP is used to compare amino acid sequences. If the two to, peripheral blood, urine, saliva, tissue biopsy, Surgical 5 compared sequences share homology, then the designated specimen, amniocentesis samples and autopsy material. In output file will present those regions of homology as aligned one example, a sample includes an ovarian cancer tissue sequences. If the two compared sequences do not share biopsy. homology, then the designated output file will not present Sensitivity: aligned sequences. 10 Once aligned, the number of matches is determined by A measurement of activity, Such as biological activity, of a counting the number of positions where an identical nucle molecule or a collection of molecules in a given condition. In otide or amino acid residue is presented in both sequences. an example, sensitivity refers to the activity of an agent. Such The percent sequence identity is determined by dividing the as a binding agent that preferentially binds to one or more number of matches either by the length of the sequence set ovarian endothelial cell tumor-associated molecules, to alter 15 forth in the identified sequence, or by an articulated length the growth, development or progression of a disease, such as (such as 100 consecutive nucleotides or amino acid residues ovarian cancer. In certain examples, sensitivity or responsive from a sequence set forth in an identified sequence), followed ness can be assessed using any endpoint indicating a benefit to by multiplying the resulting value by 100. For example, a the Subject, including, without limitation, (1) inhibition, to nucleic acid sequence that has 1166 matches when aligned Some extent, of tumor growth, including slowing down and 20 with a test sequence having 1154 nucleotides is 75.0 percent complete growth arrest; (2) reduction in the number of tumor identical to the test sequence (1166-1554*100–75.0). The cells; (3) reduction in tumor size; (4) inhibition (such as percent sequence identity value is rounded to the nearest reduction, slowing down or complete stopping) of tumor cell tenth. For example, 75.11, 75.12, 75.13, and 75.14 are infiltration into adjacent peripheral organs and/or tissues; (5) rounded down to 75.1, while 75.15, 75.16, 75.17, 75.18, and inhibition (such as reduction, slowing down or complete stop- 25 75.19 are rounded up to 75.2. The length value will always be ping) of metastasis; (6) enhancement of anti-tumor immune an integer. In another example, a target sequence containing a response, which may, but does not have to, result in the 20-nucleotide region that aligns with 20 consecutive nucle regression or rejection of the tumor; (7) relief, to Some extent, otides from an identified sequence as follows contains a of one or more symptoms associated with the tumor; (8) region that shares 75 percent sequence identity to that iden increase in the length of survival following treatment; (9) 30 tified sequence (that is, 15-20*100–75). decreased mortality at a given point of time following treat For comparisons of amino acid sequences of greater than ment; and/or (10) reducing or inhibiting angiogenesis. about 30 amino acids, the Blast 2 sequences function is Sequence Identity/Similarity: employed using the default BLOSUM62 matrix set to default The identity/similarity between two or more nucleic acid parameters, (gap existence cost of 11, and a per residue gap sequences, or two or more amino acid sequences, is expressed 35 cost of 1). Homologs are typically characterized by posses in terms of the identity or similarity between the sequences. sion of at least 70% sequence identity counted over the full Sequence identity can be measured in terms of percentage length alignment with an amino acid sequence using the identity; the higher the percentage, the more identical the NCBI Basic Blast 2.0, gapped blastp with databases such as sequences are. Sequence similarity can be measured in terms the nr or Swissprot database. Queries searched with the blastin of percentage similarity (which takes into account conserva- 40 program are filtered with DUST (Hancock and Armstrong, tive amino acid substitutions); the higher the percentage, the 1994, Comput. Appl. Biosci. 10:67-70). Other programs use more similar the sequences are. Homologs or orthologs of SEG. In addition, a manual alignment can be performed. nucleic acidoramino acid sequences possess a relatively high Proteins with even greater similarity will show increasing degree of sequence identity/similarity when aligned using percentage identities when assessed by this method, such as at standard methods. This homology is more significant when 45 least about 75%, 80%, 85%, 90%, 95%, 98%, or 99% the orthologous proteins or cDNAs are derived from species sequence identity with the proteins listed in Table 1. which are more closely related (such as human and mouse When aligning short peptides (fewer than around 30 amino sequences), compared to species more distantly related (Such acids), the alignment is be performed using the Blast 2 as human and C. elegans sequences). sequences function, employing the PAM30 matrix set to Methods of alignment of sequences for comparison are 50 default parameters (open gap 9. extension gap 1 penalties). well known in the art. Various programs and alignment algo Proteins with even greater similarity to the reference rithms are described in: Smith & Waterman, Adv. Appl. Math. sequence will show increasing percentage identities when 2:482, 1981; Needleman & Wunsch, J. Mol. Biol. 48:443, assessed by this method, such as at least about 60%, 70%, 1970: Pearson & Lipman, Proc. Natl. Acad. Sci. USA 75%, 80%, 85%, 90%. 95%, 98%, 99% sequence identity 85:2444, 1988; Higgins & Sharp, Gene, 73:237-44, 1988: 55 with the proteins listed in Table 1. When less than the entire Higgins & Sharp, CABIOS 5:151-3, 1989; Corpet et al., Nuc. sequence is being compared for sequence identity, homologs Acids Res. 16:10881-90, 1988; Huang et al. Computer Appls. will typically possess at least 75% sequence identity over in the Biosciences 8, 155-65, 1992; and Pearson et al., Meth. short windows of 10-20 amino acids, and can possess Mol. Bio. 24:307-31, 1994. Altschulet al., J. Mol. Biol. 215: sequence identities of at least 85%, 90%. 95% or 98% 403-10, 1990, presents a detailed consideration of sequence 60 depending on their identity to the reference sequence. Meth alignment methods and homology calculations. ods for determining sequence identity over Such short win The NCBI Basic Local Alignment Search Tool (BLAST) dows are described at the NCBI web site. (Altschulet al., J. Mol. Biol. 215:403-10, 1990) is available One indication that two nucleic acid molecules are closely from several sources, including the National Center for Bio related is that the two molecules hybridize to each other under logical Information (NCBI, National Library of Medicine, 65 stringent conditions, as described above. Nucleic acid Building 38A, Room 8N805, Bethesda, Md. 20894) and on sequences that do not show a high degree of identity may the Internet, for use in connection with the sequence analysis nevertheless encode identical or similar (conserved) amino US 8,440,393 B2 27 28 acid sequences, due to the degeneracy of the genetic code. cDNA, mRNA, or protein from any organism and that is Changes in a nucleic acid sequence can be made using this STC1 and is expressed or overexpressed in some examples in degeneracy to produce multiple nucleic acid molecules that ovarian tumor endothelial cells. all encode Substantially the same protein. Such homologous Nucleic acid and protein sequences for STC1 are publicly nucleic acid sequences can, for example, possess at least available. For example, GenBank Accession Nos.: about 60%, 70%, 80%, 90%, 95%, 98%, or 99% sequence NM 009285, NM 00003155, and NM 031123 disclose identity with the genes listed in Table 1 as determined by this STC1 nucleic acid sequences, and GenBank Accession Nos.: method. An alternative (and not necessarily cumulative) indi AAH21425, NP 112385, and NP 033311 disclose STC1 cation that two nucleic acid sequences are Substantially iden protein sequences, all of which are incorporated by reference tical is that the polypeptide which the first nucleic acid 10 as provided by GenBank on Feb. 14, 2007. encodes is immunologically cross reactive with the polypep In one example, STC1 includes a full-length wild-type (or tide encoded by the second nucleic acid. native) sequence, as well as STC1 allelic variants that retain One of skill in the art will appreciate that the particular the ability to be expressed in ovarian tumor endothelial cells sequence identity ranges are provided for guidance only; it is and/or modulate ovarian tumor endothelial cells, such as possible that strongly significant homologs could be obtained 15 increase vascular growth. In certain examples, STC1 has at that fall outside the ranges provided. least 80% sequence identity, for example at least 85%, 90%, Short interfering RNA (siRNA): 95%, or 98% sequence identity to GenBank Accession No.: A double stranded nucleic acid molecule capable of RNA AAH21425, NP 112385, or NP 033311. In other interference or “RNAi.” (See, for example, Bass Nature 411: examples, STC1 has a sequence that hybridizes to AFFYME 428-429, 2001; Elbashir et al., Nature 411: 494-498, 2001; TRIX(R) Probe ID No. 230746 s at, 204595 s at, and and Kreutzer et al., International PCT Publication No. WO 204597 X at and retains STC1 activity (such as the capability 00/44895; Zernicka-Goetz et al., International PCT Publica to be expressed in ovarian tumor endothelial cells and/or tion No. WO 01/36646; Fire, International PCT Publication modulate tumor and/or vascular growth). No. WO99/32619; Plaetincket al., International PCT Publi Subject: cation No.WO 00/01846; Mello and Fire, International PCT 25 Living multi-cellular vertebrate organisms, a category that Publication No. WO 01/29058: Deschamps-Depaillette, includes human and non-human mammals. International PCT Publication No. WO99/07409; and Li et Target Sequence: al., International PCT Publication No. WO 00/44914.) As A sequence of nucleotides located in a particular region in used herein, siRNA molecules need not be limited to those the that corresponds to a desired sequence, molecules containing only RNA, but further encompasses 30 Such as an ovarian endothelial cell tumor-associated chemically modified nucleotides and non-nucleotides having sequence. The target can be for instance a coding sequence; it RNAi capacity or activity. In an example, an siRNA molecule can also be the non-coding strand that corresponds to a coding is one that reduces or inhibits the biological activity or expres sequence. Examples of target sequences include those sion of one or more ovarian endothelial cell tumor-associated sequences associated with ovarian tumor endothelial cells, molecules disclosed in Tables 1, 2, 4 or 5 that are upregulated 35 such as any of those listed in Tables 1 through 5. in ovariantumor endothelial cells, such as EGFL6, TNFAIP6, Therapeutically Effective Amount: TWIST1, STC1, HOP, CSPG2, PLXDC1, EZH2, the Notch An amount of a composition that alone, or together with an ligand Jagged 1 or PTK2. additional therapeutic agent(s) (for example a chemothera Specific Binding Agent: peutic agent), induces the desired response (e.g., treatment of An agent that binds Substantially or preferentially only to a 40 a tumor). The preparations disclosed herein are administered defined target (for example, those listed in Table 1), such as a in therapeutically effective amounts. protein, enzyme, polysaccharide, oligonucleotide, DNA, In one example, a desired response is to decrease ovarian RNA, recombinant vector or a small molecule. In an example, tumor size or metastasis in a Subject to whom the therapy is a 'specific binding agent' is capable of binding to at least one administered. Tumor metastasis does not need to be com of the disclosed ovarian endothelial cell tumor-associated 45 pletely eliminated for the composition to be effective. For molecules. Thus, a RNA-specific binding agent binds Sub example, a composition can decrease metastasis by a desired stantially only to the defined RNA, or to a specific region amount, for example by at least 20%, at least 50%, at least within the RNA. For example, a “specific binding agent” 60%, at least 70%, at least 80%, at least 90%, at least 95%, at includes a siRNA that binds substantially to a specified RNA. least 98%, or even at least 100% (elimination of the tumor), as A protein-specific binding agent binds Substantially only 50 compared to metastasis in the absence of the composition. the defined protein, or to a specific region within the protein. In particular examples, it is an amount of the therapeutic For example, a “specific binding agent includes antibodies agent conjugated to the specific binding agent effective to and other agents that bind Substantially to a specified decrease a number of ovarian cancer cells, such as in a subject polypeptide. The antibodies can be monoclonal or polyclonal to whom it is administered, for example a subject having one antibodies that are specific for the polypeptide, as well as 55 or more ovarian carcinomas. The cancer cells do not need to immunologically effective portions (“fragments”) thereof. be completely eliminated for the composition to be effective. The determination that a particular agent binds Substantially For example, a composition can decrease the number of can only to a specific polypeptide may readily be made by using cer cells or growth of such cells by a desired amount, for or adapting routine procedures. One Suitable in vitro assay example by at least 20%, at least 50%, at least 60%, at least makes use of the Western blotting procedure (described in 60 70%, at least 80%, at least 90%, at least 95%, at least 98%, or many standard texts, including Harlow and Lane. Using Anti even at least 100% (elimination of detectable cancer cells), as bodies: A Laboratory Manual, CSHL, New York, 1999). compared to the number of cancer cells in the absence of the Stanniocalcin 1 (STC1): composition. A hormone that plays a role in calcium regulation, phos In other examples, it is an amount of the specific binding phate homeostasis and cell metabolism. In particular 65 agent for one or more of the disclosed ovarian endothelial cell examples, expression of STC1 is increased in ovarian tumor tumor-associated molecules capable of reducing angiogen endothelial cells. The term STC1 includes any STC1 gene, esis by least 20%, at least 50%, at least 60%, at least 70%, at US 8,440,393 B2 29 30 least 80%, at least 90%, at least 95%, at least 98%, or even at somal instability. For example, TWIST1 is capable of inhib least 100% (elimination of detectable angiogenesis) by the iting chrondrogenesis. TWIST1 protein has also been noted specific binding agent, or both, effective to decrease the to be involved in the regulation of tumor necrosis factor alpha metastasis of a tumor. production by antiinflammatory factors and pathways. In par A therapeutically effective amount of a specific binding ticular examples, expression of TWIST1 is increased in ova agent for at least one of the disclosed ovarian endothelial cell rian cancer endothelial cells. The term TWIST1 includes any tumor-associated molecules, or cancer cells lysed by a thera TWIST1 gene, cDNA, mRNA, or protein from any organism peutic molecule conjugated to the agent, can be administered and that is TWIST1 and is expressed or overexpressed in in a single dose, or in several doses, for example daily, during Some examples in ovarian cancer endothelial cells. a course of treatment. However, the therapeutically effective 10 Nucleic acid and protein sequences for TWIST1 are pub amount can depend on the Subject being treated, the severity licly available. For example, GenBank Accession Nos.: and type of the condition being treated, and the manner of NM 000474, NM 053530 and XM 001076553 and dis administration. For example, a therapeutically effective close TWIST1 nucleic acid sequences, and GenBank Acces amount of such agent can vary from about 1 ug-10 mg per 70 sion Nos.: NP 000465 and ABM87769 disclose TWIST1 kg body weight if administered intravenously and about 10 15 protein sequences, all of which are incorporated by reference ug-100 mg per 70kg body weight if administered intratumor as provided by GenBank on Feb. 14, 2007. ally. In one example, TWIST1 includes a full-length wild-type Tissue: (or native) sequence, as well as TWIST1 allelic variants that A plurality of functionally related cells. A tissue can be a retain the ability to be expressed in ovarian tumor endothelial Suspension, a semi-solid, or Solid. Tissue includes cells col cells and/or modulate ovarian tumor endothelial cells, such as lected from a Subject such as the ovaries. increase vascular growth. In certain examples, TWIST1 has at Treating a Disease: least 80% sequence identity, for example at least 85%, 90%, “Treatment” refers to a therapeutic intervention that ame 95%, or 98% sequence identity to GenBank Accession No.: liorates a sign or symptom of a disease or pathological con NP 000465 or ABM87769. In other examples, TWIST1 has dition, such as a sign or symptom of ovarian cancer. Treat 25 a sequence that hybridizes to AFFYMETRIX(R) Probe ID No. ment can also induce remission or cure of a condition, such as 206026 s at and retains TWIST1 activity (such as the capa ovarian cancer. In particular examples, treatment includes bility to be expressed in ovarian tumor endothelial cells and/ preventing a disease, for example by inhibiting the full devel or modulate tumor and/or vascular growth). opment of a disease. Prevention of a disease does not require Under Conditions Sufficient for: a total absence of disease. For example, a decrease of at least 30 A phrase that is used to describe any environment that 50% can be sufficient. permits the desired activity. In one example, includes admin Tumor: istering a test agent to an ovarian cancer cell or a subject All neoplastic cell growth and proliferation, whether sufficient to allow the desired activity. In particular examples, malignant or benign, and all pre-cancerous and cancerous the desired activity is altering the activity (such as the expres cells and tissues. In an example, a tumor is an ovarian tumor. 35 sion) of an ovarian endothelial cell tumor-associated mol Tumor-Necrosis Factor, Alpha-Induced Protein 6 (TN ecule. FAIP6): Unit Dose: A protein capable of regulating the expression of various A physically discrete unit containing a predetermined molecules involved in the control of inflammation. In particu quantity of an active material calculated to individually or lar examples, expression of TNFAIP6 is increased in ovarian 40 collectively produce a desired effect, Such as a therapeutic cancer endothelial cells. The term TNFAIP6 includes any effect. A single unit dose or a plurality of unit doses can be TNFAIP6 gene, cDNA, mRNA, or protein from any organism used to provide the desired effect, such as treatment of a and that is TNFAIP6 and is expressed or overexpressed in tumor, for example a metastatic tumor. In one example, a unit Some examples in ovarian cancer endothelial cells. dose includes a desired amount of an agent that decreases or Nucleic acid and protein sequences for TNFAIP6 are pub 45 inhibits angiogenesis. licly available. For example, GenBank Accession Nos.: Upregulated or Activation: NM 007115, BC021155 and NM 009398 disclose When used in reference to the expression of a nucleic acid TNFAIP6 nucleic acid sequences, and GenBank Accession molecule. Such as a gene, refers to any process which results Nos.: AAH21155, NP 009046 and NP 033424 disclose in an increase in production of a gene product. A gene product TNFAIP6 protein sequences, all of which are incorporated by 50 can be RNA (such as mRNA, rRNA, tRNA, and structural reference as provided by GenBank on Feb. 14, 2007. RNA) or protein. Therefore, gene upregulation or activation In one example, TNFAIP6 includes a full-length wild-type includes processes that increase transcription of a gene or (or native) sequence, as well as TNFAIP6 allelic variants that translation of mRNA. Specific examples of ovarian endothe retain the ability to be expressed in ovarian tumor endothelial lial cell tumor-associated molecules that are up-regulated in cells and/or modulate ovarian tumor endothelial cells, such as 55 ovarian tumor endothelial cells are provided in Tables 2 and 4. Suppression of vascular growth. In certain examples, For example, EZH2, EGFL6, TNFAIP6, TWIST1, STC1, TNFAIP6 has at least 80% sequence identity, for example, at HOP, CSPG2, and PLXDC1 are upregulated or increased in least 85%, 90%, 95%, or 98% sequence identity to GenBank expression in ovarian tumor endothelial cells. Accession No.: AAH21155, NP 009046 or NP 033424. In Examples of processes that increase transcription include other examples, TNFAIP6 has a sequence that hybridizes to 60 those that facilitate formation of a transcription initiation AFFYMETRIX(R) Probe ID No. 206026 s at and retains complex, those that increase transcription initiation rate, TNFAIP6 activity (such as the capability to be expressed in those that increase transcription elongation rate, those that ovariantumor endothelial cells and/or modulate tumor and/or increase processivity of transcription and those that relieve vascular growth). transcriptional repression (for example by blocking the bind Twist Homologue 1 (TWIST1): 65 ing of a transcriptional repressor). Gene upregulation can Overexpression of TWIST1 has been reported to play a include inhibition of repression as well as stimulation of role in destabilizing the genome, thus promoting chromo expression above an existing level. Examples of processes US 8,440,393 B2 31 32 that increase translation include those that increase transla In one example, EZH2 includes a full-length wild-type (or tional initiation, those that increase translational elongation native) sequence, as well as EZH2 allelic variants, fragments, and those that increase mRNA stability. homologs or fusion sequences that retain the ability to be Gene upregulation includes any detectable increase in the expressed in ovarian tumor endothelial cells and/or modulate production of agene product. In certain examples, production 5 ovarian tumor endothelial cells, such as increase vascular of a gene product increases by at least 2-fold, for example at growth. In certain examples, EZH2 has at least 80% sequence least 3-fold or at least 4-fold, as compared to a control (Such identity, for example at least 85%, 90%, 95%, or 98% an amount of gene expression in a normal endothelial cell). In sequence identity to AASO9975. In other examples, EZH2 one example, a control is a relative amount of gene expression has a sequence that hybridizes to AFFYMETRIX(R) Probe ID in a biological sample, Such as in an ovarian tissue biopsy 10 No. 203358 s at and retains EZH2 activity (such as the capa bility to be expressed in ovarian tumor endothelial cells and/ obtained from a subject that does not have ovarian cancer. or modulate tumor and/or vascular growth). Vasohibin 1 (VASH1): Additional terms commonly used in molecular genetics a protein that is expressed in a variety of tissues and inhibits can be found in Benjamin Lewin, Genes V published by functions relevant to neovascularization (migration, prolif 15 Oxford University Press, 1994 (ISBN 0-19-854287-9); Ken eration, and network formation by endothelial cells). Vaso drew et al. (eds.), The Encyclopedia of Molecular Biology, hibin also inhibits angiogenesis in vivo. The unglycosylated published by Blackwell Science Ltd., 1994 (ISBN 0-632 protein (42 kDa) does not contain a classical secretory secre 02182-9); and Robert A. Meyers (ed.), Molecular Biology tion sequence and appears in the medium as a protein of 30 and Biotechnology: a Comprehensive Desk Reference, pub kDa, Suggesting proteolytic processing during secretion. In lished by VCH Publishers, Inc., 1995 (ISBN 1-56081-569-8). particular examples, VASH1 is regulated by EZH2 wherein EZH2 binds to the VASH1 promoter and decreases or inhibits Methods of Treatment VASH1 anti-angiogenesis activity. The term VASH1 includes any VASH1 gene, cDNA, mRNA, or protein from any organ It is shown herein that ovarian cancer is associated with ism and that is VASH1 and is expressed or overexpressed in 25 differential expression of ovarian endothelial cell tumor-as Some examples in ovarian cancer endothelial cells. Sociated molecules. For example, the disclosed gene expres Nucleic acid and protein sequences for VASH1 are pub sion profile has identified ovarian endothelial cell tumor licly available. For example, GenBank Accession Nos.: associated molecules. Based on these observations, methods NP 055724 (human); NP 796328 (mouse); and of treatment to reduce or eliminate ovarian cancer are dis NP 659 128 disclose VASH1 amino acid sequences which 30 closed. For example, the method can include inhibiting the are incorporated by reference as provided by GenBank on expression or biological activity of at least one of the ovarian Aug. 14, 2009. Further, GenBank Accession Nos.: endothelial cell tumor associated molecules from Tables 1, 2, NM 014909 (human) and NM 177354 (mouse) disclose 4, and/or 5 that are upregulated in ovarian tumor cells or nucleic acid sequences which are incorporated by reference increasing the expression or biological activity of at least one as provided by GenBank on Aug. 14, 2009. 35 of the ovarian endothelial cell tumor associated molecules In one example, vasohibin includes a full-length wild-type from Tables 1, 3 and/or 5 that are downregulated in ovarian (or native) sequence, as well as VASH1 allelic variants that tumor cells or combinations thereof. As used herein, “inhibit” retain the ability to be expressed in ovarian tumor endothelial does not require 100% inhibition of expression or activity. For cells and/or modulate ovarian tumor endothelial cells, such as example, a Substantial reduction may be adequate. Such as increase vascular growth. In certain examples, VASH1 has at 40 reduction in expression or activity of at least 20%, at least least 80% sequence identity, for example at least 85%, 90%, 30%, at least 50%, at least 75%, or at least 95% may be 95%, or 98% sequence identity to GenBank Accession No.: sufficient to obtain desired therapeutic results. In some NP 055724; NP 796328 or NP 659 128 and retains examples, an “increase' in expression or activity is an VASH1 activity (such as the capability to be expressed in increase of at least 20%, at least 30%, at least 50%, at least ovariantumor endothelial cells and/or modulate tumor and/or 45 75%, or at least 95%. In some embodiments, the subject is a vascular growth). human, but the subject can alternatively be a veterinary or Zeste Homologue 2 (EZH2): laboratory Subject. In some embodiments, the ovarian cancer A member of the polycomb group of genes that has been is papillary serous ovarian cancer. reported to be involved in cell cycle regulation EZH2, a Methods are disclosed hereinfortreating an ovarian tumor, component of the polycomb repressive complex 2 (PRC2), 50 Such as ovarian cancer. In one example, the method includes has intrinsic histone methyl transferase (HMTase) activity administering a therapeutically effective amount of a compo and has been implicated in the progression and metastasis of sition to a subject. The composition can include a binding several cancers. EZH2 is also a transcriptional repressor that agent that is specific for one of the ovarian endothelial cell has multiple targets, including anti-angiogenic, pro-apop tumor-associated molecules listed in any of Tables 1, 2, 4 or 5 totic, and tumor Suppressor genes. In particular examples, 55 that are upregulated in ovarian tumor cells. Administration of expression of EZH2 is increased in ovarian cancer endothelial Such compounds decreases the expression or activity of the cells. In one example, expression of EZH2 is an indicator of molecule that is undesirably upregulated in ovarian cancer poor prognosis. The term EZH2 includes any EZH2 gene, cells. The molecules in Tables 1, 2, 4, or 5 include, for cDNA, mRNA, or protein from any organism and that is instance, nucleic acid sequences (such as DNA, cDNA, or EZH2 and is expressed or overexpressed in some examples in 60 mRNAs) and proteins. Specific genes include those listed in ovarian cancer endothelial cells. Tables 1, 2, 4 or 5 as well as fragments of the full-length Nucleic acid and protein sequences for EZH2 are publicly genes, cDNAs, or mRNAs (and proteins encoded thereby) available. For example, GenBank Accession Nos.: whose expression is upregulated in response to an ovarian NM 004456 and AY519465.1 disclose EZH2 nucleic acid tumor, Such as ovarian cancer. sequences, and GenBank Accession No. AASO9975 dis 65 In particular examples, the specific binding agent is an closes a EZH2 protein sequence, which are incorporated by inhibitor such as a siRNA or an antibody to one of the dis reference as provided by GenBank on Feb. 14, 2007. closed ovarian endothelial cell tumor-associated molecules US 8,440,393 B2 33 34 that is upregulated in ovarian tumor cells. For example, the such as, by interfering with the genes mRNA and blocking specific binding agent can be a siRNA that interferes with translation of the gene product or by post-translational modi mRNA expression of one of the disclosed ovarian endothelial fication of a gene product, or by causing changes in intracel cell tumor-associated molecules that are involved in angio lular localization. In another specific example, a specific genesis, Such as a molecule involved in regulating cell motill binding agent binds to a protein encoded by of one of the ity, cell proliferation or tube formation, thereby inhibiting cell genes listed in Table 1, 2, 4 or 5 that is upregulated in ovarian motility, cell proliferation or tube formation. For example, the tumor endothelial cells with a binding affinity in the range of specific binding agent can be a siRNA that inhibits the expres 0.1 to 20 nM and reduces or inhibits the activity of such sion of PTK2, EZH2 or Jagged 1. In other particular protein. examples, ovarian tumor growth is reduced or inhibited by 10 Examples of specific binding agents include, but are not administering a specific binding agent to inhibit or reduce the limited to, siRNAS, antibodies, ligands, recombinant pro expression or production of EZH2, EGFL6, TNFAIP6, teins, peptide mimetics, and soluble receptor fragments. One TWIST1, STC1, HOP, CSPG2, and PLXDC1. In additional example of a specific binding agent is a siRNA. Methods of examples, a composition includes at least two specific bind making siRNA that can be used clinically are known in the art. ing agents such as two specific siRNAS that each bind to their 15 Particular siRNAs and methods that can be used to produce respective ovarian endothelial cell tumor-associated nucle and administer them are described in detail below. In some otide sequences and inhibitovarian tumor growth in a subject. examples, the siRNA is incorporated into a chitosan (CH) In some examples, the composition includes at least 2, 3, 4, 5, nanoparticle. Such as chitosan obtained from shellfish or 5, 8 or 10 different siRNA molecules. For example, the com fungi. position can include PTK2, EZH2 and Jagged 1 siRNAs. Another specific example of a specific binding agent is an Treating Ovarian Cancer by Altering Activity of an Ovarian antibody, Such as a monoclonal or polyclonal antibody. Meth Endothelial Cell Tumor-Associated Molecule ods of making antibodies that can be used clinically are Methods are provided to inhibit ovarian endothelial cell known in the art. Particular antibodies and methods that can tumor-associated molecule activity or expression to treat an be used to produce them are described in detail below. ovarian tumor. Treatment of tumors by reducing the number 25 In a further example, small molecular weight inhibitors or of ovarian endothelial cell tumor-associated molecules can antagonists of the receptor protein can be used to regulate include delaying the development of the tumor in a subject activity Such as the expression or production of ovarian endot (such as preventing metastasis of a tumor). Treatment of a helial cell tumor-associated molecules. In a particular tumor also includes reducing signs or symptoms associated example, Small molecular weight inhibitors or antagonists of with the presence of Such a tumor (for example by reducing 30 the proteins encoded by the genes listed in Tables 2 and/or 4 the size, growth or Volume of the tumor or a metastasis are employed. thereof). Such reduced growth can in some examples Specific binding agents can be therapeutic, for example by decrease or slow metastasis of the tumor, or reduce the size or reducing or inhibiting the biological activity of a nucleic acid volume of the tumor by at least 10%, at least 20%, at least or protein. Complete inhibition is not required. For example, 50%, or at least 75%, such as by inhibiting angiogenesis by at 35 a reduction by at least 10%, at least 20%, at least 30%, at least least 10%, at least 20%, at least 50%, or at least 75%. For 50%, at least 70%, or even at least 90% can be sufficient. For example, ovarian endothelial cell tumor-associated mol example, a specific binding agent that binds with high affinity ecules involved in angiogenesis, Such as molecules involved to a gene listed in Tables 1, 2, 4 and/or 5 that are upregulated in promoting cell proliferation, cell motility or tube formation in ovarian tumor endothelial cells, may substantially reduce can be inhibited to treat an ovarian tumor, Such as those 40 the biological function of the gene or gene product (for provided in any of Tables 1, 2, 4 or 5 that are upregulated in example, the ability of the gene or gene product to facilitate ovarian endothelial tumor cells. In other examples, ovarian angiogenesis). In other examples, a specific binding agent tumor growth is reduced or inhibited by inhibiting the expres that binds with high affinity to one of the proteins encoded by sion or biological activity ovarian endothelial cell tumor the genes listed in Tables 1, 2, 4 and/or 5 that are upregulated associated molecules provided in any of Tables 1, 2, 4 or 5that 45 in ovarian tumor endothelial cells, may substantially reduce are upregulated in ovarian tumor endothelial cells. In further the biological function of the protein (for example, the ability examples, inhibition of ovarian endothelial cell tumor-asso of the protein to promote angiogenesis). Such agents can be ciated molecules includes reducing the invasive activity of the administered in therapeutically effective amounts to Subjects tumor in the Subject. In some examples, treatment using the in need thereof. Such as a subject having ovarian cancer. Such methods disclosed herein prolongs the time of survival of the 50 as papillary serous ovarian cancer. Subject. Pre-Screening Subjects Specific Binding Agents In some examples, Subjects are initially screened to deter Specific binding agents are agents that selectively bind mine if they have ovarian cancer. In an example, Subjects are with higher affinity to a molecule of interest, than to other initially screened for ovarian cancer by using one of the molecules. For example, a specific binding agent can be one 55 disclosed gene expression profiles (as discussed in detail that binds with high affinity to one of the genes or gene below). In some examples, if one or more of the disclosed products of the ovarian endothelial cell tumor-associated endothelial cell tumor-associated molecules upregulated in molecules listed in any of Tables 1, 2, 4 or 5 that are upregu ovarian endothelial cells (such as those listed in Tables 2 and lated in ovarian tumor endothelial cells, but does not substan 4) is detected, a specific binding agent capable of reducing or tially bind to another gene or gene product. In a specific 60 inhibiting ovarian cancer is adminstered. example, a specific binding agent binds to one gene listed in Pre-Screening Specific Binding Agents Tables 1, 2, 4 or 5 that is upregulated in ovarian tumor endot In some examples, specific binding agents are initially helial cells thereby reducing or inhibiting expression of the screened for treating ovarian cancer by use of the disclosed gene, but does not bind to the other genes (or gene product) gene expression profile (see below). For example, the dis listed in such Tables under similar conditions. For example, 65 closed gene expression profile can be used to identify specific the agent can interfere with gene expression (transcription, binding agents capable of reducing or inhibiting ovarian can processing, translation, post-translational modification), cer. In an example, the disclosed gene expression profile is US 8,440,393 B2 35 36 used to identify compositions that can be employed to reduce specific for one of the disclosed ovarian endothelial cell or inhibit angiogenesis in ovarian tumors. tumor-associated molecules further includes one or more bio Exemplary Tumors logically inactive compounds. Examples of such biologically A tumor is an abnormal growth of tissue that results from inactive compounds include, but are not limited to: carriers, excessive cell division. A particular example of a tumor is 5 thickeners, diluents, buffers, preservatives, and carriers. The cancer. For example, the current application provides meth pharmaceutically acceptable carriers useful for these formu ods for the treatment (such as the prevention or reduction of lations are conventional (see Remington's Pharmaceutical metastasis) of tumors (such as cancers) by altering the expres Sciences, by E. W. Martin, Mack Publishing Co., Easton, Pa., sion/production of one or more disclosed ovarian endothelial 19th Edition (1995)). For solid compositions (for example, cell tumor-associated molecules. In some examples, the 10 powder, pill, tablet, or capsule forms), conventional non-toxic tumor is treated in Vivo, for example in a mammalian Subject, Solid carriers can include, for example, pharmaceutical Such as a human Subject. Exemplary tumors that can be grades of mannitol, lactose, starch, or magnesium Stearate. In treated using the disclosed methods include, but are not lim addition to biologically-neutral carriers, pharmaceutical ited to ovarian cancer, including metastases of such tumors to compositions to be administered can include minor amounts other organs. Generally, the tumor is an ovarian cancer. Such 15 of non-toxic auxiliary Substances, such as wetting or emulsi as papillary serous ovarian cancer. fying agents, preservatives, and pH buffering agents and the Administration like, for example sodium acetate or Sorbitan monolaurate. Methods of administrating the disclosed compositions are Additional Treatments routine, and can be determined by a skilled clinician. For In particular examples, prior to, during, or following example, the disclosed therapies (such as those that include a administration of a therapeutic amount of an agent that binding agent specific for one of the disclosed ovarian endot reduces or inhibits ovarian cancer due to the interaction of a helial cell tumor-associated molecules listed in Tables 1, 2, 4 binding agent with one of the disclosed ovarian endothelial or 5 whose expression is increased in ovarian endothelial cell tumor-associated molecules, the Subject can receive one tumor-associated cells) can be administered via injection, or more other therapies. In one example, the Subject receives intratumorally, orally, topically, transdermally, parenterally, 25 one or more treatments to remove or reduce the tumor prior to or via inhalation or spray. In a particular example, a compo administration of a therapeutic amount of a composition sition is administered intravenously to a mammalian Subject, including a binding agent specific for one of the disclosed Such as a human. ovarian endothelial cell tumor-associated molecules. The therapeutically effective amount of the agents admin Examples of such therapies include, but are not limited to, istered can vary depending upon the desired effects and the 30 Surgical treatment for removal or reduction of the tumor (Such subject to be treated. In one example, the method includes as Surgical resection, cryotherapy, or chemoembolization), as daily administration of at least 1 lug of the composition to the well as anti-tumor pharmaceutical treatments which can Subject (such as a human Subject). For example, a human can include radiotherapeutic agents, anti-neoplastic chemothera be administered at least 1 Lug or at least 1 mg of the compo peutic agents, antibiotics, alkylating agents and antioxidants, sition daily, such as 10 ug to 100 ug daily, 100 ug to 1000 g 35 kinase inhibitors, and other agents. Particular examples of daily, for example 10 ug daily, 100 ug daily, or 1000 ug daily. additional therapeutic agents that can be used include micro In one example, the Subject is administered at least 1 Lug (Such tubule binding agents, DNA intercalators or cross-linkers, as 1-100 ug) intravenously of the composition including a DNA synthesis inhibitors, DNA and/or RNA transcription binding agent that specifically binds to one of the disclosed inhibitors, antibodies, enzymes, enzyme inhibitors, and gene ovarian endothelial cell tumor-associated molecules provided 40 regulators. These agents (which are administered at a thera herein. In one example, the Subject is administered at least 1 peutically effective amount) and treatments can be used alone mg intramuscularly (for example in an extremity) of Such or in combination. Methods and therapeutic dosages of Such composition. In a specific example, the dose is 50 to 350 agents are known to those skilled in the art, and can be ug/kg twice weekly, Such as 150 ug/kg twice weekly (for determined by a skilled clinician. example via iv injection). The dosage can be administered in 45 “Microtubule binding agent” refers to an agent that inter divided doses (such as 2, 3, or 4 divided doses per day), or in acts with tubulin to stabilize or destabilize microtubule for a single dosage daily. mation thereby inhibiting cell division. Examples of micro In particular examples, the Subject is administered the tubule binding agents that can be used in conjunction with the therapeutic composition that includes a binding agent specific disclosed therapy include, without limitation, paclitaxel, doc for one of the disclosed ovarian endothelial cell tumor-asso 50 etaxel, vinblastine, vindesine, vinorelbine (navelbine), the ciated molecules on a multiple daily dosing schedule, such as epothilones, colchicine, dolastatin 15, nocodazole, podo at least two consecutive days, 10 consecutive days, and so phyllotoxin and rhizoxin. Analogs and derivatives of Such forth, for example for a period of weeks, months, or years. In compounds also can be used and are known to those of ordi one example, the Subject is administered the therapeutic com nary skill in the art. For example, suitable epothilones and position that a binding agent specific for one of the disclosed 55 epothilone analogs are described in International Publication ovarian endothelial cell tumor-associated molecules daily for No. WO 2004/018478. Taxoids, such as paclitaxel and doc a period of at least 30 days. Such as at least 2 months, at least etaxel, as well as the analogs of paclitaxel taught by U.S. Pat. 4 months, at least 6 months, at least 12 months, at least 24 Nos. 6,610,860: 5,530,020; and 5,912,264 can be used. months, or at least 36 months. The following classes of compounds are of use in the The therapeutic compositions, such as those that include a 60 methods disclosed herein: Suitable DNA and/or RNA tran binding agent specific for one of the ovarian endothelial cell Scription regulators, including, without limitation, actinomy tumor-associated molecules, can further include one or more cin D, daunorubicin, doxorubicin and derivatives and analogs biologically active or inactive compounds (or both). Such as thereof also are suitable for use in combination with the anti-neoplastic agents and conventional non-toxic pharma disclosed therapies. DNA intercalators and cross-linking ceutically acceptable carriers, respectively. 65 agents that can be administered to a subject include, without In a particular example, a therapeutic composition that limitation, cisplatin, carboplatin, oxaliplatin, mitomycins, includes atherapeutically effective amount of a binding agent Such as mitomycin C, bleomycin, chlorambucil, cyclophos US 8,440,393 B2 37 38 phamide and derivatives and analogs thereof. DNA synthesis closed therapies (such as one including a binding agent spe inhibitors suitable for use as therapeutic agents include, with cific for one of the disclosed ovarian endothelial cell tumor out limitation, methotrexate, 5-fluoro-5'-deoxyuridine, associated molecules). In an example, one or more 5-fluorouracil and analogs thereof. Examples of suitable chemotherapeutic agents is administered following treatment enzyme inhibitors include, without limitation, camptothecin, with a binding agent specific for one of the disclosed ovarian etoposide, formestane, trichostatin and derivatives and ana endothelial cell tumor-associated molecules. In another par logs thereof. Suitable compounds that affect gene regulation ticular example, the Subject has a metastatic tumor and is include agents that result in increased or decreased expres administered radiation therapy, chemoembolization therapy, sion of one or more genes, such as raloxifene, 5-azacytidine, or both concurrently with the administration of the disclosed 5-aza-2'-deoxycytidine, tamoxifen, 4-hydroxytamoxifen, 10 therapies (such as one including a binding agent specific for mifepristone and derivatives and analogs thereof. Kinase one of the disclosed ovarian endothelial cell tumor-associated inhibitors include Gleevac, Iressa, and Tarceva that prevent molecules). phosphorylation and activation of growth factors. Other therapeutic agents, for example anti-tumor agents, Generation and Administration of siRNA that may or may not fall under one or more of the classifica 15 tions above, also are suitable for administration in combina In an example, certain inhibitors provided by this disclo tion with the disclosed therapies. By way of example, such sure are species of siRNAs. One of ordinary skill in the art can agents include adriamycin, apigenin, rapamycin, Zebularine, readily generate siRNAs which specifically bind to one of the cimetidine, and derivatives and analogs thereof. disclosed ovarian endothelial cell tumor-associated mol In one example, the therapeutic composition (such as one ecules that are upregulated in ovarian endothelial cell tumor including a binding agent specific for one or more of the cells. In an example, commercially available kits, such as disclosed ovarian endothelial cell tumor-associated mol siRNA molecule synthesizing kits from PROMEGAR) ecules) is injected into the Subject in the presence of an (Madison, Wis.) or AMBIONR) (Austin,Tex.) may be used to adjuvant. An adjuvant is an agent that when used in combi synthesize siRNA molecules. In another example, siRNAs nation with an immunogenic agent augments or otherwise 25 are obtained from commercial sources, such as from alters or modifies a resultant immune response. In some QIAGENR Inc (Germantown, Md.), INVITROGENR) examples, an adjuvant increases the titer of antibodies (Carlsbad, Calif.), AMBION (Austin, Tex.), DHARMA induced in a Subject by the immunogenic agent. In one CONR) (Lafayette, Colo.) or OPENBIOSYSTEMS(R) (Hunts example, the one or more peptides are administered to the ville, Ala.). Subject as an emulsion with an adjuvant and sterile water for 30 In certain examples, expression vectors are employed to injection (for example an intravenous or intramuscular injec express the at least one siRNA molecule. For example, an tion). Incomplete Freund's Adjuvant (Seppic, Inc.) can be expression vector can include a nucleic acid sequence encod used as the Freund's Incomplete Adjuvant (IFA) (Fairfield, ing at least one siRNA molecule corresponding to at least one N.J.). In some examples, IFA is provided in 3 ml of a mineral of the disclosed ovarian endothelial cell tumor-associated oil solution based on mannide oleate (Montanide ISA-51). At 35 molecules listed in Tables 1, 2, 4 and/or Table 5 that are the time of injection, the peptide(s) is mixed with the Mon upregulated in ovarian endothelial cell tumor cells. For tanide ISA.51 and then administered to the subject. Other example, siRNA specific for EZH2 can be generated using adjuvants can be used, for example, Freund's complete adju publicly available EZH2 nucleic acid sequences, such as vant, B30-MDP. LA-15-PH, montanide, saponin, aluminum those described above. In a particular example, the vector hydroxide, alum, lipids, keyhole lympet protein, hemocya 40 contains a sequence(s) encoding both Strands of a siRNA nin, a mycobacterial antigen, and combinations thereof. molecule comprising a duplex. In another example, the vector In some examples, the Subject receiving the therapeutic also contains sequence(s) encoding a single nucleic acid mol peptide composition (Such as one including a binding agent ecule that is self-complementary and thus forms a siRNA specific for one of the disclosed ovarian endothelial cell molecule. Non-limiting examples of such expression vectors tumor-associated molecules) is also administered interleu 45 are described in Paul et al., Nature Biotechnology 19:505, kin-2 (IL-2), for example via intravenous administration. In 2002; Miyagishi and Taira, Nature Biotechnology 19:497, particular examples, IL-2 (Chiron Corp., Emeryville, Calif.) 2002: Lee et al., Nature Biotechnology 19:500, 2002; and is administered at a dose of at least 500,000 IU/kg as an Novina et al., Nature Medicine, online publication Jun. 3, intravenous bolus over a 15 minute period every eight hours 2003. beginning on the day after administration of the peptides and 50 In other examples, siRNA molecules include a delivery continuing for up to 5 days. Doses can be skipped depending vehicle, including interalia liposomes, for administration to a on Subject tolerance. Subject, carriers and diluents and their salts, and can be In some examples, the disclosed compositions can be co present in pharmaceutical compositions. Nucleic acid mol administered with a fully human antibody to cytotoxic T-lym ecules can be administered to cells by a variety of methods phocyte antigen-4 (anti-CTLA-4). In some example Subjects 55 known to those of skill in the art, including, but not restricted receive at least 1 mg/kg anti-CTLA-4 (Such as 3 mg/kg every to, encapsulation in liposomes, by iontophoresis, or by incor 3 weeks or 3 mg/kg as the initial dose with Subsequent doses poration into other delivery vehicles, such as hydrogels, reduced to 1 mg/kg every 3 weeks). cyclodextrins, biodegradable nanocapsules, and bioadhesive In one example, at least a portion of the ovariantumor (Such microspheres, or by proteinaceous vectors (see, for example, as a metastatic tumor) is Surgically removed (for example via 60 O'Hare and Normand, International PCT Publication No. cryotherapy), irradiated, chemically treated (for example via WO 00/53722). In one specific example, siRNAs are admin chemoembolization) or combinations thereof, prior to admin istered at according to the teachings of Soutschek et al. (Na istration of the disclosed therapies (such as administration of ture Vol. 432: 173-178, 2004) or Karpillow et al. (Pharma a binding agent specific for one of the disclosed ovarian Genomics 32-40, 2004) both of which are hereinincorporated endothelial cell tumor-associated molecules). For example, a 65 by reference in their entireties. Subject having a metastatic tumor can have all or part of the In some examples, siRNAS are incorporated into neutral tumor Surgically excised prior to administration of the dis liposomes. Such as DOPC or chitosan, and injected intraperi US 8,440,393 B2 39 40 toneal or intravenously. For example, a siRNA can be admin helial cell tumor-associated molecules. These antibodies can istered at least 1 ug/kg twice weekly, such as at least 50 ug/kg be monoclonal or polyclonal. They can be chimeric or twice weekly, at least 100 g/kg twice weekly, at least 125 humanized. Any functional fragment or derivative of an anti ug/kg twice weekly, at least 150 ug/kg twice weekly, at least body can be used including Fab, Fab', Fab2, Fab'2, and single 200 ug/kg twice weekly for at least 1 week, Such as at least 2 chain variable regions. So long as the fragment or derivative weeks, at least 3 weeks, at least 4 weeks, at least 6 weeks, at retains specificity of binding for the ovarian endothelial cell least 12 weeks, or at least 24 weeks. In one example, about at tumor-associated molecule it can be used in the methods least 1-500 g/kg, Such 10-250 g/kg, is adminstered at least provided herein. Antibodies can be tested for specificity of twice weekly for at least 1 week, Such as at least 2 weeks, at binding by comparing binding to appropriate antigen to bind least 3 weeks, at least 4 weeks, at least 6 weeks, at least 12 weeks, or at least 24 weeks. In a certain example, approxi 10 ing to irrelevantantigen or antigen mixture under a given set mately 150 ug/kg is administered twice weekly, for 2 to 3 of conditions. If the antibody binds to appropriate antigen at weeks. In other examples, approximately 1 ug/kg daily for 3 least 2, at least 5, at least 7 or 10 times more than to irrelevant weeks or 50 ug/kg every other day for 3 weeks is adminis antigen orantigen mixture, then it is considered to be specific. tered. In an example, monoclonal antibodies are generated to the Alternatively, the nucleic acid/vehicle combination can be 15 ovarian endothelial cell tumor-associated molecules dis locally delivered by direct injection or by use of an infusion closed in Tables 1, 2, 4 or 5 that are upregulated in ovarian pump. Direct injection of the nucleic acid molecules of the endothelial cell tumor cells. These monoclonal antibodies disclosure, whether subcutaneous, intramuscular, or intrad each include a variable heavy (V) and a variable light (V) ermal, can take place using standard needle and Syringe meth chain and specifically bind to the specific ovarian endothelial odologies, or by needle-free technologies Such as those cell tumor-associated molecules. For example, the antibody described by Barry et al., International PCT Publication No. can bind the specific ovarian endothelial cell tumor-associ WO99/31262. Other delivery routes, but are not limited to, ated molecules with an affinity constant of at least 10 M', oral delivery (such as in tablet or pill form), intrathecal or such as at least 107M", at least 10 M', at least 5x10 M', intraperitoneal delivery. For example, intraperitoneal deliv or at least 10 M'. ery can take place by injecting the treatment into the perito 25 The specific antibodies can include a V polypeptide hav neal cavity of the subject in order to directly deliver the ing amino acid sequences of the complementarity determin molecules to the tumor site. More detailed descriptions of ing regions (CDRs) that are at least about 90% identical, such nucleic acid delivery and administration are provided in Sul as at least about 95%, at least about 98%, or at least about 99% livan et al., PCT WO 94/02595, Draper et al., PCT WO93/ identical to the amino acid sequences of the specific ovarian 23569, Beigelman et al., PCT WO99/05094, and Klimuket 30 endothelial cell tumor-associated molecules and a V al., PCT WO99/04819, all of which are incorporated by ref polypeptide having amino acid sequences of the CDRS that erence herein. are at least about 90% identical, such as at least about 95%, at Alternatively, certain siRNA molecules can be expressed least about 98%, or at least about 99% identical to the amino within cells from eukaryotic promoters. Those skilled in the acid sequences of the specific ovarian endothelial cell tumor art will recognize that any nucleic acid can be expressed in 35 associated molecules. eukaryotic cells using the appropriate DNA/RNA vector. The In one example, the sequence of the specificity determining activity of Such nucleic acids can be augmented by their regions of each CDR is determined Residues that are outside release from the primary transcript by an enzymatic nucleic the SDR (non-ligand contacting sites) are substituted. For acid (Draper et al., PCT WO 93/23569, and Sullivan et al., example, in any of the CDR sequences, at most one, two or PCT WO94/02595). 40 three amino acids can be substituted. The production of chi In other examples, siRNA molecules can be expressed mericantibodies, which include a framework region from one from transcription units (see for example, Couture et al., antibody and the CDRs from a different antibody, is well 1996, TIG 12:510) inserted into DNA or RNA vectors. The known in the art. For example, humanized antibodies can be recombinant vectors can be DNA plasmids or viral vectors. routinely produced. The antibody or antibody fragment can siRNA expressing viral vectors can be constructed based on, 45 be a humanized immunoglobulin having CDRS from a donor for example, but not limited to, adeno-associated virus, ret monoclonal antibody that binds one of the disclosed ovarian rovirus, adenovirus, lentivirus or alphavirus. In another endothelial cell tumor-associated molecules and immunoglo example, pol III based constructs are used to express nucleic bulin and heavy and light chain variable region frameworks acid molecules (see for example, Thompson, U.S. Pat. Nos. from human acceptor immunoglobulin heavy and light chain 5,902,880 and 6,146,886). 50 frameworks. Generally, the humanized immunoglobulin spe The recombinant vectors capable of expressing the siRNA cifically binds to one of the disclosed ovarian endothelial cell molecules can be delivered as described above, and persist in tumor-associated molecules with an affinity constant of at target cells. Alternatively, viral vectors can be used that pro least 107M", such as at least 10M at least 5x10'M' or at vide for transient expression of nucleic acid molecules. Such least 10 M. vectors can be repeatedly administered as necessary. Once 55 In another example, human monoclonal antibodies to the expressed, the siRNA molecule interacts with the target disclosed ovarian endothelial cell tumor-associated mol mRNA and generates an RNAi response. Delivery of siRNA ecules in Tables 1, 2, 4 and 5 that are upregulated in ovarian molecule expressing vectors can be systemic, such as by endothelial tumor cells are produced. Human monoclonal intravenous or intramuscular administration, by administra antibodies can be produced by transferring donor comple tion to target cells ex-planted from a subject followed by 60 mentarity determining regions (CDRS) from heavy and light reintroduction into the Subject, or by any other means that variable chains of the donor mouse immunoglobulin into a would allow for introduction into the desired target cell. human variable domain, and then Substituting human resi dues in the framework regions when required to retain affin Generation of Antibodies ity. The use of antibody components derived from humanized 65 monoclonal antibodies obviates potential problems associ One of ordinary skill in the art can readily generate anti ated with the immunogenicity of the constant regions of the bodies which specifically bind to the disclosed ovarian endot donor antibody. For example, when mouse monoclonal anti US 8,440,393 B2 41 42 bodies are used therapeutically, the development of human developed to crosslink the chains through glutaraldehyde, anti-mouse antibodies (HAMA) leads to clearance of the intermolecular disulfides, or a peptide linker. Thus, in one murine monoclonal antibodies and other possible adverse example, the Fv can be a disulfide stabilized Fv (dsEv), events. Chimeric monoclonal antibodies, with human con wherein the heavy chain variable region and the light chain stant regions, humanized monoclonal antibodies, retaining variable region are chemically linked by disulfide bonds. only murine CDRs, and “fully human monoclonal antibod In an additional example, the Fv fragments include V and ies made from phage libraries or transgenic mice have all V, chains connected by a peptide linker. These single-chain been used to reduce or eliminate the murine content of thera antigen binding proteins (ScPV) are prepared by constructing peutic monoclonal antibodies. a structural gene comprising DNA sequences encoding the Techniques for producing humanized monoclonal antibod 10 V and V, domains connected by an oligonucleotide. The ies are described, for example, by Jones et al., Nature 321: structural gene is inserted into an expression vector, which is 522, 1986; Riechmann et al., Nature 332:323, 1988; Verho Subsequently introduced into a host cell Such as E. coli. The eyen et al., Science 239:1534, 1988; Carter et al., Proc. Natl. recombinant host cells synthesize a single polypeptide chain Acad. Sci. U.S.A. 89:4285, 1992; Sandhu, Crit. Rev. Biotech. with a linker peptide bridging the two V domains. Methods 12:437, 1992; and Singer et al., J. Immunol. 150:2844, 1993. 15 for producing scFVs are known in the art (see Whitlow et al., The antibody may be of any isotype, but in several embodi Methods: a Companion to Methods in Enzymology, Vol. 2, ments the antibody is an IgG, including but not limited to, page 97, 1991; Bird et al., Science 242:423, 1988: U.S. Pat. IgG, IgG, IgGs and IgG4. No. 4,946,778; Packet al., Bio/Technology 11:1271, 1993: In one example, the sequence of the humanized immuno and Sandhu, Supra). globulin heavy chain variable region framework can be at Antibody fragments can be prepared by proteolytic least about 65% identical to the sequence of the donor immu hydrolysis of the antibody or by expression in E. coli of DNA noglobulin heavy chain variable region framework. Thus, the encoding the fragment. Antibody fragments can be obtained sequence of the humanized immunoglobulin heavy chain by pepsin or papain digestion of whole antibodies by conven variable region framework can be at least about 75%, at least tional methods. For example, antibody fragments can be pro about 85%, at least about 99% or at least about 95%, identical 25 duced by enzymatic cleavage of antibodies with pepsin to to the sequence of the donor immunoglobulin heavy chain provide a 5S fragment denoted F(ab'). This fragment can be variable region framework. Human framework regions, and further cleaved using a thiol reducing agent, and optionally a that can be made in a humanized antibody frame blocking group for the Sulfhydryl groups resulting from work regions, are known in the art (see, for example, in U.S. cleavage of disulfide linkages, to produce 3.5S Fab' monova Pat. No. 5,585,089). 30 lent fragments. Alternatively, an enzymatic cleavage using Antibodies, such as murine monoclonal antibodies, chi pepsin produces two monovalent Fab' fragments and an Fc meric antibodies, and humanized antibodies, include full fragment directly (see U.S. Pat. No. 4,036,945 and U.S. Pat. length molecules as well as fragments thereof. Such as Fab, No. 4331,647, and references contained therein; Nisonhoff F(ab'), and Fv, which include a heavy chain and light chain et al., Arch. Biochem. Biophy's. 89:230, 1960; Porter, Bio variable region and are capable of binding the epitopic deter 35 chem. J. 73:119, 1959; Edelman et al., Methods in Enzymol minant. These antibody fragments retain some ability to ogy, Vol. 1, page 422. Academic Press, 1967; and Coligan et selectively bind with their epitope. These fragments include: al. at sections 2.8.1-2.8.10 and 2.10.1-2.10.4). (1) Fab, the fragment which contains a monovalent antigen Other methods of cleaving antibodies, such as separation binding fragment of an antibody molecule, can be produced of heavy chains to form monovalent light-heavy chain frag by digestion of whole antibody with the enzyme papain to 40 ments, further cleavage of fragments, or other enzymatic, yield an intact light chainanda portion of one heavy chain; (2) chemical, or genetic techniques may also be used, so long as Fab', the fragment of an antibody molecule can be obtained by the fragments bind to the antigen that is recognized by the treating whole antibody with pepsin, followed by reduction, intact antibody. to yield an intact light chain and a portion of the heavy chain; One of skill will realize that conservative variants of the two Fab' fragments are obtained per antibody molecule; (3) 45 antibodies can be produced. Such conservative variants (Fab'), the fragment of the antibody that can be obtained by employed in antibody fragments, such as dsEV fragments or treating whole antibody with the enzyme pepsin without sub in ScFV fragments, will retain critical amino acid residues sequent reduction; F(ab') is a dimer of two Fab' fragments necessary for correct folding and stabilizing between the V held together by two disulfide bonds; (4) Fv, a genetically and the V, regions, and will retain the charge characteristics engineered fragment containing the variable region of the 50 of the residues in order to preserve the low pland low toxicity light chain and the variable region of the heavy chain of the molecules Amino acid Substitutions (such as at most expressed as two chains; and (5) Single chain antibody (Such one, at most two, at most three, at most four, or at most five as ScFV), defined as a genetically engineered molecule con amino acid Substitutions) can be made in the V and the V, taining the variable region of the light chain, the variable regions to increase yield. Conservative amino acid substitu region of the heavy chain, linked by a suitable polypeptide 55 tion tables providing functionally similar amino acids are linker as a genetically fused single chain molecule. Methods well known to one of ordinary skill in the art. The following of making these fragments are known in the art (see for six groups are examples of amino acids that are considered to example, Harlow and Lane, Antibodies. A Laboratory be conservative substitutions for one another: 1)Alanine (A), Manual, Cold Spring Harbor Laboratory, NewYork, 1988. Fv Serine (S), Threonine (T); 2) Aspartic acid (D), Glutamic acid antibodies are typically about 25kDa and contain a complete 60 (E); 3) Asparagine (N), Glutamine (Q); 4) Arginine (R), antigen-binding site with three CDRs per each heavy chain Lysine (K); 5) Isoleucine (I), Leucine (L), Methionine (M), and each light chain. To produce these antibodies, the V and Valine (V); and 6) Phenylalanine (F), Tyrosine (Y), Tryp the V, can be expressed from two individual nucleic acid tophan (W). constructs in a host cell. If the V and the V are expressed In Some examples, naked antibodies can be administered at non-contiguously, the chains of the Fv antibody are typically 65 least 5 mg per kg every two weeks, such as at least 10 mg per held together by noncovalent interactions. However, these kg, at least 25 mg per kg, at least 30 mg per kg, at least 50 mg chains tend to dissociate upon dilution, so methods have been per kg (for example, 6, 7, 8, 9, 10, 15, 20, 25, 30, 35, 40, 45, US 8,440,393 B2 43 44 50 mg per kg), at least once a week, at least once every 2 or decrease in production of a gene product/expression, Such weeks, at least once every 3 weeks, or at least once every as RNA or protein. For example, an alteration can include month depending upon the ovarian cancer. In an example, the processes that downregulate or decrease transcription of a antibodies are administered continuously. In another gene or translation of mRNA whose expression or activity is example, antibodies or antibody fragments conjugated to increased in ovarian endothelial tumor cells. Gene downregu cytotoxic agents (immunotoxins) are administered at least 10 lation includes any detectable decrease in the production of a ug per kg, Such as at least 20, at least 30, at least 50, at least 70, gene product. In certain examples, production/expression of a at least 100 ug per kg, at least twice a week, at least once a gene product decreases by at least 2-fold, for example at least week, at least once every two weeks, at least once every 3-fold, at least 4-fold, at least 6-fold, or at least 10-fold as month depending upon the ovarian cancer. In one example, 50 10 compared to a control. Exemplary ovarian endothelial cell ug per kg is administered twice a week for 2 to 3 weeks. In tumor-associated molecules that are up-regulated in ovarian other examples, the Subject is administered the therapeutic tumor endothelial cells are presented in Tables 1, 2, 4 and 5. composition that a binding agent specific for one or more of Thus, a decrease in the expression of one or more of the the disclosed ovarian endothelial cell tumor-associated mol molecules listed in Tables 1, 2, 4 or 5 that is noted as being ecules daily for a period of at least 30 days, such as at least 2 15 up-regulated in ovarian tumor endothelial cells following months, at least 4 months, at least 6 months, at least 12 treatment indicates that the agent is of use for treating the months, at least 24 months, or at least 36 months. Subjects can ovarian cancer. be monitored by methods known to those skilled in the art to Exemplary ovarian endothelial cell tumor-associated mol determine ovariantumorresponsiveness to the antibody treat ecules that are down-regulated in ovarian tumor endothelial ment. The subject can be monitored by non invasive tech cells are presented in Table 1 with specific examples provided niques such as CT or MRI imaging to assess tumor response. in Table 3. Thus, an increase in the expression of one or more It is contemplated that additional agents can be administered, of the molecules listed in Table 3 following administration Such as antineoplastic agents in combination with or follow indicates that the agent is effective for the treatment of ovarian ing treatment with the antibodies. CaCC. 25 In another example, an alteration can include processes Methods of Evaluating the Effectiveness of an that increase transcription of a gene or translation of mRNA. Ovarian Tumor Treatment Gene up-regulation includes any detectable increase in the production of a gene product. In certain examples, produc Methods are disclosed herein for determining the effective tion/expression of a gene product increases by at least 2-fold, ness of a binding agent specific for one of the disclosed 30 for example at least 3-fold or at least 4-fold, at least 6-fold, at ovarian endothelial cell tumor-associated molecules) for the least 10-fold or at least 28-fold following treatment as com treatment of an ovarian tumor in a subject with the ovarian pared to a control. tumor. In an example, the method includes detecting expres Detection of Ovarian Endothelial Cell Tumor-Associated sion of an ovarian endothelial cell tumor-associated molecule Nucleic Acids in a sample from the Subject following administration of the 35 Nucleic acids can be detected by any method known in the binding agent (Such as an siRNA), for example at least 24 art. In some examples, nucleic acids are isolated, amplified, or hours, at least 1 week, at least 2 weeks, or at least 4 weeks both, prior to detection. In an example, the biological sample following administration of the agent. The expression of the can be incubated with primers that permit the amplification of ovarian endothelial cell tumor-associated molecule following one or more of the disclosed ovarian endothelial cell tumor administration can be compared to a control. Such as a refer 40 associated mRNAs, under conditions sufficient to permit ence value. An alteration in the expression of the ovarian amplification of Such products. For example, the biological endothelial cell tumor-associated molecule relative to the sample is incubated with probes that can bind to one or more control following administration indicates that the agent is of the disclosed ovarian endothelial cell tumor-associated effective for the treatment of the ovarian cancer in the subject. nucleic acid sequences (such as cDNA, genomic DNA, or In a specific example, the method includes detecting and 45 RNA (such as mRNA)) under high stringency conditions. The comparing the protein expression levels of the ovarian endot resulting hybridization can then be detected using methods helial cell tumor-associated molecules. In other examples, the known in the art. In one example, the effectiveness of an method includes detecting and comparing the mRNA expres ovarian tumor treatment is identified by applying isolated sion levels of the ovarian endothelial cell tumor-associated nucleic acid molecules to an array in which the isolated molecules. In certain examples, the treatment is considered 50 nucleic acid molecules are obtained from a biological sample effective if the expression levels are altered by at least 2-fold, including ovarian endothelial cancer cells following treat such as by at least 3-fold, at least 4-fold, at least 6-fold or at ment with the ovarian tumor treatment. In Such example, the least 10-fold relative to the control. array includes oligonucleotides complementary to all ovarian In one example, the specific ovarian endothelial cell tumor endothelial cell tumor-associated genes listed in Table 1. In a associated molecule is detected in a biological sample. In a 55 particular example, the array is a commercially available particular example, the biological sample is a tumor biopsy. array such as a U133 Plus 2.0 oligonucleotide array from In another example, the ovarian endothelial cell tumor-asso AFFYMETRIXR (AFFYMETRIX(R), Santa Clara, Calif.). ciated molecule is detected in a serum sample. For example, In an example, the isolated nucleic acid molecules are the ovarian endothelial cell tumor-associated molecule is incubated with the array including oligonucleotides comple detected in a serum sample if the specific molecule is known 60 mentary to the ovarian endothelial cell tumor-associated mol to be secreted or located on a cell surface susceptible to ecules that are up-regulated in ovarian tumor endothelial enzymatic cleavage. cells, such as those listed in Table 2, Table 3, Table 4 and/or Altering Ovarian Endothelial Cell Tumor-Associated Mol Table 5 for a time sufficient to allow hybridization between ecules Activity Such as Expression the isolated nucleic acid molecules and oligonucleotide In an example, an alteration in the expression of one or 65 probes, thereby forming isolated nucleic acid molecule:oli more of the disclosed ovarian endothelial cell tumor-associ gonucleotide complexes. The isolated nucleic acid molecule: ated molecules following administration includes an increase oligonucleotide complexes are then analyzed to determine if US 8,440,393 B2 45 46 expression of the isolated nucleic acid molecules is altered. In analyzed by mass spectrometry for the presence or absence of Such example, an ovarian tumor treatment is effective if a the specific ovarian endothelial cell tumor-associated mol decrease in the expression of ovarian endothelial tumor-asso ecule. ciated molecules is observed as compared to a control (Such In one example, the antibody that specifically binds an as a normal endothelial cell) or reference value. In an addi ovarian endothelial cell tumor-associated molecule (such as tional example, the array includes oligonucleotides comple those listed in Table 1) is directly labeled with a detectable mentary to the ovarian endothelial cell tumor-associated mol label. In another example, each antibody that specifically ecules that down-regulated in ovarian tumor endothelial cells, binds an ovarian endothelial cell tumor-associated molecule such as those listed in Table 3. In this example, an ovarian (the first antibody) is unlabeled and a second antibody or tumor treatment is effective if an increase in the expression of 10 other molecule that can bind the human antibody that specifi one or more ovarian endothelial tumor-associated molecules cally binds the respective ovarian endothelial cell tumor associated molecule is labeled. As is well known to one of is observed. skill in the art, a second antibody is chosen that is able to Gene Expression Profile specifically bind the specific species and class of the first A gene expression profile is disclosed herein that can be 15 antibody. For example, if the first antibody is a human IgG, used to identify the effectiveness of an ovarian tumor treat then the secondary antibody can be an anti-human-IgG. Other ment. In an example, the gene expression profile includes at molecules that can bind to antibodies include, without limi least two of the ovarian endothelial cell tumor-associated tation, Protein A and Protein G, both of which are available molecules listed in Table 1, such as at least 5, at least 7, at least commercially. 10, at least 20, at least 30, at least 40, at least 50, at least 60, Suitable labels for the antibody or secondary antibody at least 70, at least 80, at least 90, at least 100, at least 150, at include various enzymes, prosthetic groups, fluorescent least 200, at least 250, at least 300, at least 400, at least 500, materials, luminescent materials, magnetic agents and radio at least 600, at least 700, at least 800, at least 900, at least active materials. Non-limiting examples of Suitable enzymes 1000, or at least 1100 molecules (for example, 2, 6, 8, 9, 10, include horseradish peroxidase, alkaline phosphatase, beta 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 25, 30, 35, 40, 45, 50,55, 25 galactosidase, or acetylcholinesterase. Non-limiting 60, 65, 70, 75, 80, 85,90, 95, 100,105, 110, 120, 130, 140, examples of Suitable prosthetic group complexes include 150, 175, 225, 275,325, 350, 375,450, 550, 650, 750, 850, streptavidin/biotin and avidin/biotin. Non-limiting examples 950, 1050 or 1149 of those listed). of suitable fluorescent materials include umbelliferone, fluo In a particular example, the gene expression profile rescein, fluorescein isothiocyanate, rhodamine, dichlorotri includes at least 2, at least 5, at least 7, at least 10, at least 20, 30 azinylamine fluorescein, dansyl chloride or phycoerythrin A at least 25, at least 27 molecules (for example, 2, 3, 4, 5, 6, 7, non-limiting exemplary luminescent material is luminol; a 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 22, 24, 25, 26, non-limiting exemplary magnetic agent is gadolinium, and 27, 28 or 29 molecules) listed in Table 2, Table 4 and/or Table non-limiting exemplary radioactive labels include 'I. 'I, 5 that are associated with an at least six-fold increase in SS or H. expression in tumor endothelial cells. In a particular example, 35 In an alternative example, ovarian endothelial cell tumor the at least two molecules include EGFL6 and TNFAIP6. In associated molecules can be assayed in a biological sample other particular examples, the at least two ovarian endothelial by a competition immunoassay utilizing ovarian endothelial cell tumor-associated molecules include EGFL6, TNFAIP6, cell tumor-associated molecule standards labeled with a TWIST1, STC1, HOP, CSPG2, and PLXDC1. detectable substance and unlabeled antibody that specifically In other particular examples, the gene expression profile 40 bind to the desired ovarian endothelial cell tumor-associated includes at least 2, at least 5, at least 7, at least 10, at least 13, molecule. In this assay, the biological sample (such as serum, or at least 15 molecules (for example, 2, 3, 4, 5, 6, 7, 8, 9, 10. tissue biopsy, or cells isolated from a tissue biopsy), the 11, 12, 13, 14, 15, 16 or 17 molecules) that are down-regu labeled ovarian endothelial cell tumor-associated molecule lated in ovarian tumor endothelial cells as listed in Table 3. standards and the antibody that specifically binds to ovarian For example, the profile includes the seventeen ovarian endot 45 endothelial cell tumor-associated molecule are combined and helial cell tumor-associated molecules listed in Table 3. the amount of labeled ovarian endothelial cell tumor-associ Detecting Ovarian Endothelial Cell Tumor-Associated Pro ated molecule standard bound to the unlabeled antibody is teins determined. The amount of ovarian endothelial cell tumor As an alternative to analyzing the sample for the presence associated molecule in the biological sample is inversely of nucleic acids, the presence of proteins can be determined. 50 proportional to the amount of labeled ovarian endothelial cell Proteins can be detected by any method known in the art. In tumor-associated molecule standard bound to the antibody Some examples, proteins are purified prior to detection. For that specifically binds the ovarian endothelial cell tumor example, the effect of an ovarian tumor treatment can be associated molecule. determined by incubating the biological sample with one or more antibodies that specifically binds to one of the disclosed 55 Identifying Agents to Treat Ovarian Cancer ovarian endothelial cell tumor-associated proteins encoded by the genes listed in Tables 1, Table 2, Table 3, Table 4 or Methods are provided herein for identifying agents to treat Table 5 to detect expression. The primary antibody can an ovarian cancer. For example, agents that decrease expres include a detectable label. For example, the primary antibody sion or activity of a gene that is upregulated in ovarian endot can be directly labeled, or the sample can be subsequently 60 helial tumor cells (such as those listed in Tables 2 and 4), as incubated with a secondary antibody that is labeled (for well as agents that increase activity of a gene that is down example with a fluorescent label). The label can then be regulated in ovarian endothelial tumor cells (such as those detected, for example by microscopy, ELISA, flow cytom listed in Table 3), can be identified using these methods. In an etry, or spectrophotometry. In another example, the biological example, the method includes contacting an ovarian tumor sample is analyzed by Western blotting for the presence or 65 endothelial cell with one or more test agents under conditions absence of the specific ovarian endothelial cell tumor-associ sufficient for the one or more test agents to alter the activity of ated molecule. In other examples, the biological sample is at least one ovarian endothelial cell tumor-associated mol US 8,440,393 B2 47 48 ecule listed in any of Tables 1-5. It is contemplated that listed in Table 3. For example, the nucleic acid can be part of several doses of the agent can be tested and then expression a vector Suitable for gene therapy. levels of nucleic acids or proteins can be determined. The Altering Ovarian Endothelial Cell Tumor-Associated Mol method also includes detecting the activity or expression of ecules Activity the at least one ovarian endothelial cell tumor-associated 5 In an example, an alteration in the activity of one or more molecule in the presence and absence of the one or more test of the disclosed ovarian endothelial cell tumor-associated agents. The activity or expression of the at least one ovarian molecules includes an increase or decrease in production of a endothelial cell tumor-associated molecule in the presence of gene product, Such as RNA or protein. For example, an alter the one or more test agents is then compared to the activity or ation can include processes that downregulate or decrease expression of the at least one ovarian endothelial cell tumor 10 transcription of a gene or translation of mRNA. Gene down associated molecule in the absence of Such agents to deter regulation includes any detectable decrease in the production mine if there is differential expression of the at least one of agene product. In certain examples, production/expression ovarian endothelial cell tumor associated molecule. In several of a gene product decreases by at least 2-fold, for example at examples, differential expression of the ovarian endothelial 15 least 3-fold, at least 4-fold, at least 6-fold, or at least 10-fold cell tumor-associated molecule in the presence of the agent as compared to a control (Such as a reference value or a (as compared to expression in the absence of the agent) indi normal endothelial cell). For example, a decrease in one or cates that the one or more test agents is of use to treat the more of the disclosed ovarian endothelial cell tumor-associ ovarian tumor. ated molecules up-regulated in ovarian tumor endothelial In an example, determining whether there is differential cells (such as those listed in Tables 2, 4 and 5), is indicative of expression of one or more ovarian endothelial cell tumor an agent that is effective at treating ovarian cancer. associated molecules includes generating a gene expression In another example, an alteration can include processes profile for the Subject. For example, a gene expression profile that increase transcription of a gene or translation of mRNA. for the Subject can be generated by using an array of mol Gene up-regulation includes any detectable increase in the ecules including an ovarian endothelial cell tumor-associated 25 production of a gene product. In certain examples, produc expression profile. tion/expression of a gene product increases by at least 2-fold, Ovarian Endothelial Cell Tumor-Associated Molecules for example at least 3-fold or at least 4-fold, at least 6-fold, at Ovarian endothelial cell tumor-associated molecules can least 10-fold or at least 28-fold as compared to a control. For include nucleic acid sequences (such as DNA, cDNA, or example, an increase in one or more of the disclosed ovarian mRNAs) and proteins. In a specific example, detecting dif 30 endothelial cell tumor-associated molecules down-regulated ferential expression of the ovarian endothelial cell tumor in ovarian tumor endothelial cells (such as those listed in associated molecules includes detecting differential mRNA Table 3) is indicative of an agent that is effective at treating expression of the disclosed ovarian endothelial cell tumor ovarian cancer. associated molecules. For example, such differential expres Detection of Ovarian Endothelial Cell Tumor-Associated sion can be measured by real time quantitative polymerase 35 chain reaction or microarray analysis or other methods known Nucleic Acids in the art. In another example, detecting differential expres Nucleic acids can be detected by any method known in the sion of the ovarian endothelial cell tumor-associated mol art, such as those described above. In one example, a thera ecules includes detecting differential protein expression of peutic agent is identified by applying isolated nucleic acid the disclosed ovarian endothelial cell tumor-associated mol 40 molecules to an array in which the isolated nucleic acid mol ecules. For example, protein differential expression is mea ecules are obtained from a biological sample including ova sured by Western blot analysis or a protein microarray. rian endothelial cancer cells following treatment with the one Test Agents or more test agents. In Such example, the array includes The one or more test agents can be any substance, includ oligonucleotides complementary to all ovarian endothelial ing, but not limited to, a protein (such as an antibody), a 45 cell tumor-associated genes listed in Table 1. In a particular nucleic acid molecule (such as a siRNA), an organic com example, the array is a commercially available array Such as pound, an inorganic compound, a small molecule or any other a U133 Plus 2.0 oligonucleotide array from AFFYMETRIX(R) molecule of interest. In a particular example, the test agent is (AFFYMETRIX(R), Santa Clara, Calif.). a siRNA that reduces or inhibits the activity (such as the In an example, the isolated nucleic acid molecules are expression) of one of the ovarian endothelial cell tumor 50 incubated with the array including oligonucleotides comple associated molecules listed in Tables 2, 4 or 5. For example, mentary to the ovarian endothelial cell tumor-associated mol the siRNA is directed to an ovarian endothelial cell tumor ecules listed in Table 2, 4 and/or 5 for a time sufficient to allow associated molecule listed in Table 2, 4 or 5 which is involved hybridization between the isolated nucleic acid molecules in angiogenesis, such as a molecule that is involved in at least and oligonucleotide probes, thereby forming isolated nucleic one of cell proliferation, tube formation or cell motility. 55 acid molecule:oligonucleotide complexes. The isolated In other examples, the test agent is an antibody. For nucleic acid molecule:oligonucleotide complexes are then example, the antibody is directed to specifically bind to an analyzed to determine if expression of the isolated nucleic ovarian endothelial cell tumor-associated protein encoded by acid molecules is altered. In Such example, an agent is con one of the genes listed in any of Tables 1, 2, 4 or 5 that are sidered is effective if the test agent decreases expression of upregulated in ovarian endothelial tumor cells. In a particular 60 ovarian endothelial tumor-associated molecules upregulated example, the antibody is directed to an ovarian endothelial in ovarian endothelial cell tumors relative to the absence of cell tumor-associated protein encoded by one of the genes the agent (such as a decrease of at least 2-, 3-, 4-, 5- or listed in Tables 2, 4 or 5 that is upregulated in ovarian endot 10-fold). Similarly, an agent is considered is effective if the helial tumor cells and which is involved in angiogenesis, Such test agent increases expression of ovarian endothelial tumor as a gene that is involved in at least one of cell proliferation, 65 associated molecules downregulated in ovarian endothelial tube formation or cell motility. In another example, the test cell tumors relative to the absence of the agent (such as an agent is a nucleic acid encoding one or more of the proteins increase of at least 2-, 3-, 4-, 5- or 10-fold). US 8,440,393 B2 49 50 Gene Expression Profile molecules that can bind to antibodies include, without limi The disclosed gene profile (as described above) can also be tation, Protein A and Protein G, both of which are available used to identify agents to treat an ovarian tumor, Such as a commercially. cancer, in a Subject. In an example, the gene expression pro Suitable labels for the antibody or secondary antibody file includes at least two of the ovarian endothelial cell tumor include various enzymes, prosthetic groups, fluorescent associated molecules listed in Table 1. Such as at least 5, at materials, luminescent materials, magnetic agents and radio least 7, at least 10, at least 20, at least 30, at least 40, at least active materials. Non-limiting examples of Suitable enzymes 50, at least 60, at least 70, at least 80, at least 90, at least 100, include horseradish peroxidase, alkaline phosphatase, beta at least 150, at least 200, at least 250, at least 300, at least 400, galactosidase, or acetylcholinesterase. Non-limiting at least 500, at least 600, at least 700, at least 800, at least 900, 10 examples of Suitable prosthetic group complexes include at least 1000, or at least 1100 molecules (for example, 2, 6, 8, streptavidin/biotin and avidin/biotin. Non-limiting examples 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 25, 30, 35, 40, 45, of suitable fluorescent materials include umbelliferone, fluo 50, 55,60, 65,70, 75, 80, 85,90, 95, 100,105,110, 120, 130, rescein, fluorescein isothiocyanate, rhodamine, dichlorotri 140, 150, 175, 225, 275,325, 350, 375, 450, 550, 650, 750, azinylamine fluorescein, dansyl chloride or phycoerythrin A 850, 950, 1050 or 1149 of those listed). 15 non-limiting exemplary luminescent material is luminol; a In a particular example, the gene expression profile non-limiting exemplary magnetic agent is gadolinium, and includes at least 2, at least 5, at least 7, at least 10, at least 20, non-limiting exemplary radioactive labels include 'I. 'I, at least 25, or at least 27 molecules (for example, 2, 3, 4, 5, 6, SS or H. 7,8,9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 22, 24, 25, 26, In an alternative example, ovarian endothelial cell tumor 27, 28 or 29 molecules) listed in Table 2, 4 and/or 5 that are associated molecules can be assayed in a biological sample associated with an at least six-fold increase in expression in by a competition immunoassay utilizing ovarian endothelial tumor endothelial cells. In a particular example, the at least cell tumor-associated molecule standards labeled with a two molecules include EGFL6 and TNFAIP6. In other par detectable substance and an unlabeled antibody that specifi ticular examples, the at least two ovarian endothelial cell cally binds the desired ovarian endothelial cell tumor-associ tumor-associated molecules include EGFL6, TNFAIP6, 25 ated molecule. In this assay, the biological sample (such as TWIST1, STC1, HOP, CSPG2, and PLXDC1. serum, tissue biopsy, or cells isolated from a tissue biopsy), In other particular examples, the gene expression profile the labeled ovarian endothelial cell tumor-associated mol includes at least 2, at least 5, at least 7, at least 10, at least 13, ecule standards and the antibody that specifically binds the or at least 15 molecules (for example, 2, 3, 4, 5, 6, 7, 8, 9, 10. desired ovarian endothelial cell tumor-associated molecule 11, 12, 13, 14, 15, 16 or 17 molecules) that are down-regu 30 are combined and the amount of labeled ovarian endothelial lated in ovarian tumor endothelial cells as listed in Table 3. cell tumor-associated molecule standard bound to the unla For example, the profile includes the seventeen ovarian endot beled antibody is determined. The amount of ovarian endot helial cell tumor-associated molecules listed in Table 3. helial cell tumor-associated molecule in the biological Detecting Ovarian Endothelial Cell Tumor-Associated Pro sample is inversely proportional to the amount of labeled teins 35 ovarian endothelial cell tumor-associated molecule standard As an alternative to analyzing the sample for the presence bound to the antibody that specifically binds the ovarian of nucleic acids, the presence of proteins can be determined endothelial cell tumor-associated molecule. using any method known in the art. In some examples, pro teins are purified before detection. For example, the effect of Methods of Diagnosing and Prognosing an Ovarian one or more test agents on an ovarian tumor can be deter 40 Tumor mined by incubating the biological sample with an antibody that specifically binds to one of the disclosed ovarian endot Metastasis is a major complication in the pathogenesis of helial cell tumor-associated proteins encoded by the genes tumors, such as ovarian cancer, and is typically indicative of listed in Tables 1-5. The primary antibody can include a poor prognosis. It is also known that angiogenesis is a crucial detectable label. For example, the primary antibody can be 45 factor in the progression of Solid tumors and metastases, directly labeled, or the sample can be subsequently incubated including ovarian cancer. The formation of the vascular with a secondary antibody that is labeled (for example with a stroma plays an important role in the pathophysiology of fluorescent label). The label can then be detected, for example malignancy. For instance, in the absence of vascular Support by microscopy, ELISA, flow cytometry, or spectrophotom tumors may become necrotic, or even apoptotic. In contrast, etry. In another example, the biological sample is analyzed by 50 the onset of angiogenesis marks a phase of rapid proliferation, Western blotting for the presence or absence of the specific local invasion, and ultimately metastasis. ovarian endothelial cell tumor-associated molecule. In some Without wishing to be bound to a particular theory, it is examples, the biological sample is analyzed by mass spec proposed that an alteration in the expression of the disclosed trometry. ovarian endothelial tumor-associated molecules associated In one example, the antibody that specifically binds an 55 with angiogenesis, Such as molecules involved in cell prolif ovarian endothelial cell tumor-associated molecule (such as eration, cell motility or tube formation, including those dis those listed in Table 1) is directly labeled with a detectable closed in FIG. 5 (such as EZH2), is related to enhanced label. In another example, each antibody that specifically ovarian tumor cell metastasis and a poor clinical outcome. binds an ovarian endothelial cell tumor-associated molecule Thus, methods of diagnosing or prognosing an ovarian tumor (the first antibody) is unlabeled and a second antibody or 60 that overexpresses at least one pro-angiogenic ovarian endot other molecule that can bind the human antibody that specifi helial cell tumor-associated molecule (such as those listed in cally binds the respective ovarian endothelial cell tumor Tables 1, 2, 4 and 5 that are upregulated in ovarian endothelial associated molecule is labeled. As is well known to one of tumor cells; e.g., EZH2) or underexpresses at least one proan skill in the art, a second antibody is chosen that is able to giogenic ovarian endothelial cell tumor associated molecule specifically bind the specific species and class of the first 65 (such as those listed in Tables 1 and 3 that are downregulated antibody. For example, if the first antibody is a human IgG, in ovarian endothelial tumor cells), are disclosed. In some then the secondary antibody can be an anti-human-IgG. Other examples, such methods can be used to identify those Subjects US 8,440,393 B2 51 52 that will benefit from the disclosed treatment methods. For angiogenic ovarian endothelial cell tumor-associated mol example, such diagnostic methods can be performed prior to ecules detected in a tumor sample are compared to the level of the Subject undergoing the treatments described above. In Such molecules detected in a sample obtained from a subject other examples, these methods are utilized to predict the that does not have an ovarian tumor or has a non-metastatic metastatic potential of the ovarian cancer, a poor prognosis, or 5 ovarian tumor. In certain examples, detection of at least a combinations thereof. In one particular example, these meth 2-fold, such as by at least 3-fold, at least 4-fold, at least 6-fold ods are utilized to predicta poor prognosis, such as to indicate or at least 10-fold alteration in the relative amount of the a decreased Survival time. pro-angiogenic ovarian endothelial cell tumor-associated In an example, the method includes detecting expression of molecules in a tumor sample, as compared to the relative at least one angiogenic ovarian endothelial cell tumor-asso 10 amount of Such molecules in a control indicates that the ciated molecule listed in Tables 1-5, such as at least two, at least 5, at least 7, at least 10, at least 20, at least 30, at least 40, Subject has a tumor with metastatic potential, has a tumor that at least 50, at least 60, at least 70, at least 80, at least 90, at least has metastasized, has a poor prognosis, or combinations 100, at least 150, at least 200 ovarian endothelial cell tumor thereof. In some examples, detection of statistically similar associated molecules related to angiogenesis (for example, 1. 15 relative amounts of pro-angiogenic ovarian endothelial cell 2, 3, 4, 5, 6, 7, 8, 9, 10, 12, 15, 17, 20, 25, 30, 35, 40, 45, 50, tumor-associated molecules observed in a tumor sample, as 55, 60, 65,70, 75, 80, 85,90, 100,125, 150, 175, 200, or 250 compared to the relative amount of such molecules in a con listed in Tables 1-5) in a sample from the subject exhibiting trol sample that is not metastatic, indicates that that Subject one or more symptoms associated with ovarian cancer. does not have a tumor with metastatic potential, does not have In an example, the specific angiogenic ovarian endothelial a tumor that has metastasized, has a good prognosis, or com cell tumor-associated molecule, such as EZH2, is detected in binations thereof. a biological sample. In a particular example, the biological In a specific example, the method includes detecting and sample is a sample taken from a subject with ovarian epithe comparing the nucleic acid expression levels of the pro-an lial cancer. In a particular example, the biological sample is a giogenic ovarian endothelial cell tumor-associated molecules tumor biopsy. In another example, the angiogenic ovarian 25 such as DNA, cDNA, or mRNAs. In a specific example, the endothelial cell tumor-associated molecule is detected in a method includes detecting and comparing the mRNA expres serum sample. For example, the ovarian endothelial cell sion levels of the pro-angiogenic ovarian endothelial cell tumor-associated molecule is detected in a serum sample if tumor-associated molecules. For example, Such expression the specific molecule is secreted or located on a cell Surface can be measured by real time quantitative polymerase chain Susceptible to enzymatic cleavage. 30 reaction or microarray analysis. In a particular example, the In one example, detection of at least one angiogenic ova disclosed gene expression profile is utilized to diagnosis and/ rian endothelial cell tumor-associated molecule listed in any or prognosis an ovarian tumor. of Tables 1, 2, 3, 4 or 5, such as detection of EZH2, in a Detection of Ovarian Endothelial Cell Tumor-Associated biological sample from the Subject is used to diagnose or Nucleic Acids prognose an ovarian tumor. Methods of detecting Such mol 35 Nucleic acids can be detected by any method known in the ecules in a sample are known in the art and are routine. In art. In some examples, nucleic acids are isolated, amplified, or Some examples, the relative amount of pro-angiogenic ova both, prior to detection. In an example, the biological sample rian endothelial cell tumor-associated molecules present is can be incubated with primers that permit the amplification of determined, for example by quantitating the expression level one or more of the disclosed ovarian endothelial cell tumor of such molecules. For example, the relative or absolute quan 40 associated mRNAs, under conditions sufficient to permit tity of the at least one angiogenic ovarian endothelial cell amplification of Such products. For example, the biological tumor-associated molecule in a sample can be determined. sample is incubated with probes that can bind to one or more The activity such as the expression level of the disclosed of the disclosed ovarian endothelial cell tumor-associated pro-angiogenic ovarian endothelial cell tumor-associated nucleic acid sequences (such as cDNA, genomic DNA, or molecules in a sample obtained from a subject is compared to 45 RNA (such as mRNA)) under high stringency conditions. The a control. In one example, an increase in expression of one or resulting hybridization can then be detected using methods more of the angiogenic ovarian endothelial cell tumor-asso known in the art. In one example, a therapeutic agent is ciated molecules upregulated in ovarian tumor endothelial identified by applying isolated nucleic acid molecules to an cells (such as those listed in Table 2) as compared to a non array in which the isolated nucleic acid molecules are tumor control or reference value indicates the presence of an 50 obtained from a biological sample including ovarian endot ovarian tumor, the ovarian tumor is metastatic, the ovarian helial cancer cells for example following treatment with the tumor is likely to become metastatic, a poor prognosis or a one or more test agents. In Such example, the array includes combination thereof. In some examples, a decrease in expres oligonucleotides complementary to all ovarian endothelial sion of one or more of the angiogenic ovarian endothelial cell cell tumor-associated genes listed in Table 1. In a particular tumor-associated molecules that is downregulated in ovarian 55 example, the array is a commercially available array Such as tumor endothelial cells (such as those listed in Table 3 or a U133 Plus 2.0 oligonucleotide array from AFFYMETRIX(R) VASH1) as compared to a non-tumor control or reference (AFFYMETRIX(R), Santa Clara, Calif.). value indicates the presence of an ovarian tumor, the ovarian In an example, the isolated nucleic acid molecules are tumor is metastatic, the ovarian tumor is likely to become incubated with the array including oligonucleotides comple metastatic, a poor prognosis or a combination thereof. 60 mentary to the ovarian endothelial cell tumor-associated mol For example, the level of the angiogenic ovarian endothe ecules listed in Tables 2, 3, 4 and/or 5 for a time sufficient to lial cell tumor-associated molecules, such as the level of allow hybridization between the isolated nucleic acid mol EZH2, detected can be compared to a non-tumor control or ecules and oligonucleotide probes, thereby forming isolated reference value, such as a value that represents a level of nucleic acid molecule: oligonucleotide complexes. In a par angiogenic ovarian endothelial cell tumor-associated mol 65 ticular example, the isolated nucleic acid molecules are incu ecules expected if an ovarian tumor is or is not metastatic or bated with the array including oligonucleotides complemen is a low grade tumor or early stage tumor. In one example, the tary to at least EZH2. The isolated nucleic acid molecule: US 8,440,393 B2 53 54 oligonucleotide complexes are then analyzed to determine if immunobeads (P1H12 antibody was from Chemicon, expression of the isolated nucleic acid molecules is altered. Temecula, Calif.), and the beads linked to secondary antibody Gene Expression Profile were from Dynal Biotech. Immunostaining was then per The disclosed gene profile (as described above) can also be formed using von Willebrand factor and 4,6-diamidino-2- used in the diagnosis and prognosis of an ovarian tumor in a phenylindole nuclear staining to confirm the purification of Subject. In an example, the gene expression profile includes at endothelial cells. least two of the ovarian endothelial cell tumor-associated Total RNA Amplification for AFFYMETRIXR GENE molecules listed in Table 1, such as at least 5, at least 7, at least CHIP(R) Hybridization and Image Acquisition. 10, at least 20, at least 30, at least 40, at least 50, at least 60, To successfully generate sufficient labeled cRNA for at least 70, at least 80, at least 90, at least 100, at least 150, at 10 microarray analysis from 25 ng of total RNA, two rounds of least 200, at least 250, at least 300, at least 400, at least 500, amplification were necessary. For the first round synthesis of at least 600, at least 700, at least 800, at least 900, at least 1000, or at least 1100 molecules (for example, 2, 6, 8, 9, 10, double-stranded cDNA, 25 ng of total RNA was reverse tran 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 25, 30, 35, 40, 45, 50,55, scribed using the Two-Cycle cDNA Synthesis Kit (AF 60, 65, 70, 75, 80, 85,90, 95, 100,105, 110, 120, 130, 140, 15 FYMETRIX(R), Santa Clara, Calif.) and oligo-dT24-T7 150, 175, 225, 275,325, 350, 375,450, 550, 650, 750, 850, 950, 1050 or 1149 of those listed). (SEQ ID NO: 1: 5 - GGCCAGTGAATTGTAATACGACT CACTATAGG In a particular example, the gene expression profile GAGGCGG-3 '' ) includes at least 1, at least 3, at least 5, at least 7, at least 10, at least 20, at least 25, or at least 27 molecules (for example, primer according to the manufacturers instructions followed 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, by amplification with the MEGA script T7 Kit (Ambion, Inc., 22, 24, 25, 26, 27, 28 or 29 molecules) listed in Table 2, 4 Austin, Tex.). After cleanup of the cRNA with a GENE and/or 5 that are associated with angiogenesis, such as mol CHIPR) Sample Cleanup Module IVT column (AFFYME ecules involved in cell proliferation, cell motility and/or tube TRIX(R), second round double stranded cDNA was amplified formation. In a particular example, the at least one molecule 25 using the IVT Labeling Kit (AFFYMETRIX(R). A 15.0 ug includes EZH2 aliquot of labeled product was fragmented by heat and ion In other particular examples, the gene expression profile mediated hydrolysis at 94°C. for 35 minutes in 24 ul HO and includes at least 2, at least 5, at least 7, at least 10, at least 13, 6 ul of 5x Fragmentation Buffer (AFFYMETRIX(R). The or at least 15 molecules (for example, 2, 3, 4, 5, 6, 7, 8, 9, 10. fragmented cRNA was hybridized for 16 hr at 45° C. in a 11, 12, 13, 14, 15, 16 or 17 molecules) that are down-regu 30 Hybridization Oven 640 to a U133 plus 2.0 oligonucleotide lated in ovarian tumor endothelial cells as listed in Table 3. array (AFFYMETRIX(R). Washing and staining of the arrays For example, the profile includes the seventeen ovarian endot with phycoerythrin-conjugated streptavidin (Molecular helial cell tumor-associated molecules listed in Table 3. Probes, Eugene, Oreg.) was completed in a Fluidics Station Detecting Ovarian Endothelial Cell Tumor-Associated Pro 450 (AFFYMETRIX(R). The arrays were then scanned using teins 35 a confocal laser GENECHIP(R) Scanner 3000 and GENE As an alternative to analyzing the sample for the presence CHIPR Operating Software (AFFYMETRIX(R). of nucleic acids, alterations in protein expression can be mea Data Normalization and Filtering. sured by methods known in the art such as Western blot Global normalization at a target value of 500 was applied to analysis, mass spectrometry, immunoassay or a protein all 15 of the arrays under consideration using GENECHIPR) microarray (as described above). For example, the metastatic 40 Operating Software (AFFYMETRIX(R). Normalized data potential of an ovarian tumor can be determined by using a were uploaded into the National Cancer Institute's Microar protein array that includes one or more capture agents, such as ray Analysis Database (mAdb) for quality control screening antibodies that are specific for the one or more disclosed and collation prior to downstream analyses. Biometric ovarian endothelial tumor-associated molecules that are Research Branch (BRB) ArrayTools version 3.2.2 software related to angiogenesis, such as molecules that play a role in 45 developed by Drs. Richard Simon and Amy Peng Lam of the cell proliferation, cell motility or tube formation, such as Biometrics Research Branch of the National Cancer Institute EZH2. was used to filter and complete the statistical analysis of the The disclosure is further illustrated by the following non array data. BRB-ArrayTools is a multifunctional Excel add-in limiting Examples. that contains utilities for processing and analyzing microar 50 ray data using the R version 2.0.1 environment (R Develop Example 1 ment Core Team, 2004). Of the 47,000 transcripts represented on the array, hybridization control probe sets and probe sets Materials and Methods for Examples 2-7 scored as absent at C.1=0.05 or marginal (M) at C. 2–0.065 were excluded. In addition, only those transcripts present in Sample Preparation. 55 greater than 50% of the arrays and displaying a variance in the Fresh tissue samples (5 normal ovaries and 10 epithelial high top 50th percentile were evaluated. grade, stage III or IV invasive serous ovarian cancers) were Class Comparison Analysis. obtained from patients undergoing primary Surgical explora Differentially expressed genes were identified for tumor and tion at the M.D. Anderson Cancer Center. The minced tissue normal endothelial cell specimens using a multivariate per was digested with collagenase A, elastase and DNase 1 at 37° 60 mutation test in BRB-ArrayTools (Simon et al., “Design and C. for 90 minutes to yield a single cell suspension. A number Analysis of DNA Microarray Investigations' Springer-Ver of negative selections followed including removal of platelets lag, 2003). A total of 2000 permutations were completed to and red blood cells (RBCs) by Percoll separation, removal of identify the list of probe sets with a false discovery rate less epithelial cells using M450 beads, which are prebound to than 10% at a confidence of 95%. Differential expression was BerEP4 antibody, removal of leukocytes using anti-CD-14, 65 considered significant at a p<0.001. A random-variance t-test CD-45, and CD-64 beads (Dynal Biotech, Brown Deer, Wis.). was selected to permit the sharing of information among Positive selection was performed with P1H12 (CD 146) probe sets within class variation without assuming that all of US 8,440,393 B2 55 56 the probe sets possess the same variance (Wright et al., Bio mined by BLAST search), and separate sequences designed informatics 19: 2448-2455, 2003). A global assessment of to target EZH2, Jagged 1, or PTK2 mRNA. The Jagged 1 whether expression profiles were different between classes siRNA target sequence is SEQ ID NO: 2 (5'-CTGCATT was also performed. During each permutation the class labels TAGGGAGTATTCTA-3). The EZH2 siRNA was targeted to were reassigned randomly and the p-value for each probe set the region corresponding to residues 85-106 of human EZH2 recalculated. The proportion of permutations yielding at least (Gene accession No. NM004456; 5'-AACCATGTTTA as many significant genes as the actual data set at a CAACTATCAA-3: SEQID NO:3). The EZH2 siRNA sense p-values0.001 was reported as the significance level of the sequence was 5'-CCAUGUUUACAACUAUCAAtt-3: SEQ global test. ID NO: 4) and the antisense sequence was 3'-ttGGUA Pathway Analysis. 10 Differentially regulated genes identified in a series of 48 CAAAUGUUGAUAGUU-5; SEQ ID NO: 5). For in vitro late-stage (III and IV) high-grade (Hurwitz et al., N. Engl. J. delivery, siRNA (5ug) was incubated with 30 ul, RNAi Fect Med. 350: 2335-2342, 2004) microdissected papillary serous transfection reagent (QIAGENR) for 10 min at room tem ovarian carcinomas, as compared to 10 normal ovarian Sur perature and added to cells in culture at 80% confluence in 35 face epithelial brushings (Bonome et al., Cancer Res. 65: 15 mm culture plates. 10602-10612, 2005), were categorized by cellular compo Cell Migration Assay. nent according to the (GO) ontological hier Unstimulated motility was determined in membrane invasion archy. Epithelial genes associated with the cell membrane, culture system chambers containing polycarbonate filter extracellular matrix, and extracellular region were used as (with 10 um pores) that had been soaked in 0.1% gelatin, as central nodes to identify signaling pathways modulated in described previously (Sood et al., Am. J. Pathol. 165: 1087 tumor-associated endothelial cell isolates. This was accom 1095, 2004). HUVECs (1x105) were seeded in each upper plished using Pathway Assist version 3.0 software (Iobion well, allowed to incubate at 37° C. for 6 hr in Dulbecco's Informatics LLC, La Jolla, Calif.). This software package modified Eagle's medium (DMEM) containing 15% serum, contains over 500,000 documented protein interactions and Subsequently processed as described for the invasion acquired from MedLine using the natural language process 25 assay. ing algorithm MEDSCAN. The proprietary database can be Tube Formation Assay. used to develop a biological association network (BAN) to Matrigel (12.5 mg/ml) was thawed at 4°C. and 50 ul were identify putative co-regulated signaling pathways using quickly added to each well of a 96-well plate and allowed to expression data. solidify for 10 minat37°C. The wells were then incubated for qRT-PCR Validation. 30 6 h at 37° C. with HUVECs (20,000 cells/well), which had Quantitative real-time PCR (qRT-PCR) was performed on previously been treated for 18 h with the indicated siRNA. 100 ng of double-amplified product from the 15 specimens The formation of capillary-like structures was examined using primer sets specific for 23 select genes, and the house microscopically and photographs (50x) were taken using a keeping genes GAPDH, GUSB, and cyclophilin. An iCycler RETIGAR 1300 camera and a ZEISS.R. Axiovert S100 iQ Real-Time PCR Detection System (BIORADR Laborato 35 microscope. The extent to which capillary-like structures ries, Hercules, Calif.) was used in conjunction with the Quan formed in the gel was quantified by analysis of digitized tiTect SYBR Green RT-PCR Kit (QIAGENR, Inc., Valencia, images to determine the thread length of the capillary-like Calif.) according to previously described cycling conditions network, using a commercially available image analysis pro (Donninger et al. Oncogene 23: 8065-8077, 2004). To cal gram (Northern Eclipse, North Tonawanda, N.Y.). culate the relative expression for each gene, the 2-, 40 method was used averaging the C values for the three house Example 2 keeping genes for a single reference gene value (Livak and Schmittgen, Methods 25: 402-408, 2001). Purity of Isolated Endothelial Cells Immunohistochemical Staining. Paraffin sections were stained for the following antibodies: 45 This example illustrates the purity of the endothelial cell rabbit anti-Fyn at 1:400 (Santa Cruz Biotechnology, Inc., samples utilized in the disclosed microarray analyses. Santa Cruz, Calif.), or rabbit anti-Fak at 1:50, mouse anti According to the methods described in Example 1, Samples MMP-9 at 1:40 (Oncogene-Research Products, Boston, were immunostained with endothelial cell markers P1H12 Mass.), anti-B2-Arrestin at 1:200 (Santa Cruz Biotechnology, and von Willebrand factor to determine endothelial cell Inc.), anti-PLXDC1 at 1:200 (Abcam, Inc., Cambridge, 50 purity. Immunostaining revealed that the employed purifica Mass.), or anti-Jagged 1 at 1:200 (Santa Cruz, Biotechnology, tion technique yielded endothelial cell purity of >95% in all Inc.) diluted in PBS at 4° C. After three washes in PBS, samples. Thus, the disclosed isolation technique resulted in a sections were incubated with secondary antibody for 1 hr at highly pure population of endothelial cells. room temperature. Positive reactions were rendered visible by incubating the slides with stable 3.3-diaminobenzidine for 55 Example 3 5-10 min. The sections were rinsed with distilled water, coun terstained with Gill's hematoxylin for 30s, and mounted with Development of Gene Expression Profile for Ovarian Universal Mount (Research Genetics, Huntsville, Ala.). The Tumor-Endothelial Cells intensity of protein expression in the endothelial cells was evaluated using OPTIMAS 6.5 software and the mean optical 60 This example provides a gene expression profile for ova density (OD) was calculated from 5 normal ovarian and 5 rian tumor endothelial cells. ovarian cancer samples. Ten vessels were selected randomly According to the methods described in Example 1, gene from each sample for the measurements. expression differences in purified endothelial cells from 10 Small interfering RNA (siRNA). invasive epithelial ovarian cancers and 5 normal (non-tumor) The small interfering RNA (siRNA) constructs were pur 65 ovaries were determined by using the AFFYMETRIX(R) chased from QIAGENR) (Germantown, Md.): a control Human U133 Plus 2.0 Gene Chip platform. The nucleic acid sequence with no homology to any human mRNA (as deter sequence of each AFFYMETRIX(R) probe listed in the tables US 8,440,393 B2 57 58 below is herein incorporated by reference, and is available helial tumor sample and a negative fold change indicates the from the AFFYMETRIXR website. As illustrated in Table 1, gene was downregulated in Such sample. A multivariate per 1149 genes were identified as being differentially regulated mutation t-test (p<0.001) providing 95% confidence that the 22-fold in endothelium derived from epithelial ovarian can number of false discoveries did not exceed 10% of the com cers compared to normal ovarian tissue. A positive fold plete gene list identified. In addition, global analysis of the change indicates the gene was upregulated in ovarian endot gene list returned ap value-5x10'. TABLE 1 Gene expression profile. Affymetrix Parametric Fold Change Probe Set P-Value (Tumor? Normal) Gene Symbol Description Map 17 at O.OOOS215 4.4 HSPA6 heat shock 70 kDa protein 6 1q23 (HSP7OB') (HSPA6), mRNA. 552365 at O.OOO8711 6.6 SCIN scinderin (SCIN), mRNA. 7p21.3 552767 a at 0.0002277 -6.5 HS6ST2 heparan Sulfate 6-O- Xq26.2 sulfotransferase 2 (HS6ST2), mRNA. 552790 a. at 0.0009635 -2.2 TLOC1 translocation protein 1 3q26.2 (TLOC1), mRNA. 55.2889 a. at 0.0003238 2.5 XTP7 protein 7 transactivated by 19q13.32 hepatitis B virus X antigen (HBxAg) (XTP7), mRNA. 553185 at O.OOO128 4.6 RASEF RAS and EF-hand domain 9q21.32 containing 553186 X at 3.33E-05 5.3 RASEF RAS and EF-hand domain 9q21.32 containing 553407 at O.OOO3912 2.5 MACF1 Glycine-rich protein (GRP3S) 1p32-p31 553538. S. at 0.000594 2.2 Unknown 553569 at O.OOO1391 2.6 Unknown 553.570 x at O.OOO2337 2.5 Unknown 5.53575 at 2.8OE-06 4.6 Unknown 553909 X at 0.0005681 3.8 C10orf6 10 open reading 10q24.32 rame 6 553959 a. at O.OOO9778 2.3 B3GALT6 UDP-Gal:beta Gal beta1,3- 1p36.33 galactosyltransferase polypeptide 6 (B3GALT6), mRNA. 553983 at O.OOO152 2.5 DTYMK deoxythymidylate kinase 2a37.3 (thymidylate kinase) (DTYMK), mRNA. 554168 a at 0.0003894 2.1 SH3KBP1 SH3-domain kinase binding Xp22.1-p21.3 protein 1 (SH3KBP1), transcript variant 2, mRNA. 554.309 at OOOO6114 2.3 EIF4G3 Eukaryotic translation initiation 1p36.12 factor 4 gamma, 3 554334 a at 7.15E-05 5.2 DNAJA4 DnaJ (Hsp40) homolog, 15q25.1 subfamily A, member 4 (DNAJA4), mRNA. 554455 at O.OOO7696 2.4 LINS1 lines homolog 1 (Drosophila) 15q26.3 (LINS1), transcript variant 2, mRNA. 554.464 a at 0.000263 -2.8 CRTAP cartilage associated protein 3p22.3 (CRTAP), mRNA. 554595 at O.OOOS225 2.6 SYMPK Symplekin 19q13.3 554640 at O.OOO4537 2.6 PALM2-AKAP2 Paralemmin 2 9q31-q33 554678 s at 0.0002435 -2 HNRPDL Heterogeneous nuclear 4q13-q21 ribonucleoprotein D-like 554703 at O.OOO8778 2.1 ARHGEF10 Rho guanine nucleotide 8p23 exchange factor (GEF) 10 555014 x at 0.0004322 3.5 OKASW-cl.92 555.241 at O.OOO1337 3.2 Hypothetical gene Supported by 8q21.2 BCOSSO92 555.243 x at 0.0001753 2.8 Hypothetical gene Supported by 8q21.2 BCOSSO92 555374 at O.OOO3108 4 TTL Tubulin tyrosine ligase 2d 13 555823 at O.OOO1833 -2 BS3076 14 556126 s at 1.04E-05 3.5 GPATC2 G patch domain containing 2 1q41 556138 a. at 0.0001589 2.9 COLSA1 Collagen, type V, alpha 1 9q34.2-q34.3 556185 a. at 0.0002463 2.6 CDNA clone IMAGE: 5260162 7 556242 a. at 3.82E-05 2.1 Homo sapiens, clone 8 IMAGE: 3885623, mRNA 556316 s at 0.0005561 3.3 MIF Macrophage migration 22q11.23 inhibitory factor (glycosylation inhibiting factor) 556499 s at 0.0005246 2.5 COL1A1 collagen, type I, alpha 1 17q21.3-q22.1 (COL1A1), mRNA. 556835 S. at 0.0002018 2.2 Transcribed locus 11 557432 at O.OOO1321 3.6 RASAL2 RAS protein activator like 2 1q24 US 8,440,393 B2 59 60 TABLE 1-continued Gene expression profile. Affymetrix Parametric Fold Change Probe Set P-Value (Tumor? Normal) Gene Symbol Description Map 557527 at O.OOO3572 2.5 RUNX1 Runt-related transcription factor 21q22.3 (acute myeloid leukemia 1: aml1 oncogene) 558O19 at 2.09E-OS -2.9 DST Dystonin 6p12-p11 558048 x at 2.01E-05 4.1 Unknown 558292 S at 0.0007105 2 PIGW phosphatidylinositol glycan, 17q12 class W (PIGW), mRNA. 558426 X at 0.000189 2.9 Chromosome 7 open reading 7 rame 19 558487 a. at 0.0002657 -2.6 TMED4 Transmembrane emp24 protein 7p13 transport domain containing 4 558836 a S.35E-05 3.2 MRNA, cDNA 2 DKFZp667A182 (from clone DKFZp667A182) 559060 a. at 5.16E-05 3 KIAA1961 KIAA1961 gene 5q23.3 559078 a 6.58E-OS 4.6 BCL11A B-cell CLLlymphoma 1 1A 2p16.1 (Zinc finger protein) 559101 a 9.12E-OS 2.6 FYN FYN oncogene related to SRC, 6q21 FGR, YES 5594.10 a 4.66E-OS 3.6 Unknown 559436 X at 4.54E-05 4.8 ARRB2 Arrestin, beta 2 17p13 559585 a. O.OOO489 3.2 FLJ31033 Hypothetical protein FLJ31033 4q32.3 559593 a at 0.0005238 2.9 CRSP7 Cofactor required for Sp1 19p 13.11 transcriptional activation, subunit 7, 70 kDa 560817 a O.OOO1564 2.5 MOV10 Mov10, Moloney leukemia virus 1p13.2 0, homolog (mouse) 562062 a O.OOO114 4 Homo sapiens transcribed sequence with weak similarity to protein ref: NP O55301.1 (H. Sapiens) neuronal thread protein Homo sapiens 1562063 x at 0.0003481 3.4 Homo sapiens transcribed sequence with weak similarity to protein ref: NP O55301.1 (H. Sapiens) neuronal thread protein Homo sapiens 562270 at O.OOO382 4 ARHGEF7 Rho guanine nucleotide 13q34 exchange factor (GEF) 7 562271 x at 1.81E-05 3.9 ARHGEF7 Rho guanine nucleotide 13q34 exchange factor (GEF) 7 562456 at O.OOO8617 2.3 MRNA, cDNA 11 DKFZp566C0924 (from clone DKFZp566C0924) 563357 at O.OOO2718 3.4 SERPINB9 Serpin peptidase inhibitor, clade 6p25 B (ovalbumin), member 9 (SERPINB9), mRNA. 565579 at O.OOO3108 3.7 TATDN2 TatD DNase domain containing 2 3p25.3 565823 at 1.91E-05 4.8 7-Sep septin 7 (SEPT7), transcript 7p14.3-p14.1 variant 2 mRNA. 565974 at O.OOO8723 2.8 SUV42OH1 Suppressor of variegation 4-20 11 q13.2 homolog 1 (Drosophila) 56.6887 X at 8.32E-05 4.3 KIAAO284 KIAAO284 14q32.33 568619 S at 0.0002665 2.2 LOC162O73 Hypothetical protein 16p12.3 LOC162O73 568954 S at 0.0001751 2.2 Unknown 56.9519 at O.OOO61.99 2.5 FL21272 hypothetical protein FLJ21272 1q21.2 56.9872 a. at 0.0004892 2.4 Homo sapiens, clone 16 MAGE: 5242623 570061 at 8.83E-OS 3.2 CDNA clone IMAGE: 4555.030 3 570143 at 0.0005727 2.4 Homo sapiens, clone 8 MAGE: 3932570, mRNA 57O185 at O.OOO3O81 3 Homo sapiens, clone 10 MAGE: 5766850, mRNA 200005 at O.OOOSSO9 -2 EIF3S7 eukaryotic translation initiation 22d 13.1 actor 3, subunit 7 zeta, 66/67 kDa (EIF3S7), mRNA. 200010 at O.OOO1997 -2.2 RPL11 Ribosomal protein L11 1p36.1-p35 200013 at 2. SOE-OS -2.6 RPL24 ribosomal protein L24 (RPL24), 3q12 mRNA. 200021 at O.OOO1O3S 2.5 CFL1 cofilin 1 (non-muscle) (CFL1), 11 q13 mRNA. 200022 at O.OOO2003 -2.1 RPL18 ribosomal protein L18 (RPL18), 19q13 mRNA. US 8,440,393 B2 61 62 TABLE 1-continued Gene expression profile. Affymetrix Parametric Fold Change Probe Set P-Value (Tumor? Normal) Gene Symbol Description Map 200023 s at 1.61E-05 -2.4 EIF3S5 eukaryotic translation initiation 11 p 15.4 factor 3, Subunit 5 epsilon, 47 kDa (EIF3S5), mRNA. 200024 at O.OOO4964 -2.5 RPS5 ribosomal protein S5 (RPS5), 19q13.4 mRNA. 20007.4 S at 0.0001721 -2.2 RPL14 ribosomal protein L14 (RPL14), 3p22-p21.2 mRNA. 200081 s at 6.2OE-06 -3.3 RPS6 ribosomal protein S6 (RPS6), 9p21 mRNA. 200642 at O.OOO7047 -2 SOD1 Superoxide dismutase 1, Soluble 21q22.11 (amyotrophic lateral sclerosis 1 (adult)) (SOD1), mRNA. 200651 at O.OOO1597 -2.3 GNB2L1 guanine nucleotide binding 5q.35.3 protein (G protein), beta polypeptide 2-like 1 (GNB2L1), mRNA. 200665 S at 0.00039 2.7 SPARC Secreted protein, acidic, 5q31.3-q32 cysteine-rich (osteonectin) (SPARC), mRNA. 200676 S. at 0.0007037 -2.5 UBE2IL3 ubiquitin-conjugating enzyme 22d 11.21 E2L 3 (UBE2IL3), transcript variant 1 mRNA. 200700 s at 4.15E-05 2.3 KDELR2 KDEL (Lys-Asp-Glu-Leu) 7p22.1 endoplasmic reticulum protein retention receptor 2 (KDELR2), mRNA. 200734 S. at 6.61E-05 2.5 ARF3 ADP-ribosylation factor 3 12q13 (ARF3), mRNA. 20073S X at 9.63E-05 -2.1 NACA nascent-polypeptide-associated 12q23-q24.1 complex alpha polypeptide (NACA), mRNA. 200755 s at 0.0007055 2.1 CALU calumenin (CALU), mRNA. 7q32 200760 S at 5.53E-05 -2.6 ARL6IPS ADP-ribosylation-like factor 6 3p14 interacting protein 5 (ARL6IP5), mRNA. 200806 s at 0.0008155 -2 HSPD1 heat shock 60 kDa protein 1 2a33.1 (chaperonin) (HSPD1), nuclear gene encoding mitochondrial protein, transcript variant 2, mRNA. 200811 at 1.2SE-OS -3.6 CIRBP cold inducible RNA binding 9p13.3 protein (CIRBP), mRNA. 200827 at O.OOO4906 2.3 PLOD1 procollagen-lysine 1,2- p36.3-p36.2 oxoglutarate 5-dioxygenase 1 (PLOD1), mRNA. 2008.66 s at 0.0004.451 -2.1 PSAP prosaposin (variant Gaucher Oq21-q22 disease and variant metachromatic leukodystrophy) (PSAP), mRNA. 200883 at 7.19E-OS -3.2 UQCRC2 ubiquinol-cytochrome c 6p12 reductase core protein II (UQCRC2), mRNA. 200906 S. at 0.0002674 -3.2 KIAAO992 palladin (KIAA0992), mRNA. 4q32.3 200920 S. at 0.0001341 -2.3 BTG B-cell translocation gene 1, anti- 2q22 proliferative (BTG1), mRNA. 200937 S at 5.83E-05 -2.4 RPL5 ribosomal protein L5 (RPL5), p22.1 mRNA. 200951 S. at 0.000314 4.1 CCND2 cyclin D2 (CCND2), mRNA. 2p13 200953 S at 0.0001997 2.6 CCND2 cyclin D2 (CCND2), mRNA. 2p13 200965 S at 0.000112 -5.9 ABLIM1 actin binding LIM protein 1 Oq25 (ABLIM1), transcript variant 4, mRNA. 200999 S at 0.0009198 2 CKAP4 cytoskeleton-associated protein 2d23.3 4 (CKAP4), mRNA. 2010O8 s at 0.0009045 -2.7 TXNIP thioredoxin interacting protein q21.1 (TXNIP), mRNA. 201009 s at 0.0004645 -2.8 TXNIP thioredoxin interacting protein q21.1 (TXNIP), mRNA. 201018 at O.OOO7096 -2.3 EIF1AX eukaryotic translation initiation Xp22.12 factor 1A, X-linked (EIF1AX), mRNA. 201023 at 2.69E-OS -2.4 TAF7 TAF7 RNA polymerase II, 5q31 TATA box binding protein (TBP)-associated factor, 55 kDa (TAF7), mRNA. US 8,440,393 B2 63 64 TABLE 1-continued Gene expression profile. Affymetrix Parametric Fold Change Probe Set P-Value (Tumor? Normal) Gene Symbol Description Map O30 x at 2.4OE-05 -2.2 lactate dehydrogenase B 12p12.2-p12.1 (LDHB), mRNA. O36 s at O.OOOS/34 -3.7 HADHSC L-3-hydroxyacyl-Coenzyme A dehydrogenase, short chain (HADHSC), mRNA. 054 at 1.41E-05 -2.2 HNRPAO heterogeneous nuclear ribonucleoprotein AO (HNRPAO), mRNA. O76 at 182E-05 -2.4 NHP2L1 NHP2 non-histone chromosome 22d 13.2-q13.31 protein 2-like 1 (S. cerevisiae) (NHP2L1), transcript variant 2, mRNA. 085 s at O.OOO222 -2.3 SON SON DNA binding protein 21q22.11 (SON), transcript varianta, mRNA. 088 at 9.1OE-06 2.7 KPNA2 karyopherin alpha 2 (RAG 17q23.1-q23.3 cohort 1, importin alpha 1) (KPNA2), mRNA. 101 S. at 3.64E-05 -2.9 BCLAF1 BCL2-associated transcription factor 1 (BCLAF1), mRNA. 129 at O.OOO2614 -2.6 SFRS7 splicing factor, arginine?serine 2p22.1 rich 7, 35 kDa (SFRS7), transcript variant 1, mRNA. 133 s at O.OOO1686 -2.5 PA2 praja 2, RING-H2 motif 2p22.1 containing (PJA2), mRNA. 154 X at O.OOO1159 -2.2 RPL4 ribosomal protein L4 (RPL4), 15q22 mRNA. 163 s at O.OOO3882 2.2 insulin-like growth factor binding protein 7 (IGFBP7), mRNA. 193 at O.OOO4945 -2.9 IDEH1 isocitrate dehydrogenase 1 2a33.3 (NADP+), soluble (IDH1), mRNA. 204 S at 1.43E-05 2.2 RR BP1 Ribosome binding protein 1 20p12 homolog 180 kDa (dog) 206 s at 2.5 RR BP1 ribosome binding protein 1 20p12 homolog 180 kDa (dog) (RRBP1), mRNA. 250 S at O.OOO1573 3.1 solute carrier family 2 1p35-p31.3 (facilitated glucose transporter), member 1 (SLC2A1), mRNA. 261 X at O.OOO3831 4.3 BGN biglycan (BGN), mRNA. 302 at 1.57E-OS -3.1 ANXA4 annexin A4 (ANXA4), mRNA. 370 s at O.OOO2857 -2.3 CUL3 cullin 3 (CUL3), mRNA. 2a36.3 376 s at O.OOO1535 -2.4 HNRPF heterogeneous nuclear 10q11.21-q11.22 ribonucleoprotein F (HNRPF), mRNA. 408 at 3.9SE-OS -3.2 protein phosphatase 1, catalytic 2p23 subunit, beta isoform (PPP1CB), transcript variant 2, mRNA. 425 at 18OE-06 -2.9 ALDH2 aldehyde dehydrogenase 2 12q24.2 family (mitochondrial) (ALDH2), nuclear gene encoding mitochondrial protein, mRNA. 427 s at O.OOO4665 -2.5 SEPP1 Selenoprotein P. plasma, 1 (SEPP1), mRNA. 431 S at O.OOO3957 DPYSL3 dihydropyrimidinase-like 3 (DPYSL3), mRNA. 432 at 2.09E-OS -3.3 CAT catalase (CAT), mRNA. 11 p.13 455 s at O.OOO4393 -2.7 NPEPPS aminopeptidase puromycin 17q21 sensitive (NPEPPS), mRNA. 482 at O.OOO4221 -2.9 quiescin Q6 (QSCN6), transcript variant 2 mRNA. 484 at O.OOO1619 -2.2 SUPT4H1 Suppressor of Ty 4 homolog 1 (S. cerevisiae) (SUPT4H1), mRNA. 487 at O.OOO2325 3.1 CTSC cathepsin C (CTSC), transcript 11 q14.1-q14.3 variant 1 mRNA. 496 x at O.OOOS336 -6.1 MYH11 myosin, heavy polypeptide 11, 16p13.13-p13.12 Smooth muscle (MYH11), transcript variant SM2, mRNA. 506 at O.OOO3278 3.3 TGFBI transforming growth factor, beta-induced, 68 kDa (TGFBI), mRNA. US 8,440,393 B2 65 66 TABLE 1-continued Gene expression profile. Affymetrix Parametric Fold Change Probe Set P-Value (Tumor? Normal) Gene Symbol Description Map 201529 S. at 0.0001633 -3 RPA1 replication protein A1, 70 kDa 17p13.3 (RPA1), mRNA. 201535 at 6.72E-OS -2.3 UBL3 ubiquitin-like 3 (UBL3), 13q12-q13 mRNA. 201554 X at 0.0003257 -2.5 GYG glycogenin (GYG), mRNA. 3q24-q25.1 201579 at O.OOO7798 2.6 FAT FAT tumor Suppressor homolog 4q34-q35 1 (Drosophila) (FAT), mRNA. 201581 at 2.11E-05 -2.4 DJ971N18.2 Hypothetical protein 20p12 DJ971N18.2 201584 S at 0.0001736 2.3 DDX39 DEAD (Asp-Glu-Ala-Asp) box 19p 13.12 polypeptide 39 (DDX39), transcript variant 1, mRNA. 201596 X at 1.62E-05 4.3 KRT18 keratin 18 (KRT18), transcript 12q13 variant 2 mRNA. 201600 at 3.48E-05 -2.3 PHB2 prohibitin 2 (PHB2), mRNA. 12p13 201666 at O.OOO4812 3.4 TIMP1 TIMP metallopeptidase inhibitor Xp11.3-p11.23 1 (TIMP1), mRNA. 201674 S at 0.0001795 -3 AKAP1 A kinase (PRKA) anchor protein 17q21-q23 1 (AKAP1), nuclear gene encoding mitochondrial protein, transcript variant 1, mRNA. 201696 at O.OOO121 -2 SFRS4 splicing factor, arginine?serine- 1p35.3 rich 4 (SFRS4), mRNA. 201697 s at 0.0001.256 2.4 DNMT1 DNA (cytosine-5-)- 19p 13.2 methyltransferase 1 (DNMT1), mRNA. 201712 S at 0.0008422 -2.2 RANBP2 RAN binding protein 2 2d 12.3 (RANBP2), mRNA. 201737 S at 5.10E-06 -3.1 6-Mar membrane-associated ring finger 5p15.2 (C3HC4) 6 (MARCH6), mRNA. 201756 at O.OOO3S33 -2 RPA2 replication protein A2, 32 kDa 1p35 (RPA2), mRNA. 201810 S. at 0.0001904 -3.2 SH3BP5 SH3-domain binding protein 5 3p24.3 (BTK-associated) (SH3BP5), transcript variant 2, mRNA. 201816 s at 0.0004168 -2.2 GBAS glioblastoma amplified sequence 7p12 (GBAS), mRNA. 201871 s at 0.0001348 -2 LOCS1035 ORF (LOC51035), mRNA. 11 q12.3 201891 s at 0.0007795 2 B2M beta-2-microglobulin (B2M), 15q21-q22.2 mRNA. 201893 x at 0.0006272 -3.1 DCN decorin (DCN), transcript 12q21.33 variant B, mRNA. 201911 s at 0.0001467 3.4 FARP1 FERM, RhoGEF (ARHGEF) 13q32.2 and pleckstrin domain protein 1 (chondrocyte-derived) (FARP1), transcript variant 1, mRNA. 201922 at O.OOO1538 -2 TINP1 TGFbeta-inducible nuclear 5q13.3 protein 1 (TINP1), mRNA. 201960 S at 0.0002203 -2 MYCBP2 MYC binding protein 2 13q22 (MYCBP2), mRNA. 201973 s at 0.0004504 2 C7orf28A chromosome 7 open reading 7p22.1 rame 28A (C7orf28A), mRNA. 202016 at O.OOO1808 3.2 MEST mesoderm specific transcript 7q32 homolog (mouse) (MEST), transcript variant 3, mRNA. 202028 s at 0.0002331 3 RPL38 ribosomal protein L38 (RPL38), 17q23-q25 mRNA. 202029 X at 1.93E-05 -2.1 RPL38 ribosomal protein L38 (RPL38), 17q23-q25 mRNA. 202037 S at 0.0002558 -3.1 SFRP1 secreted frizzled-related protein 8p12-p11.1 1 (SFRP1), mRNA. 202068 s at 0.0001171 -4.1 LDLR low density lipoprotein receptor 19p13.3 (familial hypercholesterolemia) (LDLR), mRNA. 202073 at 4.49E-05 -2.6 OPTN optineurin (OPTN), transcript 10p13 variant 2 mRNA. 202105 at 3.OOE-07 -3.2 IGBP1 immunoglobulin (CD79A) Xq13.1-q13.3 binding protein 1 (IGBP1), mRNA. 202119 S at 0.0002779 -3.2 CPNE3 copine III (CPNE3), mRNA. 8q21.3 202139 at O.OOO4872 -2.1 AKR7A2 aldo-keto reductase family 7, 1p35.1-p36.23 member A2 (aflatoxin aldehyde reductase) (AKR7A2), mRNA. US 8,440,393 B2 67 68 TABLE 1-continued Gene expression profile. Affymetrix Parametric Fold Change Probe Set P-Value (Tumor? Normal) Gene Symbol Description Map 202148 S. at 0.0006205 2.2 PYCR1 pyrroline-5-carboxylate 17q25.3 reductase 1 (PYCR1), transcript variant 2 mRNA. 202156 S at 0.0004765 -2.8 CUGBP2 CUG triplet repeat, RNA 10p13 binding protein 2 (CUGBP2), transcript variant 2, mRNA. 202157 s at 9.00E-07 -6.1 CUGBP2 CUG triplet repeat, RNA 10p13 binding protein 2 (CUGBP2), transcript variant 2, mRNA. 20215.8 s at 2.99E-05 -4.2 CUGBP2 CUG triplet repeat, RNA 10p13 binding protein 2 (CUGBP2), transcript variant 2, mRNA. 202172 at O.OOO6154 -2.4 ZNF161 Zinc finger protein 161 17q23.2 (ZNF161), mRNA. 2022.02 s at 0.0002996 3.4 LAMA4 aminin, alpha 4 (LAMA4), 6q21 mRNA. 202214 S at 0.0005849 -2 CUL4B Cullin 4B Xq23 202232 S at 0.0002955 -2.2 lf-BS dendritic cell protein (hfl-B5), 11 p.13 mRNA. 202259 S at 3.74E-05 -3.2 PFAAP5 phosphonoformate immuno- 13q12-q13 associated protein 5 (PFAAP5), mRNA. 202260 s at 0.0005909 -2.2 STXBP1 Syntaxin binding protein 1 9q34.1 (STXBP1), transcript variant 2, mRNA. 202286 s at 0.0003361 5.5 TACSTD2 tumor-associated calcium signal 1p32-p31 transducer 2 (TACSTD2), mRNA. 202292 X at 0.0001802 2.4 LYPLA2 lysophospholipase II (LYPLA2), 1p36.12-p35.1 mRNA. 202297 s at 6.25E-05 2.2 RER1 RER1 retention in endoplasmic 1pter-q24 reticulum 1 homolog (S. cerevisiae) (RER1), mRNA. 202314 at O.OOO1115 -2.6 CYP51A1 cytochrome P450, family 51, 7q21.2-q21.3 Subfamily A, polypeptide 1 (CYP51A1), mRNA. 202350 S at 0.0002121 -3.7 MATN2 matrilin 2 (MATN2), transcript 8q22 variant 2 mRNA. 202364 at O.OOO2246 -2.5 MXI1 MAX interactor 1 (MXI1), 10q24-q25 transcript variant 3, mRNA. 202378 s at 6.45E-05 -2.2 LEPROT eptin receptor overlapping 1p31.2 transcript (LEPROT), mRNA. 202404 S at 0.000452 2.9 COL1A2 collagen, type I, alpha 2 7q22.1 (COL1A2), mRNA. 202429 S. at 0.0001.255 -2.4 PPP3CA protein phosphatase 3 (formerly 4q21-q24 2B), catalytic Subunit, alpha isoform (calcineurin Aalpha) (PPP3CA), mRNA. 202464 S at 0.00098.23 3.5 PFKFB3 6-phosphofructo-2- 10p14-p15 kinase? fructose-2,6- biphosphatase 3 2O2465 at 9.8OE-06 4.8 PCOLCE procollagen C-endopeptidase 7q22 enhancer (PCOLCE), mRNA. 202468 S. at 7.84E-05 -2.5 CTNNAL1 catenin (cadherin-associated 9q31.2 protein), alpha-like 1 (CTNNAL1), mRNA. 2025.02 at S.O8E-OS -2 ACADM acyl-Coenzyme A 1p31 dehydrogenase, C-4 to C-12 straight chain (ACADM), nuclear gene encoding mitochondrial protein, mRNA. 20251.0 s at 1.14E-05 3.7 TNFAIP2 tumor necrosis factor, alpha- 14q32 induced protein 2 (TNFAIP2), mRNA. 202512 s at 6.24E-05 -2.2 APGSL APG5 autophagy 5-like (S. cerevisiae) 6q21 (APGSL), mRNA. 202536 at O.OOO2234 -2 CHMP2B chromatin modifying protein 2B 3p12.1 (CHMP2B), mRNA. 202546 at O.OOO4572 S.6 WAMP8 vesicle-associated membrane 2p12-p11.2 protein 8 (endobrevin) (VAMP8), mRNA. 202547 s at 6.98E-05 3 ARHGEF7 Rho guanine nucleotide 13q34 exchange factor (GEF) 7 (ARHGEF7), transcript variant 1, mRNA. US 8,440,393 B2 69 70 TABLE 1-continued Gene expression profile. Affymetrix Parametric Fold Change Probe Set P-Value (Tumor? Normal) Gene Symbol Description Map 202573 at O.OOO2474 2 CSNK1 G2 casein kinase 1, gamma 2 19p13.3 (CSNK1G2), mRNA. 202581 at O.OOO2675 2.8 HSPA1B heat shock 70 kDa protein 1B 6p21.3 (HSPA1B), mRNA. 202630 at 6.64E-OS -2.7 APPBP2 amyloid beta precursor protein 17q21-q23 (cytoplasmic tail) binding protein 2 (APPBP2), mRNA. 202665 S at 0.0003645 2.9 WASPIP Wiskott-Aldrich syndrome 2a31.1 protein interacting protein (WASPIP), mRNA. 20272.2 s at 0.0002949 2.1 GFPT1 glutamine-fructose-6-phosphate 2p13 transaminase 1 (GEPT1), mRNA. 202723 s at 2.20E-06 -4 FOXO1A orkhead box O1A 13q14.1 (rhabdomyosarcoma) (FOXO1A), mRNA. 202724 S at 6.00E-07 -3.3 FOXO1A orkhead box O1A 13q14.1 (rhabdomyosarcoma) (FOXO1A), mRNA. 202731 at O.OOO8149 -2.6 PDCD4 programmed cell death 4 10q24 (neoplastic transformation inhibitor) (PDCD4), transcript variant 1 mRNA. 202733 at 0.0005751 3 P4HA2 procollagen-proline, 2- 5q31 oxoglutarate 4-dioxygenase (proline 4-hydroxylase), alpha polypeptide II (P4HA2), transcript variant 3, mRNA. 202746 at O.OOOS864 -6.4 ITM2A integral membrane protein 2A Xq13.3-Xq21.2 (ITM2A), mRNA. 202749 at O.OOO9431 -2.2 WRB Tryptophan rich basic protein 21q22.3 202761 S. at 0.0003908 -2.3 SYNE2 spectrin repeat containing, 14q23.2 nuclear envelope 2 (SYNE2), transcript variant 4, mRNA. 202780 at O.OOO9309 -2 OXCT1 3-oxoacid CoA transferase 1 5p13.1 (OXCT1), nuclear gene encoding mitochondrial protein, mRNA. 202820 at 1.90E-OS 3.3 AHR aryl hydrocarbon receptor 7p15 (AHR), mRNA. 202825 at O.OOO4002 -2.1 SLC25A4 solute carrier family 25 4q35 (mitochondrial carrier; adenine nucleotide translocator), member 4 (SLC25A4), nuclear gene encoding mitochondrial protein, mRNA. 202888. S. at 0.0005702 4.7 ANPEP alanyl (membrane) 15q25-q26 aminopeptidase (aminopeptidase N, aminopeptidase M, microsomal aminopeptidase, CD13, p15O) (ANPEP), mRNA. 202899 s at 5.55E-05 -2.8 SFRS3 splicing factor, arginine?serine- 6p21 rich 3 (SFRS3), mRNA. 202908 at S.90E-06 -3.4 WFS1 Wolfram syndrome 1 4p16 (wolframin) (WFS1), mRNA. 202911 at O.OOO8837 -2.1 MSH6 mutS homolog 6 (E. coli) 2p16 (MSH6), mRNA. 202920 at 4.6OE-06 -4.2 ANK2 ankyrin 2, neuronal (ANK2), 4q25-q27 transcript variant 2, mRNA. 202952 S at 2.37E-05 7.6 ADAM12 ADAM metallopeptidase 10q26.3 domain 12 (meltrin alpha) (ADAM12), transcript variant 1, mRNA. 202954 at O.OOO6434 2.8 UBE2C ubiquitin-conjugating enzyme 20d 13.12 E2C (UBE2C), transcript variant 1, mRNA. 202957 at O.OOO374 2.7 HCLS1 hematopoietic cell-specific Lyn 3q13 Substrate 1 (HCLS1), mRNA. 202968 s at 0.0001551 2.3 DYRK2 dual-specificity tyrosine-(Y)- 12q15 phosphorylation regulated kinase 2 (DYRK2), transcript variant 1 mRNA. 202975 s at 2.87E-05 -3.6 RHOBTB3 Rho-related BTB domain 5q15 containing 3 (RHOBTB3), mRNA. US 8,440,393 B2 71 72 TABLE 1-continued Gene expression profile. Affymetrix Parametric Fold Change Probe Set P-Value (Tumor? Normal) Gene Symbol Description Map 202992 at O.OOOS 19 -5.4 C7 complement component 7 (C7), 5p13 mRNA. 202998 s at 0.0006468 3 LOXL2 lysyl oxidase-like 2 (LOXL2), 8p21.3-p21.2 mRNA. 203088 at O.OOO8436 -4.5 FBLNS fibulin 5 (FBLN5), mRNA. 14q32.1 203156 at 3.37E-OS -2.4 AKAP11 A kinase (PRKA) anchor protein 13q14.11 11 (AKAP11), transcript variant 1, mRNA. 203166 at O.OOO1939 -2.3 CFDP1 craniofacial development protein 16q22.2-q22.3 1 (CFDP1), mRNA. 203178 at O.OOO2428 -3.3 GATM glycine amidinotransferase (L- 15q21.1 arginine:glycine amidinotransferase) (GATM), mRNA. 203249 at O.OOO4O75 -2.5 EZH1 enhancer of Zeste homolog 1 17q21.1-q21.3 (Drosophila) (EZH1), mRNA. 203297 s at 4.33E-05 2.4 ARID2 Jumonji, AT rich interactive 6p24-p23 domain 2 (JARID2), mRNA. 203298 s at 0.0005468 2.1 ARID2 Jumonji, AT rich interactive 6p24-p23 domain 2 (JARID2), mRNA. 203349 s at 7.26E-05 2.8 ETV5 ets variant gene 5 (ets-related 3q28 molecule) (ETV5), mRNA. 203356 at 7.77E-05 -2.6 CAPN7 calpain 7 (CAPN7), mRNA. 3p24 203358. S. at 2.44E-05 2.9 EZH2 enhancer of Zeste homolog 2 7q35-q36 (Drosophila) (EZH2), transcript variant 2 mRNA. 203401 at O.OOO3991 -3.8 PRPS2 phosphoribosyl pyrophosphate Xp22.3-p22.2 synthetase 2 (PRPS2), mRNA. 203423 at O.OOO8061 -3.9 RBP1 retinol binding protein 1, 3q23 cellular (RBP1), mRNA. 203424 S at 0.0007037 -3.8 GFBP5 insulin-like growth factor 2a33-q36 binding protein 5 (IGFBP5), mRNA. 203427 at 7. OSE-OS -2.6 ASF1A ASF1 anti-silencing function 1 6q22.31 homolog A (S. cerevisiae) (ASF1A), mRNA. 203450 at O.OOO7322 -2.1 PGEA1 PKD2 interactor, golgi and 22q12 endoplasmic reticulum associated 1 (PGEA1), transcript variant 1 mRNA. 203455 S at 0.0005583 2 SAT spermidine?spermine N1- Xp22.1 acetyltransferase (SAT), mRNA. 203459 s at 8.97E-05 2.1 VPS16 vacuolar protein sorting 16 20p13-p12 (yeast) (VPS16), transcript variant 2 mRNA. 203468 at O.OOO9911 2.3 CDK10 cyclin-dependent kinase (CDC2- 16q24 like) 10 (CDK10), transcript variant 2 mRNA. 203476 at 1.6SE-OS 2.6 TPBG trophoblast glycoprotein 6q14-q15 (TPBG), mRNA. 203493 s at 0.0009.017 -2.1 PIG8 translokin (PIG8), mRNA. 11q21 203494 S at 0.0001.224 -2.2 PIG8 translokin (PIG8), mRNA. 11q21 203505 at O.OOO1234 2.7 ABCA1 ATP-binding cassette, Sub- 9q31.1 family A (ABC1), member 1 203549 s at 0.0001406 3.9 LPL lipoprotein lipase (LPL), 8p22 mRNA. 203599 s at 0.0004137 -2 WBP4. WW domain binding protein 4 13q14.11 (formin binding protein 21) (WBP4), mRNA. 203640 at O.OOO2347 -2 MBNL2 muscleblind-like 2 (Drosophila) 13q32.1 (MBNL2), transcript variant 3, mRNA. 203657 s at 0.0007798 -2.3 CTSF cathepsin F (CTSF), mRNA. 11 q13 203680 at 9.OOE-07 -6.1 PRKAR2B protein kinase, cAMP- 7q22 dependent, regulatory, type II, beta (PRKAR2B), mRNA. 2036.92 s at 0.0005974 2.2 E2F3 E2F transcription factor 3 6p22 (E2F3), mRNA. 203695 s at 0.0009082 -2.2 DFNAS deafness, autosomal dominant 5 7p15 (DFNA5), mRNA. 203758 at O.OOO2672 -2 CTSO cathepsin O (CTSO), mRNA. 4q31-q32 203762 s at 0.0003113 -2.1 D2LIC dynein 2 light intermediate chain 2p25.1-p24.1 (D2LIC), transcript variant 1, mRNA. US 8,440,393 B2 73 74 TABLE 1-continued Gene expression profile. Affymetrix Parametric Fold Change Probe Set P-Value (Tumor? Normal) Gene Symbol Description Map 203799 at 1.06E-05 -3.3 CD302 CD302 antigen (CD302), 2d24.2 mRNA. 203803 at 3.79E-05 -3.9 PCYOX1 prenylcysteine oxidase 1 2p13.3 (PCYOX1), mRNA. 203845 at O.OOO1562 -2.8 PCAF p300, CBP-associated factor 3p24 (PCAF), mRNA. 203878 s at 2.11E-05 3.7 MMP11 matrix metallopeptidase 11 22d 11.23 (stromelysin 3) (MMP11), mRNA. 203888 at O.OOO8536 THEBD hrombomodulin (THBD), 20p12-cen mRNA. 203897 at 7.35E-05 -2.2 LOCS7149 hypothetical protein A-211C6.1 16p11.2 (LOC57149), mRNA. 203908 at 2.7 Solute carrier family 4, Sodium bicarbonate cotransporter, member 4 (SLC4A4), mRNA. 203936 s at 9.4 MMP9 matrix metallopeptidase 9 20d 11.2-q13.1 (gelatinase B, 92 kDa gelatinase, 92 kDa type IV collagenase) (MMP9), mRNA. 204004 at 1SOE-06 -3.2 PAWR PRKC, apoptosis, WT1, regulator 204029 at O.OOOS472 2.4 CELSR2 cadherin, EGF LAG seven-pass G-type receptor 2 (flamingo homolog, Drosophila) (CELSR2), mRNA. 204041 at -6.4 MAOB Monoamine oxidase B Xp11.23 204045 at -4.5 TCEAL1 Transcription elongation factor Xq22.1 A (SII)-like 1 204078 at O.OOO1645 2.6 SC65 synaptonemal complex protein 17q21.2 SC65 (SC65), mRNA. 204082 at O.OOO2318 -2.8 PBX3 pre-B-cell leukemia transcription factor 3 (PBX3), mRNA. 204109 s at O.OOO6487 2.5 NFYA nuclear transcription factor Y, 6p21.3 alpha (NFYA), transcript variant 2, mRNA. 204136 at 7.18E-OS 3.4 COL7A1 collagen, type VII, alpha 1 3p21.1 (epidermolysis bullosa, dystrophic, dominant and recessive) (COL7A1), mRNA. 20418.4 S at 9.26E-OS ADRBK2 adrenergic, beta, receptor kinase 22q12.1 2 (ADRBK2), mRNA. 204235 s at 2.4SE-OS GULP GULP, engulfment adaptor PTB domain containing 1 (GULP1), mRNA. 204237 at 1.32E-OS -3.2 GULP GULP, engulfment adaptor PTB domain containing 1 (GULP1), mRNA. 204256 at O.OOO8289 2.1 ELOVL6 ELOVL family member 6, elongation of long chain fatty acids (FEN1/Elo2, SUR4/Elo3 like, yeast) (ELOVL6), mRNA. 204285 S at 1.2OE-06 7.7 PMAIP1 phorbol-12-myristate-13 18q21.32 acetate-induced protein 1 (PMAIP1), mRNA. 204286 s at 3. SOE-06 8.5 PMAIP1 phorbol-12-myristate-13 18q21.32 acetate-induced protein 1 (PMAIP1), mRNA. 20438.7 x at 7.88E-OS 2.6 MRP63 mitochondrial ribosomal protein 13p11.1-q11 63 (MRP63), nuclear gene encoding mitochondrial protein, mRNA. 204454 at -2.6 LDOC1 leucine Zipper, down-regulated in cancer 1 (LDOC1), mRNA. 204473 s at O.OOO9569 ZNF592 Zinc finger protein 592 15q25.3 (ZNF592), mRNA. 204493 at 2.7 BID BH3 interacting domain death 22q11.1 agonist (BID), transcript variant 3, mRNA. 204495 s at O.OOO6319 2.8 DKFZP434H132 DKFZP434H132 protein 204531 s at O.OOO4972 2.5 BRCA1 breast cancer 1, early onset (BRCA1), transcript variant BRCA1-delta 9-11, mRNA. US 8,440,393 B2 75 76 TABLE 1-continued Gene expression profile. Affymetrix Parametric Fold Change Probe Set P-Value (Tumor? Normal) Gene Symbol Description Map 204595 S at 1.40E-06 1O.S STC1 stanniocalcin 1 (STC1), mRNA. 8p21-p11.2 204597 X at 0.0001207 7.5 STC1 stanniocalcin 1 (STC1), mRNA. 8p21-p11.2 204619 s at 0.0005599 4.6 CSPG2 Chondroitin sulfate 5q14.3 proteoglycan 2 (versican) 204639 at 1.47E-OS 4.2 ADA adenosine deaminase (ADA), 20d 12-q13.11 mRNA. 204641 at O.OOO6499 4.6 NEK2 NIMA (never in mitosis genea)- 1q32.2-q241 related kinase 2 (NEK2), mRNA. 204669 S at 0.0008759 2 RNF24 ring finger protein 24 (RNF24), 20p13-p12.1 mRNA. 204731 at O.OOO2926 -3.2 TGFBR3 transforming growth factor, beta 1p33-p32 receptor III (betaglycan, 300 kDa) (TGFBR3), mRNA. 204735 at 3.38E-OS 3.5 PDE4A phosphodiesterase 4A, cAMP- 19p 13.2 specific (phosphodiesterase E2 dunce homolog, Drosophila) (PDE4A), mRNA. 204749 at O.OOOS462 -3 NAP1L3 nucleosome assembly protein 1- Xq21.3-q22 ike 3 (NAP1L3), mRNA. 204786 s at 1.59E-05 2.8 IFNAR2 interferon (alpha, beta and 21q22.11 omega) receptor 2 (IFNAR2), transcript variant 1, mRNA. 204793 at 1.31E-OS -4.2 GPRASP1 G protein-coupled receptor Xq22.1 associated sorting protein 1 (GPRASP1), mRNA. 204939 S at 0.0007407 -5.7 PLN phospholamban (PLN), mRNA. 6q22.1 204994 at 4.91E-OS 4.5 MX2 myxovirus (influenza virus) 21q22.3 resistance 2 (mouse) (MX2), mRNA. 205068. S. at 2.40E-06 3.8 ARHGAP26 Rho GTPase activating protein 5q31 26 (ARHGAP26), mRNA. 205079 s at 0.0003985 -2 MPDZ multiple PDZ domain protein 9p24-p22 (MPDZ), mRNA. 205226 at O.OOO4657 -3 PDGFRL platelet-derived growth factor 8p22-p21.3 receptor-like (PDGFRL), mRNA. 205231 s at 0.0008.847 -2 EPM2A epilepsy, progressive myoclonus 6q24 type 2A, Lafora disease (laforin) (EPM2A), transcript variant 1, mRNA. 205241 at 4.2OE-05 4.5 SCO2 SCO cytochrome oxidase 22d 13.33 deficient homolog 2 (yeast) (SCO2), nuclear gene encoding mitochondrial protein, mRNA. 205259 at O.OOO2O39 -2.5 NR3C2 nuclear receptor subfamily 3, 4q31.1 group C, member 2 (NR3C2), mRNA. 205269 at 4.OOE-06 8.7 LCP2 ymphocyte cytosolic protein 2 5q33.1-qter (SH2 domain containing eukocyte protein of 76 kDa) (LCP2), mRNA. 205270 s at 0.0006743 6.7 LCP2 ymphocyte cytosolic protein 2 5q33.1-qter (SH2 domain containing eukocyte protein of 76 kDa) (LCP2), mRNA. 205304 S at 0.0007499 2.4 KCN8 potassium inwardly-rectifying 12p11.23 channel, Subfamily J, member 8 (KCNJ8), mRNA. 205353 S at 3.92E-05 -3.3 PBP prostatic binding protein (PBP), 12q24.23 mRNA. 205367 at O.OOO1487 2.7 APS adaptor protein with pleckstrin 7q22 homology and Src homology 2 domains (APS), mRNA. 205370 X at 0.0004381 2.9 DBT dihydrolipoamide branched 1p31 chain transacylase E2 (DBT), mRNA. 205381 at O.OOO1503 -5.4 LRRC17 leucine rich repeat containing 17 7q22.1 (LRRC17), transcript variant 1, mRNA. 2054.06 s at 3.91E-05 2.5 SPA17 sperm autoantigenic protein 17 11q24.2 (SPA17), mRNA. US 8,440,393 B2 77 78 TABLE 1-continued Gene expression profile. Affymetrix Parametric Fold Change Probe Set P-Value (Tumor? Normal) Gene Symbol Description Map 205412 at 6.3OE-06 -2.7 ACAT1 acetyl-Coenzyme A 11q22.3-q23.1 acetyltransferase 1 (acetoacetyl Coenzyme A thiolase) (ACAT1), nuclear gene encoding mitochondrial protein, mRNA. 205463 s at 0.000255 3.1 PDGFA Platelet-derived growth factor 7p22 alpha polypeptide 205466 s at 6.17E-05 -4.7 HS3ST1 heparan Sulfate (glucosamine) 3- 4p16 O-sulfotransferase 1 (HS3ST1), mRNA. 205479 s at 0.000306 3.8 PLAU plasminogen activator, 10q24 urokinase (PLAU), mRNA. 205483 s at 0.0001.248 6.9 G1P2 interferon, alpha-inducible 1p36.33 protein (clone IFI-15K) (G1P2), mRNA. 205532 S at 0.0004946 2.8 CDH6 cadherin 6, type 2, K-cadherin 5p15.1-p14 (fetal kidney) (CDH6), mRNA. 205572 at O.OOO3915 5.4 ANGPT2 angiopoietin 2 (ANGPT2), 8p23.1 mRNA. 205687 at O.OOO463 2 UBPH similar to ubiquitin binding 16p12 protein (UBPH), mRNA. 205771 s at 0.0009058 -2.3 AKAP7 A kinase (PRKA) anchor protein 6q23 7 (AKAP7), transcript variant alpha, mRNA. 205812 S at 9.00E-06 2.4 TMED9 transmembrane emp24 protein 5q.35.3 transport domain containing 9 (TMED9), mRNA. 205849 s at 0.0005492 -2 UQCRB ubiquinol-cytochrome c 8q22 reductase binding protein (UQCRB), mRNA. 205862 at 7.6SE-OS -8.6 GREB1 GREB1 protein (GREB1), 2p25.1 transcript varianta, mRNA. 205871 at O.OOO3369 3.1 PLGLB1 plasminogen-like B1 (PLGLB1), 2 mRNA. 205943 at O.OOO4032 4.4 TDO2 tryptophan 2,3-dioxygenase 4q31-q32 (TDO2), mRNA. 205961 s at 0.0002574 -2.9 PSIP1 PC4 and SFRS1 interacting 9p22.3 protein 1 (PSIP1), transcript variant 2 mRNA. 206026 s at p < 1e–07 29.1 TNFAIP6 tumor necrosis factor, alpha- 2d23.3 induced protein 6 (TNFAIP6), mRNA. 20615.8 s at 0.0006831 -2.2 ZNF9 Zinc finger protein 9 (a cellular 3q21 retroviral nucleic acid binding protein) (ZNF9), mRNA. 206169 X at 0.00084.28 3.9 ROXaN Zinc finger CCCH-type 22d 13.2 containing 7B 206211 at O.OOO2O57 -8.9 SELE selectin E (endothelial adhesion 1q22-q25 molecule 1) (SELE), mRNA. 206247 at 9.38E-OS 3.3 MICB MHC class I polypeptide-related 6p21.3 sequence B (MICB), mRNA. 206359 at O.OOO4232 -3.5 SOCS3 Suppressor of cytokine signaling 3 17q25.3 206377 at 2.4OE-06 6 FOXF2 forkhead box F2 (FOXF2), 6p25.3 mRNA. 206435 at O.OOO9321 3.2 GALGT UDP-N-acetyl-alpha-D- 12q13.3 galactosamine:(N- acetylneuraminyl)- galactosylglucosylceramide N acetylgalactosaminyltransferase (GalNAc-T) (GALGT), mRNA. 206483 at O.OOO9412 2 LRRC6 leucine rich repeat containing 6 8q24.22 (LRRC6), mRNA. 206,551 X at 0.0001129 2.7 DRE1 DRE1 protein 3q27.1 206571 s at 0.000131 2.7 MAP4K4 mitogen-activated protein kinase 2d 11.2-q12 kinase kinase kinase 4 (MAP4K4), transcript variant 1, mRNA. 206621 s at 0.0002653 -2.3 WBSCR1 Williams-Beuren syndrome 7q11.23 chromosome region 1 (WBSCR1), transcript variant 2, mRNA. 206637 at 3.S2E-OS 4.8 P2RY14 purinergic receptor P2Y, G- 3q21-q25 protein coupled, 14 (P2RY14), mRNA. US 8,440,393 B2 79 80 TABLE 1-continued Gene expression profile. Affymetrix Parametric Fold Change Probe Set P-Value (Tumor? Normal) Gene Symbol Description Map 206792 x at 6.82E-05 3.7 PDE4C phosphodiesterase 4C, cAMP- 19p 13.11 specific (phosphodiesterase E1 dunce homolog, Drosophila) (PDE4C), mRNA. 206.809 S at 0.0004239 -2.8 HNRPA3P1 Heterogeneous nuclear 10q11.21 ribonucleoprotein A3 pseudogene 1 206.857 s at 0.0002413 2.5 FKBP1B FK506 binding protein 1B, 12.6 kDa 2p23.3 (FKBP1 B), transcript variant 1 mRNA. 206874 s at 9.71E-05 -2.9 SLK STE20-like kinase (yeast) 10q25.1 206927 s at 0.0004319 3.3 GUCY1A2 guanylate cyclase 1, Soluble, 11q21-q22 alpha 2 (GUCY1A2), mRNA. 207040 S at 1.56E-05 -2.6 ST13 Suppression of tumorigenicity 13 22d 13.2 (colon carcinoma) (Hsp70 interacting protein) (ST13), mRNA. 207132 X at 0.0002292 -2.5 PFDN5 prefoldin 5 (PFDN5), transcript 12q12 variant 1 mRNA. 207147 at O.OOO7438 3.3 DLX2 distal-less homeo box 2 (DLX2), 2a32 mRNA. 207170 S. at 1.76E-05 -2.6 LETMD1 LETM1 domain containing 1 12q13.12 (LETMD1), transcript variant 3, mRNA. 207239 s at 0.0002121 2.2 PCTK1 PCTAIRE protein kinase 1 Xp11.3-p11.23 (PCTK1), transcript variant 1, mRNA. 207365 x at 6.06E-05 3.7 USP34 Ubiquitin specific protease 34 2p15 207386 at O.OOO9685 2.7 CYP7B1 cytochrome P450, family 7, 8q21.3 Subfamily B, polypeptide 1 (CYP7B1), mRNA. 2O7598 x at 0.0001463 3.5 XRCC2 X-ray repair complementing 7q36.1 defective repair in Chinese hamster cells 2 (XRCC2), mRNA. 2O7688. S. at 0.0006478 2.4 LOC387933 PREDICTED: Similar to 13 heterogeneous nuclear ribonucleoprotein A3 (LOC387933), mRNA. 2O7730 x at 0.0002746 3.1 FL2O700 hypothetical protein FLJ20700 19p13.3 207761 S. at 1.38E-05 -4.2 DKFZP586A0522 DKFZP586 AO522 protein 12q13.12 (DKFZP586A0522), mRNA. 207961 x at O.OOO6679 -3.8 MYH11 myosin, heavy polypeptide 11, 16p13.13-p13.12 Smooth muscle (MYH11), transcript variant SM1 mRNA. 207974 s at 0.0001511 -2.1 SKP1A S-phase kinase-associated 5q31 protein 1A (p19A) (SKP1A), transcript variant 2, mRNA. 207983 s at 0.0009431 -2.3 STAG2 stromal antigen 2 (STAG2), Xq25 mRNA. 208092 S at 0.00041 78 2.9 FAM49A amily with sequence similarity 2p24.3-p24.2 49, member A (FAM49A), mRNA. 208137 X at 0.0008157 2.6 ZNF611 Zinc finger protein 611 19q13.41 208.238 x at 0.0007.093 2.4 LZLP eucine Zipper-like protein 11 q13.1 208.246 x at 8.09E-05 3.6 FL2OOO6 hypothetical protein FLJ20006 16q23.1 208.248 x at 5.35E-05 -3.5 APLP2 amyloid beta (A4) precursor-like 11q24 protein 2 (APLP2), mRNA. 208.540 x at 0.0007507 2 S100A14 S100 calcium binding protein 7q22-q31.1 A14 (calgizZarin) 208.626 s at 0.000509 -2.1 WAT1 vesicle amine transport protein 1 17q21 homolog (T californica) (VAT1), mRNA. 208.631 s at 0.0003804 -2.3 HADHA hydroxyacyl-Coenzyme A 2p23 dehydrogenase 3-ketoacyl Coenzyme A thiolasetenoyl Coenzyme A hydratase (trifunctional protein), alpha subunit (HADHA), mRNA. 208635 x at 8.57E-05 -2 NACA nascent-polypeptide-associated 12q23-q24.1 complex alpha polypeptide (NACA), mRNA. US 8,440,393 B2 81 82 TABLE 1-continued Gene expression profile. Affymetrix Parametric Fold Change Probe Set P-Value (Tumor? Normal) Gene Symbol Description Map 208.643 s at 2.41E-05 -2.9 XRCCS X-ray repair complementing 2a35 defective repair in Chinese hamster cells 5 (double-strand break rejoining; Ku autoantigen, 80 kDa) (XRCC5), mRNA. 208647 at O.OOO4172 -2.1 FDFT1 farnesyl-diphosphate 8p23.1-p22 farnesyltransferase 1 (FDFT1), mRNA. 208653 S at 0.0002993 3 CD164 CD164 antigen, sialomucin 6q21 (CD164), mRNA. 208655 at O.OOO1893 -2.8 CCNI Cyclin I 4q21.1 208.658 at O.OOOS959 2 PDIA4 protein disulfide isomerase 7q35 family A, member 4 (PDIA4), mRNA. 208662 s at 0.0009682 -2.1 TTC3 tetratricopeptide repeat domain 21q22.2 3 (TTC3), transcript variant 2, mRNA. 208666 s at 5.00E-07 -4.5 ST13 Suppression of tumorigenicity 13 22d 13.2 (colon carcinoma) (Hsp70 interacting protein) (ST13), mRNA. 208.667 s at 4.53E-05 -2.8 ST13 Suppression of tumorigenicity 13 22d 13.2 (colon carcinoma) (Hsp70 interacting protein) (ST13), mRNA. 208671 at O.OOOS2OS -2 TDE2 tumor differentially expressed 2 6q22.31 (TDE2), mRNA. 208673 s at 0.0005001 -2.2 SFRS3 splicing factor, arginine?serine- 6p21 rich 3 (SFRS3), mRNA. 208697 s at 4.85E-05 -2.4 EIF3S6 eukaryotic translation initiation 8q22-q23 actor 3, subunit 6 48 kDa (EIF3S6), mRNA. 2087.03 s at 0.0002231 -3.7 APLP2 amyloid beta (A4) precursor-like 11q24 protein 2 (APLP2), mRNA. 208704 x at 4.90E-05 -3.6 APLP2 amyloid beta (A4) precursor-like 11q24 protein 2 (APLP2), mRNA. 208740 at 7.4OE-OS -2.2 SAP18. sin3-associated polypeptide, 13q12.11 18 kDa (SAP18), mRNA. 208760 at 7.17E-05 -2.7 UBE2I Ubiquitin-conjugating enzyme 16p13.3 E2I (UBC9 homolog, yeast) 20877O S at 0.0005681 -2 EIF4EBP2 eukaryotic translation initiation 10q21-q22 actor 4E binding protein 2 (EIF4EBP2), mRNA. 208771 S. at 2.10E-05 -2.6 LTA4H eukotriene A4 hydrolase 12q22 (LTA4H), mRNA. 208781 x at 0.0006007 -2.2 SNX3 sorting nexin 3 (SNX3), 6q21 transcript variant 1, mRNA. 208791 at O.OOO1028 -4.4 CLU clusterin (complement lysis 8p21-p12 inhibitor, SP-40.40, sulfated glycoprotein 2, testosterone repressed prostate message 2, apolipoprotein J) (CLU), transcript variant 1, mRNA. 208792 S at 0.0001175 -4.5 CLU clusterin (complement lysis 8p21-p12 inhibitor, SP-40.40, sulfated glycoprotein 2, testosterone repressed prostate message 2, apolipoprotein J) (CLU), transcript variant 1, mRNA. 208794 S at 1.47E-05 2.4 SMARCA4 SWI/SNF related, matrix 19p 13.2 associated, actin dependent regulator of chromatin, Subfamily a member 4 (SMARCA4), mRNA. 208796 S. at 0.0001639 -2.6 CCNG1 cyclin G1 (CCNG1), transcript 5q32-q34 variant 2 mRNA. 208848 at 1SOE-06 -3.7 ADHS alcohol dehydrogenase 5 (class 4q21-q25 III), chi polypeptide (ADH5), mRNA. 208860 S at 0.0001719 -2 ATRX alpha thalassemiamental Xq13.1-q21.1 retardation syndrome X-linked (RAD54 homolog, S. cerevisiae) (ATRX), transcript variant 2, mRNA. US 8,440,393 B2 83 84 TABLE 1-continued Gene expression profile. Affymetrix Parametric Fold Change Probe Set P-Value (Tumor? Normal) Gene Symbol Description 208873 s at -2.1 CSOrf18 open reading rame 18 (C5orf18), mRNA. 208920 at O.OOO3111 -2.3 SRI Sorcin (SRI), transcript variant 2, mRNA. 208925 at O.OOOS429 -2.3 C3orfA. chromosome 3 open reading rame 4 (C3orf4), mRNA. 208950 s at O.OOO4293 -2.3 ALDHAA1 aldehyde dehydrogenase 7 amily, member A1 (ALDH7A1), mRNA. 208951 at 4.2OE-05 -2.8 ALDHAA1 aldehyde dehydrogenase 7 amily, member A1 (ALDH7A1), mRNA. 208962 s at -2.4 FADS1 atty acid desaturase 1 (FADS1), 11 q12.2-q13.1 mRNA. 208990 s. at -2.2 HNRPH3 heterogeneous nuclear 10q22 ribonucleoprotein H3 (2H9) (HNRPH3), transcript variant 2H9A, mRNA. 209009 at 1SOE-06 -3 ESD esterase D formylglutathione 13q14.1-q14.2 hydrolase (ESD), mRNA. 209030 s at 3.07E-OS 2.6 IGSF4 immunoglobulin Superfamily, 11q23.2 member 4 (IGSF4), mRNA. 209034 at 3.2OE-06 -4.5 PNRC1 proline-rich nuclear receptor coactivator 1 (PNRC1), mRNA. 209068 at 1.OOE-06 -3.5 HNRPDL heterogeneous nuclear ribonucleoprotein D-like (HNRPDL), transcript variant 2, mRNA. 209081 s at 8.3OE-06 5 COL18A1 collagen, type XVIII, alpha 1 21q22.3 (COL18A1), transcript variant 2, mRNA. 209082 s at 4.04E-OS S.1 COL18A1 collagen, type XVIII, alpha 1 21q22.3 (COL18A1), transcript variant 2, mRNA. 209137 s at O.OOO7917 -2.1 USP10 ubiquitin specific peptidase 10 16q24.1 (USP10), mRNA. 209143 s at O.OOO9179 -2.1 CLNS1A chloride channel, nucleotide 11 q13.5-q14 sensitive, 1A (CLNS1A), mRNA. 209146 at O.OOO2317 -3.2 SC4MOL sterol-C4-methyl oxidase-like (SC4MOL), transcript variant 2, mRNA. 2091.69 at O.OOO3677 4.2 GPM6B glycoprotein M6B (GPM6B), Xp22.2 transcript variant 1, mRNA. 209170 s at O.OOO14 5.8 GPM6B glycoprotein M6B (GPM6B), Xp22.2 transcript variant 4, mRNA. 209243 s at 3.90E-06 -7.7 PEG3 paternally expressed 3 (PEG3), 19q13.4 mRNA. 2093.05 s at O.OOO1816 -3.7 GADD45B growth arrest and DNA-damage 19p13.3 inducible, beta (GADD45B), mRNA. 209337 at 4.01E-OS -3 PSIP1 PC4 and SFRS1 interacting 9p22.3 protein 1 (PSIP1), transcript variant 2 mRNA. 209357 at -3.3 CITED2 Cbp/p300-interacting transactivator, with Glu/Asp rich carboxy-terminal domain, 2 (CITED2), mRNA. 209360 s at 7.44E-05 4.8 RUNX1 runt-related transcription factor 21q22.3 (acute myeloid leukemia 1: aml1 oncogene) (RUNX1), transcript variant 1, mRNA. 209384 at -2 PROSC proline synthetase co-transcribed 8p11.2 homolog (bacterial) (PROSC), mRNA. 209385 S at 3.63E-OS -2.6 PROSC proline synthetase co-transcribed 8p11.2 homolog (bacterial) (PROSC), mRNA. 209447 at 3.84E-05 -2.5 SYNE1 spectrin repeat containing, nuclear envelope 1 (SYNE1), transcript variant alpha, mRNA. 209512 at 3.90E-06 -3.9 HSDL2 hydroxysteroid dehydrogenase ike 2 (HSDL2), mRNA. 209513 s at 1.94E-05 -3.9 HSDL2 hydroxysteroid dehydrogenase ike 2 (HSDL2), mRNA. US 8,440,393 B2 85 86 TABLE 1-continued Gene expression profile. Affymetrix Parametric Fold Change Probe Set P-Value (Tumor? Normal) Gene Symbol Description Map 209596 at 4.8OE-06 6.9 MXRAS matrix-remodelling associated 5 Xp22.33 (MXRA5), mRNA. 209605 at O.OOO4157 -2.4 TST thiosulfate sulfurtransferase 22d 13.1 (rhodanese) (TST), nuclear gene encoding mitochondrial protein, mRNA. 20961.2 s at 0.0001328 -4.5 ADH1B alcohol dehydrogenase IB (class 4q21-q23 ), beta polypeptide (ADH1B), mRNA. 209.613 s at 0.0003092 -5.8 ADH1B alcohol dehydrogenase IB (class 4q21-q23 ), beta polypeptide (ADH1B), mRNA. 209633 at O.OOO6246 -2.2 PPP2R3A protein phosphatase 2 (formerly 3q22.1 2A), regulatory subunit B", alpha (PPP2R3A), transcript variant 2 mRNA. 209657 s at 0.0003701 -2.1 HSF2 heat shock transcription factor 2 6q22.31 (HSF2), mRNA. 209685 s at 1.1OE-05 3.7 PRKCB1 protein kinase C, beta 1 16p11.2 (PRKCB1), transcript variant 2, mRNA. 209733 at 164E-05 -2.7 LOC286440 Hypothetical protein Xq22.3 LOC28644O 209737 at O.OOO39SS -2.2 MAGI2 membrane associated guanylate 7q21 kinase, WW and PDZ domain containing 2 (MAGI2), mRNA. 209875 S. at 9.40E-06 9.5 SPP1 Secreted phosphoprotein 1 4q21-q25 (Osteopontin, bonesialoprotein early T-lymphocyte activation ) (SPP1), mRNA. 2098.94 at O.OOO1625 -4.2 LEPR eptin receptor (LEPR), 1p31 transcript variant 2, mRNA. 209897 s at 0.0001202 2.5 SLIT2 slit homolog 2 (Drosophila) 4p15.2 (SLIT2), mRNA. 209969 s at 1.51E-05 3.7 STAT1 signal transducer and activator 2a32.2 of transcription 1, 91 kDa (STAT1), transcript variant beta, mRNA. 210048 at O.OOO8484 2.2 NAPG N-ethylmaleimide-sensitive 18p11.22 factor attachment protein, gamma (NAPG), mRNA. 210.069 at 3.07E-OS 2.8 CPT1B carnitine palmitoyltransferase 22d 13.33 1B (muscle) (CPT1B), nuclear gene encoding mitochondrial protein, transcript variant 3, mRNA. 210365 at O.OOO2181 2.6 RUNX1 Runt-related transcription factor 21q22.3 (acute myeloid leukemia 1: aml1 oncogene) 210438 x at 0.0003684 -2.7 TROVE2 TROVE domain family, 1q31 member 2 (TROVE2), mRNA. 210621 S at 0.0004.078 -2.1 RASA1 RAS p21 protein activator 5q13.3 (GTPase activating protein) 1 (RASA1), transcript variant 2, mRNA. 210664 S at 0.0006973 2.4 TFPI issue factor pathway inhibitor 231-q32.1 (lipoprotein-associated coagulation inhibitor) (TFPI), transcript variant 1, mRNA. 210665 at O.OOO3981 3.1 TFPI issue factor pathway inhibitor 231-q32.1 (lipoprotein-associated coagulation inhibitor) (TFPI), transcript variant 2, mRNA. 210679 X at 0.000265 3.4 BCL7A B-cell CLLlymphoma 7A 12q24.13 210788 s at 1.26E-05 -3.1 DHRS7 dehydrogenase/reductase (SDR 14q23.1 amily) member 7 (DHRS7), mRNA. 210800 at O.OOO97O6 4.7 MGC12262 hypothetical protein MGC12262 210809 s at 0.0002365 6.7 POSTN periostin, osteoblast specific 13q13.3 actor (POSTN), mRNA. 210944 S at 0.0009253 2.3 CAPN3 calpain 3, (p94) (CAPN3), 15q15.1-q21.1 transcript variant 7, mRNA. 210950 s at 1.38E-05 -3.3 FDFT1 airnesyl-diphosphate 8p23.1-p22 arnesyltransferase 1 (FDFT1), mRNA. US 8,440,393 B2 87 88 TABLE 1-continued Gene expression profile. Affymetrix Parametric Fold Change Probe Set P-Value (Tumor? Normal) Gene Symbol Description Map 211040 x at 0.0006364 2.4 GTSE1 G-2 and S-phase expressed 1 22d 13.2-q13.3 (GTSE1), mRNA. 211276 at 8.52E-OS -5.6 TCEAL2 transcription elongation factor A Xq22.1-q22.3 (SII)-like 2 (TCEAL2), mRNA. 211423 s at 0.000164 -2.8 SCSDL sterol-C5-desaturase (ERG3 11q23.3 delta-5-desaturase homolog, ungal)-like (SCSDL), transcript variant 1 mRNA. 211445 X at 0.0007153 3.6 FKSG17 FKSG17 8q22.3 211452 x at 5.23E-05 3.4 LRRFIP1 eucine rich repeat (in FLII) 2a37.3 interacting protein 1 (LRRFIP1), mRNA. 211454 X at 0.0006917 3.7 211569 s at 2.66E-05 -5.7 HADHSC L-3-hydroxyacyl-Coenzyme A 4q22-q26 dehydrogenase, short chain (HADHSC), mRNA. 211597 s at 1.00E-07 13.1 HOP homeodomain-only protein 4q11-q12 (HOP), transcript variant 2, mRNA. 211623 s at 0.0008386 -2.2 FBL fibrillarin (FBL), mRNA. 19q13.1 211666 X at 9.6OE-06 -2.6 RPL3 ribosomal protein L3 (RPL3), 22d 13 mRNA. 211673 s at 5.65E-05 3.5 MOCS1 Molybdenum cofactor synthesis 1 6p21.3 211710 X at 0.0001136 -2.2 RPL4 ribosomal protein L4 (RPL4), 15q22 mRNA. 211725 S. at 0.0006346 2.4 BID BH3 interacting domain death 22q11.1 agonist (BID), transcript variant 3, mRNA. 211727 s at 0.0004584 -2.5 COX11 COX11 homolog, cytochrome c 17q22 oxidase assembly protein (yeast) (COX11), nuclear gene encoding mitochondrial protein, mRNA. 211749 s at 7.41E-05 -2.6 WAMP3 vesicle-associated membrane 1p36.23 protein 3 (cellubrevin) (VAMP3), mRNA. 211762 s at 0.0002377 2.3 KPNA2 karyopherin alpha 2 (RAG 17q23.1-q23.3 cohort 1, importin alpha 1) (KPNA2), mRNA. 211769 X at 4.67E-05 -2.4 TDE1 tumor differentially expressed 1 20131-13.3 (TDE1), transcript variant 1, mRNA. 211813 X at 1.69E-05 -4.8 DCN decorin (DCN), transcript 2d21.33 variant D, mRNA. 211896 S. at 0.0006287 -3.7 DCN decorin (DCN), transcript 2d21.33 variant C. mRNA. 211937 at O.OOO1012 -3.2 EIF4B eukaryotic translation initiation 2d 13.13 factor 4B (EIF4B), mRNA. 211938 at 3.OOE-07 -3.2 EIF4B eukaryotic translation initiation 2d 13.13 factor 4B (EIF4B), mRNA. 211941 s at 2.01E-05 -2.3 PBP prostatic binding protein (PBP), 2d24.23 mRNA. 211942 X at 7.02E-05 -2.8 RPL13A Ribosomal protein L13a 9d 13.3 211964 at O.OOO6935 3.4 COL4A2 Collagen, type IV, alpha 2 3d34 211980 at O.OOOS493 3.9 COL4A1 collagen, type IV, alpha 1 3d34 (COL4A1), mRNA. 211986 at 3.42E-05 -3.3 AHNAK AHNAKnucleoprotein 1a12.2 (desmoyokin) (AHNAK), transcript variant 1, mRNA. 211988 at 1.3OE-OS -2.2 SMARCE1 SWI/SNF related, matrix 7q21.2 associated, actin dependent regulator of chromatin, Subfamily e, member 1 (SMARCE1), mRNA. 211994 at O.OOO12S6 -2 Transcribed locus, strongly 12 similar to XP 508919.1 PREDICTED: similar to protein kinase, lysine deficient 1; kinase deficient protein Pan troglodytes 211997 X at 5.21E-05 -2.5 H3F3B H3 histone, family 3B (H3.3B) 17q25 (H3F3B), mRNA. 211998 at 7.8OE-OS -3 H3F3B H3 histone, family 3B (H3.3B) 17q25 US 8,440,393 B2 89 90 TABLE 1-continued Gene expression profile. Affymetrix Parametric Fold Change Probe Set P-Value (Tumor? Normal) Gene Symbol Description Map 212037 at O.OOO1271 -2.7 PNN Pinin, desmosome associated 14q21.1 protein 212044 S at 9.00E-07 3.4 RPL27A ribosomal protein L27a 11 p 15 (RPL27A), mRNA. 212052 S at 0.0005363 2 KIAAO676 KIAA0676 protein 5q.35.3 (KIAAO676), transcript variant 2, mRNA. 212094 at O.OOO1189 -4.8 PEG10 PREDICTED: paternally 7 expressed 10 (PEG10), mRNA. 212096 s at 0.0002831 -2.7 MTUS1 mitochondrial tumor suppressor 8p22 (MTUS1), nuclear gene encoding mitochondrial protein, transcript variant 5, mRNA. 212131 at 4.28E-OS -2 FAM61A amily with sequence similarity 19q13.11 61, member A (FAM61A), mRNA. 212134 at O.OOO3O33 2 PHILDB1 pleckstrin homology-like 11q23.3 domain, family B, member 1 (PHLDB1), mRNA. 212144 at O.OOOS431 -2.1 UNC84B unc-84 homolog B (C. elegans) 22d 13.1 (UNC84B), mRNA. 212148 at 9.2OE-06 -2.9 PBX1 Pre-B-cell leukemia 1q23 transcription factor 1 212151 at O.OOO1748 -2.5 PBX1 Pre-B-cell leukemia 1q23 transcription factor 1 212171 x at 0.0006474 3.3 VEGF vascular endothelial growth 6p12 actor (VEGF), transcript variant 6, mRNA. 212179 at O.OOO3184 -2.5 C6orf111 Chromosome 6 open reading 6q16.3 rame 111 212188 at O.OOO184 -2.9 KCTD12 potassium channel 13q22.3 etramerisation domain containing 12 (KCTD12), mRNA. 212195 at O.OOO2328 -2.1 IL6ST interleukin 6 signal transducer 5q11 (gp130, oncostatin M receptor) 212199 at O.OOO1637 -2.5 MRFAP1L1 Morfa family associated protein 4p16.1 -like 1 (MRFAP1L1), transcript variant 2, mRNA. 212215 at O.OOO1021 -2.8 PREPL prolyl endopeptidase-like 2p22.1 (PREPL), mRNA. 212224 at 1.18E-05 -4.3 ALDH1A1 aldehyde dehydrogenase 1 9q21.13 amily, member A1 (ALDH1A1), mRNA. 212236 X at 0.0002768 5.3 KRT17 keratin 17 (KRT17), mRNA. 17q12-q21 212254 s at 4.00E-06 -2.7 DST dystonin (DST), transcript 6p12-p11 variant lea, mRNA. 212256 at O.OOO7645 -2.9 GALNT10 UDP-N-acetyl-alpha-D- 5q33.2 galactosamine:polypeptide N acetylgalactosaminyltransferase O (GalNAc-T10) 212323 s at 2.11E-05 2.5 VPS13D vacuolar protein sorting 13D 1p36.22-p36.21 (yeast) (VPS13D), transcript variant 2 mRNA. 212351 a 8.OOE-06 2.2 EIF2B5 eukaryotic translation initiation 3q27.1 factor 2B, subunit 5 epsilon, 82 kDa (EIF2B5), mRNA. 212353 a O.OOO225 3.6 SULF1 Sulfatase 1 (SULF1), mRNA. 8q13.2-q13.3 212354 a O.OOO2219 2.9 SULF1 Sulfatase 1 (SULF1), mRNA. 8q13.2-q13.3 212365 a. O.OOO3347 4.1 MYO1B myosin IB (MYO1 B), mRNA. 2012-q34 212368 a O.OOO2867 -2.3 ZNF292 PREDICTED: zinc finger 6 protein 292 (ZNF292), mRNA. 212408 a 1.41E-05 -2.6 TOR1AIP1 torsin A interacting protein 1 1q24.2 (TOR1AIP1), mRNA. 212413 a O.OOOS848 2 6-Sep septin 6 (SEPT6), transcript Xq24 variant II, mRNA. 212414 S at 0.0001527 2.4 6-Sep septin 6 (SEPT6), transcript Xq24 variant II, mRNA. 212435 a. O.OOO3489 -2.3 TRIM33 tripartite motif-containing 33 1p13.1 (TRIM33), transcript variant b, mRNA. 212468 a O.OOO1247 -2 SPAG9 Sperm associated antigen 9 17q21.33 212498 a 1SOE-06 -2.7 MARCH-VI Membrane-associated ring 5p15.2 finger (C3HC4) 6 US 8,440,393 B2 91 92 TABLE 1-continued Gene expression profile. Affymetrix Parametric Fold Change Probe Set P-Value (Tumor? Normal) Gene Symbol Description Map 212510 at O.OOO4844 -3.1 GPD1 L, glycerol-3-phosphate 3p24.1 dehydrogenase 1-like (GPD1L), mRNA. 212520 S. at 3.25E-05 3.1 SMARCA4 SWI/SNF related, matrix 19p 13.2 associated, actin dependent regulator of chromatin, Subfamily a member 4 (SMARCA4), mRNA. 212526 at 6.11E-OS -2.3 SPG20 spastic paraplegia 20, Spartin 13q13.3 (Troyer syndrome) (SPG20), mRNA. 212549 at 6.64E-OS -2.4 STATSB signal transducer and activator 17q11.2 of transcription 5B (STATSB), mRNA. 212556 at O.OOO3S6S 2.8 SCRIB scribbled homolog (Drosophila) 8q24.3 (SCRIB), transcript variant 2, mRNA. 212586 at O.OOO4909 -2.1 CAST calpastatin (CAST), transcript 5q15-q21 variant 2 mRNA. 212595 S at 0.0007053 -2 DAZAP2 DAZ associated protein 2 12q12 (DAZAP2), mRNA. 212609 S at 1.85E-05 -2.4 AKT3 V-akt murine thymoma viral 1q43-q14 oncogene homolog 3 (protein kinase B, gamma) 212624 S at 5.4OE-06 6.9 CHN1 chimerin (chimaerin) 1 (CHN1), 231-q32.1 transcript variant 2, mRNA. 2 2632 at O.OOO8287 -2 STX7 Syntaxin 7 6q23.1 212638 s at 0.0007524 -2.2 WWP1 WW domain containing E3 8q21 ubiquitin protein ligase 1 (WWP1), mRNA. 212644 S at 0.0001184 -2.3 C14orf32 chromosome 14 open reading 14q22.2-q22.3 frame 32 (C14orf32), mRNA. 212646 at O.OOO3295 2.6 RAFTLIN raft-linking protein (RAFTLIN), 3p25.1-p24.3 mRNA. 212653 S at 3.00E-07 -3.8 EHBP1 EH domain binding protein 1 2p15 (EHBP1), mRNA. 212675 S. at 0.0003318 -3.5 KIAAOS82 KIAAOS82 2p14 212730 at O.OOO67O6 -3.4 DMN desmuslin (DMN), transcript 15p26.3 variant B, mRNA. 212731 at O.OOO1 OS -3.3 ANKRD46 ankyrin repeat domain 46 8q22.3 (ANKRD46), mRNA. 212751 at O.OOO3336 -2.7 UBE2IN ubiquitin-conjugating enzyme 12q22 E2N (UBC13 homolog, yeast) (UBE2N), mRNA. 212769 at O.OOO184 2.7 TLE3 Transducin-like enhancer of 15q22 split 3 (E(Sp1) homolog, Drosophila) 212776 s at 0.0004769 -3 KIAAO657 PREDICTED: KIAAO657 2 protein (KIAAO657), mRNA. 212779 at O.OOOS/62 -2 KIAA1109 PREDICTED: hypothetical 4 protein KIAA1109 (KLAA1109), mRNA. 212798 s at 3.32E-05 -2 ANKMY2 ankyrin repeat and MYND 7p21 domain containing 2 (ANKMY2), mRNA. 212809 at 7.32E-OS 2.6 NFATC2IP nuclear factor of activated T- 16p11.2 cells, cytoplasmic, calcineurin dependent 2 interacting protein (NFATC2IP), mRNA. 212841 s at 0.0004662 2.6 PPFIBP2 PTPRF interacting protein, 11 p 15.4 binding protein 2 (liprin beta 2) (PPFIBP2), mRNA. 212943 at O.OOO9511 -2.2 KIAAOS28 KIAA0528 gene product 12p12.1 (KIAAO528), mRNA. 212971 at O.OOO4892 2.2 CARS Cysteinyl-tRNA synthetase 11 p 15.5 213002 at 7.97E-05 3.8 MARCKS myristoylated alanine-rich 6q22.2 protein kinase C Substrate (MARCKS), mRNA. 213005 s at 0.0003887 -2.3 ANKRD15 ankyrin repeat domain 15 9p24.3 (ANKRD15), transcript variant 1, mRNA. 213027 at O.OOO1503 -2.1 SSA2 TROVE domain family, 1q31 member 2 US 8,440,393 B2 93 94 TABLE 1-continued Gene expression profile. Affymetrix Parametric Fold Change Probe Set P-Value (Tumor? Normal) Gene Symbol Description Map 213029 at 6.18E-OS -2.4 NFIB Nuclear factor IB 9p24.1 213047 X at 0.0003826 -2.6 SET SET translocation (myeloid 9q34 eukemia-associated) (SET), mRNA. 213074 at O.OOO2338 -2.6 PHIP Pleckstrin homology domain 6q14 interacting protein 213085 s at 0.0005289 3.8 KIBRA KIBRA protein (KIBRA), 5q34 mRNA. 213093 at 18OE-05 -2.8 PRKCA protein kinase C, alpha 17q22-q23.2 (PRKCA), mRNA. 213110 s. at 5.00E-06 -3.9 COL4A5 collagen, type IV, alpha 5 Xq22 (Alport syndrome) (COL4A5), transcript variant 1, mRNA. 213111 a O.OOO6152 -2 PIPSK3 phosphatidylinositol-3- 2a34 phosphate:phosphatidylinositol 5-kinase, type III (PIP5K3), transcript variant 1, mRNA. 213117 a O.OOO8783 -2 KLHL9 kelch-like 9 (Drosophila) 9p22 (KLHL9), mRNA. 213139 a O.OOO8306 2.6 SNAI2 Snail homolog 2 (Drosophila) 8q11 (SNAI2), mRNA. 213146 a O.OOO12S 2.9 KIAAO346 KIAA0346 protein 17p13.1 213224 s at 0.0005954 -2 LOC92482 PREDICTED: hypothetical 10 protein LOC92482 (LOC92482), mRNA. 213227 a O.OOO1677 -2 PGRMC2 Progesterone receptor membrane 4q26 component 2 213248 a O.OOO6637 2.4 LOC221362 Hypothetical protein 6p12.3 LOC221362 213258 a 3.1OE-OS 2.5 TFPI Tissue factor pathway inhibitor 231-q32.1 (lipoprotein-associated coagulation inhibitor) 213272 s at 6.80E-06 -4 LOCS7146 promethin (LOC57146), mRNA. 16p12 213344 s at 3.91E-05 2.7 H2AFX H2A histone family, member X 11q23.2-q23.3 (H2AFX), mRNA. 213350 at 1.2OE-05 6.1 RPS11 ribosomal protein S11 (RPS11), 19q13.3 mRNA. 213364 S at 4.00E-07 -3.7 SNX1 sorting nexin 1 (SNX1), 15q22.31 transcript variant 2, mRNA. 213370 S. at 0.0001347 2.1 SFMBT1 Scm-like with four mbt domains 3p21.1 1 (SFMBT1), transcript variant 3, mRNA. 213397 X at 4.3OE-06 -5.2 RNASE4 ribonuclease, RNase A family, 4 14q11.1 (RNASE4), transcript variant 3, mRNA. 213405 at O.OOO663 -2.4 RAB22A RAB22A, member RAS 20d 13.32 oncogene family (RAB22A), mRNA. 213413 at O.OOOS89 -2.4 SBLF stoned B-like factor (SBLF), 2p16.3 mRNA. 213418 at 3.79E-05 6.8 HSPA6 heat shock 70 kDa protein 6 1q23 (HSP7OB') (HSPA6), mRNA. 213464 at O.OOOS/18 2.7 SHC2 SHC (Src homology 2 domain 19p13.3 containing) transforming protein 2 213479 at 2.75E-05 4.3 NPTX2 neuronal pentraxin II (NPTX2), 7q21.3-q22.1 mRNA. 213523 at 5.86E-OS 2.6 CCNE1 cyclin E1 (CCNE1), transcript 19q12 variant 2 mRNA. 213560 at 0.0005733 2.4 GADD45B growth arrest and DNA-damage- 19p13.3 inducible, beta (GADD45B), mRNA. 213574 s at 1.OOE-06 -2.4 KPNB1 Karyopherin (importin) beta 1 17q21.32 213605 S at 0.0009961 2.5 FLJ40092 FLJ40092 protein 5q13.2 213661 at 8.OOE-07 6.2 DKFZP586H2123 regeneration associated muscle 11 p.13 protease (DKFZP586H2123), transcript variant 2, mRNA. 213687 s at 0.0009746 -2.1 RPL3SA ribosomal protein L35a 3q29-qter (RPL35A), mRNA. 213693 s at 7.45E-05 6.2 MUC1 mucin 1, transmembrane 1q21 (MUC1), transcript variant 4, mRNA. 213778 X at 0.000467 2.2 ZFP276 Zinc finger protein 276 homolog 16q24.3 (mouse) (ZFP276), mRNA. US 8,440,393 B2 95 96 TABLE 1-continued Gene expression profile. Affymetrix Parametric Fold Change Probe Set P-Value (Tumor? Normal) Gene Symbol Description Map 213790 at 7.S1E-OS 4.4 ADAM12 A disintegrin and 10q26.3 metalloproteinase domain 12 (meltrin alpha) 213803 at p < 1e–07 -3.6 KPNB1 Karyopherin (importin) beta 1 17q21.32 213836 s at 0.0001.261 2.2 WIPI49 WD40 repeat protein Interacting 17q24.2 with phosphoInositides of 49 kDa (WIPI49), mRNA. 213848 at 2.OOE-07 3.1 DUSP7 Dual specificity phosphatase 7 3p21 213869 X at 0.0002103 4.6 THY1 Thy-1 cell Surface antigen 11q22.3-q23 (THY1), mRNA. 213895 at O.OOO44-85 -2.5 EMP1 Epithelial membrane protein 1 12p12.3 213900 at 3.89E-OS -2.7 C9orf61 chromosome 9 open reading 9q13-q21 rame 61 (C9orf61), mRNA. 213905 X at 0.0001976 4.6 BGN Biglycan Xq28 213943 at 4.OOE-07 19 TWIST1 twist homolog 1 7p21.2 (acrocephalosyndactyly 3: Saethre-Chotzen syndrome) (Drosophila) (TWIST1), mRNA. 213979 S at 0.0002226 3.3 CTBP1 C-terminal binding protein 1 4p16 (CTBP1), transcript variant 1, mRNA. 214023 x at 0.000646 2.8 MGC8685 Tubulin, beta polypeptide 6p25 paralog 214041 x at 0.000362 2.8 RPL37A Ribosomal protein L37a 2a35 214052 x at 0.0005361 2.7 BAT2D1 BAT2 domain containing 1 1q23.3 (BAT2D1), mRNA. 214057 at O.OOO5414 2.3 MCL1 myeloid cell leukemia sequence 1q21 1 (BCL2-related) (MCL1), transcript variant 2, mRNA. 214081 at 4.OOE-07 10.2 PLXDC1 plexin domain containing 1 17q21.1 (PLXDC1), mRNA. 214097 at 2.29E-OS -2.3 RPS21 Ribosomal protein S21 20d 13.3 214110 s. at 0.0001953 3 ESTs, Highly similar to A43542 lymphocyte-specific protein 1 H. Sapiens 214140 at O.OOO4058 2 SLC25A16 solute carrier family 25 10q21.3 (mitochondrial carrier; Graves disease autoantigen), member 16 (SLC25A16), nuclear gene encoding mitochondrial protein, mRNA. 214149 s at 0.0003428 3.4 ATP6VOE ATPase, H+ transporting, 5q.35.1 lysosomal 9 kDa, VO subunite 214177 s at 0.0001027 -2.2 PBXIP1 pre-B-cell leukemia 1q22 transcription factor interacting protein 1 (PBXIP1), mRNA. 214264 S at 0.0002244 2.1 C14orf143 chromosome 14 open reading 14q32.11 frame 143 (C14orf143), mRNA. 214316 x at 7.97E-05 2.4 CALR Calreticulin 19p13.3-p13.2 214359 is at 0.0001495 -3 HSPCB heat shock 90 kDa protein 1, beta 6p12 (HSPCB), mRNA. 214426 X at 0.0008417 2.1 CHAF1A chromatin assembly factor 1, 19p13.3 subunit A (p150) (CHAF1A), mRNA. 214435 X at 0.0005753 2.8 RALA v-ral simian leukemia viral 7p15-p13 oncogene homolog A (ras related) (RALA), mRNA. 214438 at O.OOO956S 3.4 HLX1 H2.0-like homeo box 1 1q41-q42.1 (Drosophila) (HLX1), mRNA. 214527 s at 4.2OE-06 -2.3 PQBP1 polyglutamine binding protein 1 Xp11.23 (PQBP1), transcript variant 5, mRNA. 214594 X at 0.0002968 2.9 ATP8B1 ATPase, Class I, type 8B, 18q21-q22 member 1 214707 X at 0.0003366 2.8 ALMS1 Alstrom syndrome 1 2p13 214715 x at 0.0002241 4.3 ZNF160 Zinc finger protein 160 19q13.41 (ZNF160), transcript variant 1, mRNA. 214724 at O.OOO6038 -2.1 DIXDC1 DIX domain containing 1 11q23.2 (DDXDC1), mRNA. 214737 X at 0.0004018 -2.2 HNRPC heterogeneous nuclear 14q11.2 ribonucleoprotein C (C1/C2) (HNRPC), transcript variant 2, mRNA. US 8,440,393 B2 97 98 TABLE 1-continued Gene expression profile. Affymetrix Parametric Fold Change Probe Set P-Value (Tumor? Normal) Gene Symbol Description Map 214855 S at 4.03E-05 -2.6 GARNL1 GTPase activating Rap/RangAP 14q13.2 domain-like 1 (GARNL1), transcript variant 2, mRNA. 214862 X at 0.0009591 -2.8 MRNA, cDNA 10 DKFZp564G 1162 (from clone DKFZp564G 1162) 214924 S at 2.90E-06 3.8 OIP106 OGT(O-Glc-NAc transferase)- 3p25.3-p24.1 interacting protein 106 KDa (OIP106), mRNA. 215016 X at 1.5OE-05 -2.3 DST dystonin (DST), transcript 6p12-p11 variant 1 mRNA. 215046 at O.OOO6908 -2.2 FLU23861 hypothetical protein FLJ23861 2a34 (FLJ23861), mRNA. 215073 s at 3.83E-05 -2.1 NR2F2 nuclear receptor subfamily 2, 15q26 group F, member 2 (NR2F2), mRNA. 215076 s at 0.000876 3 COL3A1 collagen, type III, alpha 1 2a31 (Ehlers-Danlos syndrome type IV, autosomal dominant) (COL3A1), mRNA. 215179 X at 6.80E-05 4.5 PGF Placental growth factor, vascular 14q24-q31 endothelial growth factor-related protein 215203 at O.OOO626S 2.2 GOLGA4 Golgi autoantigen, golgin 3p22-p21.3 Subfamily a 4 215206 at O.OOO1618 3.1 EXT1 Exostoses (multiple) 1 8q24.11-q24.13 2 5208 X at 0.0006292 3.4 RPL3SA Ribosomal protein L35a 3q29-qter 215294 S at 8.91E-05 -2.9 SMARCA1 SWI/SNF related, matrix Xq25 associated, actin dependent regulator of chromatin, Subfamily a member 1 (SMARCA1), transcript variant 2, mRNA. 2153.06 at O.OOO8848 -3.4 LHCGR. Luteinizing 2p21 hormone? choriogonadotropin receptor 215336 at 9.2iSE-OS 2.7 AKAP11 A kinase (PRKA) anchor protein 13q14.11 1 (AKAP11), transcript variant 2, mRNA. 215373 X at 0.0004506 2.8 SET8 PR/SET domain containing 12q24.31 protein 8 215383 x at 0.0009763 2.6 SPG21 Spastic paraplegia 21 15q21-q22 (autosomal recessive, Mast syndrome) 215404 x at 0.0005104 3.2 FGFR1 Fibroblast growth factor 8p11.2-p11.1 receptor 1 (fms-related tyrosine kinase 2, Pfeiffer syndrome) 215467 X at 0.0004682 3.9 DHX9 DEAH (Asp-Glu-Ala-His) box 1q25 polypeptide 9 215504 x at 0.0006038 2.5 ANKRD10 Ankyrin repeat domain 10 13q34 215529 X at 0.0007.444 2.6 C21orf106 hromosome 21 open reading 21q22.3 .ame 106 215566 X at 0.0007954 2.2 LYPLA2 sophospholipase II (LYPLA2), 1p36.12-p35.1 RNA. 215577 at S.29E-OS 2.4 UBE2E1 biquitin-conjugating enzyme 3p24.2 E 2E 1 (UBC45 homolog, yeast) 215588 x at 6.38E-05 3.7 RIOK3 Okinase 3 (yeast) 18q11.2 215599 at 4.1OE-06 3.9 SMA4 MA4 5q13 215600 X at 0.0005929 3.3 FBXW12 s box and WD-40 domain 3p21.31 protein 12 215604 x at 0.0003643 3.8 UBE2D2 Ubiquitin-conjugating enzyme 5q31.2 E2D2 (UBC45 homolog, yeast) 215628 x at 0.0005005 2.7 PPP2CA Protein phosphatase 2 (formerly 5q31.1 2A), catalytic Subunit, alpha isoform 215978 x at 9.04E-05 S.6 LOC152719 ATP-binding cassette, Sub- 4p16.3 amily A (ABC1), member 11 (pseudogene) 216035 x at 0.000632 -2 TCF7L2 Transcription factor 7-like 2 (T. 10q25.3 cell specific, HMG-box) 216051 X at 5.83E-05 5 KIAA1217 KIAA1217 10p12.31 21618.7 x at 0.0001456 3.7 XRCC3 X-ray repair complementing 14q32.3 defective repair in Chinese hamster cells 3 US 8,440,393 B2 99 100 TABLE 1-continued Gene expression profile. Affymetrix Parametric Fold Change Probe Set P-Value (Tumor? Normal) Gene Symbol Description Map 216215 S. at 7.67E-05 -2.1 RBM9 RNA binding motif protein 9 22d 13.1 216221 S at 0.0007062 -2 PUM2 pumilio homolog 2 (Drosophila) 2p22-p21 (PUM2), mRNA. 216241 s at 0.0005875 -2.1 TCEA1 transcription elongation factor A 8q11.2 (SII), 1 (TCEA1), transcript variant 2 mRNA. 216246 at O.OOO4251 2.7 RPS2O ribosomal protein S20 (RPS20), 8q12 mRNA. 216274 S at 0.0003437 -2.3 SEC11L1 SEC11-like 1 (S. cerevisiae) 15q25.3 (SEC11L1), mRNA. 216733 s at 0.0005519 -3.3 GATM glycine amidinotransferase (L- 15q21.1 arginine:glycine amidinotransferase) (GATM), mRNA. 216858 x at 0.0002005 4.1 216859 X at 0.0004599 3.7 216944 S at 0.0001923 -2.7 ITPR1 inositol 14,5-triphosphate 3p26-p25 receptor, type 1 (ITPR1), mRNA. 217028 at O.OOO384 3.4 CXCR4 chemokine (C X—C motif) 2d21 receptor 4 (CXCR4), transcript variant 2 mRNA. 217118 s at 2.15E-05 2.2 C22Orf open reading 22d 13.31 frame 9 (C22orf), transcript variant 2 mRNA. 217466 X at 0.0009002 -2.2 RPS2 Ribosomal protein S2 16p13.3 217497 at O.OOO2299 3.3 ECGF1 Endothelial cell growth factor 1 22d 13 (platelet-derived) 217579 x at 0.0006875 2.5 ARL6IP2 ADP-ribosylation factor-like 6 2p22.2-p22.1 interacting protein 2 217586 X at 0.000568 2.9 ESTs 217679 X at 7.86E-05 4.8 ESTs, Weakly similar to hypothetical protein FLJ20489 Homo sapiens H. Sapiens 217713 X at 0.0001301 3.4 ESTs, Weakly similar to ALU6 HUMAN ALU SUBFAMILY SP SEQUENCE CONTAMINATION WARNING ENTRY H. Sapiens 217715 x at 0.0002336 3.9 ESTs 217773 s at 0.0002007 -2.4 NDUFA4 NADH dehydrogenase 7p21.3 (ubiquinone) 1 alpha Subcomplex, 4, 9 kDa (NDUFA4), nuclear gene encoding mitochondrial protein, mRNA. 217774 s at 0.0007.057 -2.1 HSPC152 hypothetical protein HSPC152 11 q13.1 (HSPC152), mRNA. 217781 s at 0.0003457 -2.5 ZFP106 Zinc finger protein 106 homolog 15q15.1 (mouse) (ZFP106), mRNA. 217787 s at 8.07E-05 2.7 GALNT2 UDP-N-acetyl-alpha-D- 1q41-q42 galactosamine:polypeptide N acetylgalactosaminyltransferase 2 (GalNAc-T2) (GALNT2), mRNA. 217795. S. at 0.0001341 -2.1 TMEM43 transmembrane protein 43 3p25.1 (TMEM43), mRNA. 217814 at O.OOO4705 -2.2 GKOO1 GKOO1 protein (GKO01), 17q23.3 mRNA. 217833 at O.OOOS95 -2.5 SYNCRIP Synaptotagmin binding, 6q14-q15 cytoplasmic RNA interacting protein 217862 at O.OOO3S1 -2.6 PIAS1 Protein inhibitor of activated 15q. STAT, 1 217864 S at 0.0004-053 -2.2 PIAS1 protein inhibitor of activated 15q. STAT, 1 (PIAS1), mRNA. 21788.8 s at 0.0003955 2.2 ARFGAP1 ADP-ribosylation factor GTPase 20d 13.33 activating protein 1 (ARFGAP1), transcript variant 1, mRNA. 217915 S. at 9.25E-05 -2.2 C1 Sorf15 open reading 15q21 frame 15 (C15orf15), mRNA. US 8,440,393 B2 101 102 TABLE 1-continued Gene expression profile. Affymetrix Parametric Fold Change Probe Set P-Value (Tumor? Normal) Gene Symbol Description Map 217989 at 2.94E-05 -2.7 DHRS8 dehydrogenase/reductase (SDR 4q22.1 amily) member 8 (DHRS8), mRNA. 217992 S at 0.0003726 2.8 EFEHD2 EF-hand domain family, 1p36.21 member D2 (EFHD2), mRNA. 218018 at 3.6SE-OS 2.2 PDXK pyridoxal (pyridoxine, vitamin 21q22.3 B6) kinase (PDXK), mRNA. 218025 S. at 2.20E-06 -4.3 PECI peroxisomal D3.D2-enoyl-CoA 6p24.3 isomerase (PECI), transcript variant 1 mRNA. 218031 s at 4.OOE-07 -4.3 CHES1 checkpoint Suppressor 1 14q24.3-q32.11 (CHES1), mRNA. 218113 at O.OOO146 2 TMEM2 transmembrane protein 2 9q13-q21 (TMEM2), mRNA. 218130 at O.OOO4049 2 MGC4368 hypothetical protein MGC4368 17q25.3 (MGC4368), mRNA. 218131 S. at 2.87E-05 2.9 GATAD2A GATA zinc finger domain 19p 13.11 containing 2A (GATAD2A), mRNA. 218151 X at 0.0005594 2.2 GPR172A G protein-coupled receptor 8q24.3 72A (GPR172A), mRNA. 21815S X at 0.000544 2.6 FLJ10534 hypothetical protein FLJ10534 17p13.3 (FLJ10534), mRNA. 21815.8 s at 0.0001389 -2.7 APPL adaptor protein containing pH 3p21.1-p14.3 domain, PTB domain and eucine zipper motif 1 (APPL), mRNA. 218167 at 7.25E-05 -2.4 AMZ2 archaemetZincins-2 (AMZ2), 17q24.2 mRNA. 218191 s at 6.81E-05 -2.8 LMBRD1 LMBR1 domain containing 1 6q13 (LMBRD1), mRNA. 218193 s at 0.0002316 2.8 GOLT1B golgi transport 1 homolog B (S. cerevisiae) 12p12.1 (GOLT1B), mRNA. 218204 s at 8.33E-05 -2.3 FYCO1 FYVE and coiled-coil domain 3p21.31 containing 1 (FYCO1), mRNA. 218311 at O.OOO1916 -2.3 MAP4K3 mitogen-activated protein kinase 2p22.1 kinase kinase kinase 3 (MAP4K3), mRNA. 218373 at O.OOO1922 -2.3 FTS used toes homolog (mouse) 16q12.2 (FTS), transcript variant 2, mRNA. 218383 at O.OOO3931 -2.7 C14orf4 chromosome 14 open reading 14q11.2 rame 94 (C14orf)4), mRNA. 218432 at O.OOO3175 -2.1 FBXO3 F-box protein 3 (FBXO3), 11 p.13 transcript variant 1, mRNA. 218450 at O.OOO2984 -2.5 HEBP1 heme binding protein 1 12p13.1 (HEBP1), mRNA. 218504 at O.OOOSS44 -2 FAHD2A umarylacetoacetate hydrolase 2p24.3-p11.2 omain containing 2A (FAHD2A), mRNA. 218528 S. at 4.23E-05 -2.3 RNF38 ring finger protein 38 (RNF38), 9p13-p12 transcript variant 4, mRNA. 2186.38 S. at 0.0008018 4 SPON2 spondin 2, extracellular matrix 4p16.3 protein (SPON2), mRNA. 218730 S. at 0.0002624 -6.8 OGN osteoglycin (osteoinductive 9q22 actor, mimecan) (OGN), transcript variant 3, mRNA. 218739 at O.OOO2299 2.6 ABHDS abhydrolase domain containing 3p21 5 (ABHD5), mRNA. 218804 at 1.58E-OS 3.3 TMEM16A transmembrane protein 16A 11 q13.3 (TMEM16A), mRNA. 218817 at O.OOOSS21 2 SPCS3 signal peptidase complex 4q34.2 Subunit 3 homolog (S. cerevisiae) (SPCS3), mRNA. 218820 at O.OOO7651 -3 C14orf132 chromosome 14 open reading 14q32.2 frame 132 (C14orf132), mRNA. 218831 s at 7.99E-05 -2.6 FCGRT Fc fragment of IgG, receptor, 19q13.3 transporter, alpha (FCGRT), mRNA. 218856 at 2.2OE-06 6.6 TNFRSF21 tumor necrosis factor receptor 6p21.1-12.2 Superfamily, member 21 (TNFRSF21), mRNA. 218902 at 4.33E-OS 2.5 NOTCH1 Notch homolog 1, translocation- 9q34.3 associated (Drosophila) (NOTCH1), mRNA. US 8,440,393 B2 103 104 TABLE 1-continued Gene expression profile. Affymetrix Parametric Fold Change Probe Set P-Value (Tumor? Normal) Gene Symbol Description Map 218919 at O.OOO1794 -3.1 ZFAND1 Zinc finger, AN1-type domain 1 8q21.13 (ZFAND1), mRNA. 218929 at 3.34E-OS -2.1 CARF collaborates, cooperates with 4q35.1 ARF (alternate reading frame) protein (CARF), mRNA. 218961 S. at 0.0007861 2 PNKP polynucleotide kinase 3'- 19q13.3-q13.4 phosphatase (PNKP), mRNA. 219023 at 7.64E-OS -3.1 C4orf16 open reading 4q25 rame 16 (C4orf16), mRNA. 219025 at 5.55E-05 4.3 CD248 CD248 antigen, endosialin 11 q13 (CD248), mRNA. 219033 at O.OOO7081 2.3 PARP8 poly (ADP-ribose) polymerase 5q11.1 amily, member 8 (PARP8), mRNA. 219054 at 4.17E-OS -2.6 FLJ14054 hypothetical protein FLJ14054 5p13.3 (FLJ14054), mRNA. 219092 S at 2.14E-05 2.2 C9orf12 chromosome 9 open reading 9q21.33-q22.31 rame 12 (C9orf12), mRNA. 219099 at 9.73E-OS 2 C12Orfs chromosome 12 open reading 12p13.3 rame 5 (C12orf5), mRNA. 219102 at 1.21E-05 3 RCN3 reticulocalbin 3, EF-hand 19q13.33 calcium binding domain (RCN3), mRNA. 219105 X at 9.00E-07 3.8 ORC6L origin recognition complex, 16q12 Subunit 6 homolog-like (yeast) (ORC6L), mRNA. 219117 s at 0.0002973 2.4 FKBP11 FK506 binding protein 11, 19 kDa 12q13.12 (FKBP11), mRNA. 219263 at S.92E-OS 4.2 RNF128 ring finger protein 128 Xq22.3 (RNF128), transcript variant 2, mRNA. 219279 at O.OOO2O23 2.7 DOCK10 dedicator of cytokinesis 10 2a36.3 (DOCK10), mRNA. 219289 at O.OOO1862 2.3 FL2O718 hypothetical protein FLJ20718 16q12.1 (FLJ20718), transcript variant 1, mRNA. 219290 X at 0.0009817 3.8 DAPP1 dual adaptor of phosphotyrosine 4q25-q27 and 3-phosphoinositides (DAPP1), mRNA. 21935.8 s at 1.6OE-06 6.1 CENTA2 centaurin, alpha 2 (CENTA2), 17q11.2 mRNA. 219359 at 4.OOE-06 4.4 FLJ2263S hypothetical protein FLJ22635 11 p 15.5 (FLJ22635), mRNA. 219368 at O.OOO9624 -2.9 NAP1L2 nucleosome assembly protein 1- Xq13 ike 2 (NAP1L2), mRNA. 219392 X at 3.14E-05 4.3 FLJ11029 hypothetical protein FLJ11029 17q23.2 (FLJ11029), mRNA. 219407 s at 0.0009809 4.1 LAMC3 aminin, gamma 3 (LAMC3), 9q31-q34 mRNA. 219449 s at 0.0006528 -2.5 TMEM70 transmembrane protein 70 8q21.11 (TMEM70), mRNA. 219454 at 4.OOE-07 36.8 EGFL6 EGF-like-domain, multiple 6 Xp22 (EGFL6), mRNA. 219493 at O.OOO4597 2.6 SHCBP1 SHC SH2-domain binding 16q11.2 protein 1 (SHCBP1), mRNA. 21951.1 s at 7.7OE-06 -4.1 SNCAIP Synuclein, alpha interacting 5q23.1-q23.3 protein (synphilin) (SNCAIP), mRNA. 219522 at 1.OOE-07 7.7 FJX1 our jointed box 1 (Drosophila) 11 p.13 (FJX1), mRNA. 219549 s at 0.0004199 -2.2 RTN3 reticulon 3 (RTN3), transcript 11 q13 variant 4, mRNA. 219582 at O.0004924 3 OGFRL1 opioid growth factor receptor- 6q13 like 1 (OGFRL1), mRNA. 219634 at 8.1OE-06 3.8 CHST11 carbohydrate (chondroitin 4) 12q Sulfotransferase 11 (CHST11), mRNA. 219641 at O.OOO33O8 -2.1 DET1 de-etiolated homolog 1 15q25.3 (Arabidopsis) (DET1), mRNA. 219700 at 1.1OE-06 7.7 PLXDC1 plexin domain containing 1 17q21.1 (PLXDC1), mRNA. 219764 at 9.OOE-07 6.7 FZD10 frizzled homolog 10 12q24.33 (Drosophila) (FZD10), mRNA. US 8,440,393 B2 105 106 TABLE 1-continued Gene expression profile. Affymetrix Parametric Fold Change Probe Set P-Value (Tumor? Normal) Gene Symbol Description Map 21993.9 s at 2.10E-06 -3.5 CSDE1 cold shock domain containing 1p22 E1, RNA-binding (CSDE1), transcript variant 2, mRNA. 219958 at O.OOO873 3.2 C20orfA6 open reading 20p13 rame 46 (C20orf46), mRNA. 21996.1 s at 0.0003792 -2.5 C20orf19 chromosome 20 open reading 20pter-q11.23 rame 19 (C20orf19), mRNA. 220014 at 1.38E-OS 4.6 LOCS1334 mesenchymal stem cell protein 5q23.1 DSC54 (LOC51334), mRNA. 220094 S at 0.0002242 2 C6orf79 chromosome 6 open reading 6p24.3-p23 rame 79 (C6orf79), transcript variant 1 mRNA. 22O113 x at 0.0004.047 3.1 POLR1B polymerase (RNA) I polypeptide 2d 13 B, 128 kDa (POLR1B), mRNA. 22O167 s at 0.0008187 2 TP53TG3 TP53TG3 protein (TP53TG3), 16p13 transcript variant 2, mRNA. 220232 at 4.54E-OS 4.8 SCD5 stearoyl-CoA desaturase 5 4q21.3 (SCD5), mRNA. 220242 x at 0.0001347 2.2 ZNF701 Zinc finger protein 701 19q13.41 (ZNF701), mRNA. 220266 s at 1.44E-05 -3.3 KLF4 Kruppel-like factor 4 (gut) 9q31 (KLF4), mRNA. 220301 at 1.76E-OS 5.8 C18orf14 chromosome 18 open reading 18q22.1 rame 14 (C18orf14), mRNA. 220327 at 8.86E-OS -2.7 WGL-3 vestigial-like 3 (VGL-3), 3p12.1 mRNA. 220334 at O.OOO3O43 3.1 RGS17 regulator of G-protein signalling 6q25.3 7 (RGS17), mRNA. 220432 S at 0.000473 -3.2 CYP39A1 cytochrome P450, family 39, 6p21.1-p11.2 Subfamily A, polypeptide 1 (CYP39A1), mRNA. 220575 at 4.48E-OS 3.4 FLJ11800 hypothetical protein FLJ1 1800 7p11.2 (FLJ11800), mRNA. 220603 s at O.OOO1617 3.9 MCTP2 Multiple C2-domains with two 5q26.2 transmembrane regions 2 220720 x at 0.0005432 3.3 FLJ14346 Hypothetical protein FLJ14346 2d21.1 220796 X at 0.0002356 4.1 SLC3SE1 Solute carrier family 35, 9p 13.11 member E1 220817 at 4.5OE-06 4.1 TRPC4 transient receptor potential 3q13.1-q13.2 cation channel, Subfamily C, member 4 (TRPC4), mRNA. 220952 S at 0.000293 -2.1 PLEKHAS pleckstrin homology domain 2p12 containing, family A member 5 (PLEKHA5), mRNA. 220992 S at 0.0003763 -2.1 C1orf25 Chromosome 1 open reading q25.2 rame 25 221012 s at 0.0004784 2 TRIM8 tripartite motif-containing 8 Oq24.3 (TRIM8), mRNA. 22105.9 s at 2.20E-06 4.1 COTL1 coactosin-like 1 (Dictyostelium) 6q24.1 (COTL1), mRNA. 221127 s at 0.000454 3.7 DKK3 dickkopfhomolog 3 (Xenopus 1p 15.2 aevis) (DKK3), transcript variant 3, mRNA. 221222 S at 0.0005338 2.2 C1orf56 chromosome 1 open reading q21.2 rame 56 (C1orf56), mRNA. 221476 s at 4.74E-05 -2.4 RPL15 ribosomal protein L15 (RPL15), 3p24.2 mRNA. 221486 at O.OOO4058 -2 ENSA endosulfine alpha (ENSA), q21.2 transcript variant 7, mRNA. 221538. S. at 0.0004626 2.4 PLXNA1 plexin A1 (PLXNA1), mRNA. 3q21.3 221558. S. at 1.86E-05 4.6 LEF1 ymphoid enhancer-binding 4q23-q25 actor 1 (LEF1), mRNA. 221577 X at 0.0006888 4.5 GDF15 growth differentiation factor 15 19p 13.1-13.2 (GDF15), mRNA. 221588 x at 0.0001441 -2.4 ALDH6A1 aldehyde dehydrogenase 6 14q24.3 amily, member A1 (ALDH6A1), nuclear gene encoding mitochondrial protein, mRNA. 221589 s at 1.20E-06 -4.3 ALDH6A1 Aldehyde dehydrogenase 6 14q24.3 family, member A1 221590 S. at 0.0008671 -2.3 ALDH6A1 Aldehyde dehydrogenase 6 14q24.3 family, member A1 US 8,440,393 B2 107 108 TABLE 1-continued Gene expression profile. Affymetrix Parametric Fold Change Probe Set P-Value (Tumor? Normal) Gene Symbol Description Map 221689 s at 0.000155 -2 DSCRS Down syndrome critical region 21q22.2 gene 5 (DSCR5), transcript variant 2 mRNA. 221 691 x at 1.46E-05 -3.2 NPM1 nucleophosmin (nucleolar 5q.35 phosphoprotein B23, numatrin) (NPM1), mRNA. 221725 at 3.6OE-OS -2.1 WASF2 WAS protein family, member 2 1p36.11-p34.3 221726 at 1.93E-OS -2.9 RPL22 ribosomal protein L22 (RPL22), 1p36.3-p36.2 mRNA. 221727 at O.OOO1534 -2.3 PC4 Activated RNA polymerase II 5p13.3 transcription cofactor 4 221729 at O.OOO4594 3.1 COLSA2 collagen, type V, alpha 2 214-q32 (COL5A2), mRNA. 221730 at O.OOO7972 3.4 COLSA2 collagen, type V, alpha 2 214-q32 (COL5A2), mRNA. 221731 X at 2.57E-05 10.4 CSPG2 chondroitin Sulfate proteoglycan 5q14.3 2 (versican) (CSPG2), mRNA. 221747 at O.OOO3412 -2.2 TNS Tensin 1 2a35-q36 221771 s at 0.000498 -2.1 HSMPP8 M-phase phosphoprotein, mpp8 3q12.11 221840 at O.OOO2221 3.2 PTPRE protein tyrosine phosphatase, Oq26 receptor type, E (PTPRE), transcript variant 2, mRNA. 221882 s at 0.0003013 2.2 TMEM8 transmembrane protein 8 (five 6p13.3 membrane-spanning domains) (TMEM8), mRNA. 221943 x at 0.0002266 2.4 RPL38 ribosomal protein L38 (RPL38), 7q23-q25 mRNA. 221988 at 1.59E-OS -2.3 MGC2747 Hypothetical protein MGC2747 9p 13.11 222108 at 5.57E-05 -4.4 AMIGO2 adhesion molecule with Ig-like 2d 13.11 domain 2 (AMIGO2), mRNA. 222207 X at 0.0003362 3.2 CDNA: FLJ20949 fis, clone 7 ADSEO1902 222212 s at 0.0001439 -2.3 LASS2 LAG1 longevity assurance q21.2 homolog 2 (S. cerevisiae) (LASS2), transcript variant 3, mRNA. 222252 x at 5.29E-05 4 LRRCS1 eucine rich repeat containing 51 1q13.4 (LRRC51), mRNA. 222253 S at 0.000969 4.6 DKFZP434P211 POM121-like protein 22d 11.22 222358 x at 7.05E-05 3.1 ESTs, Weakly similar to hypothetical protein FLJ20378 Homo sapiens H. sapiens 222372 a O.OOO6654 2.5 BALAP1 Membrane associated guanylate 3p 14.1 kinase, WW and PDZ domain containing 1 222379 a 1.13E-OS 3.5 KCNE4 Potassium voltage-gated 2a36.3 channel, Isk-related family, member 4 222394 a 1.44E-05 -2.8 PDCD6IP programmed cell death 6 3p23 interacting protein (PDCD6IP), mRNA. 222423 a 9.81E-OS -2.8 NDFIP1 Nedd4 family interacting protein 1 5q31.3 222431 a O.OOO2108 -2.2 SPIN Spindlin 9q22.1-q22.3 222437 S at 0.0006141 -2 VPS24 vacuolar protein sorting 24 2p24.3-p24.1 (yeast) (VPS24), transcript variant 2 mRNA. 222449 a 7.8OE-06 4.1 TMEPAI transmembrane, prostate 20d 13.31-q13.33 androgen induced RNA (TMEPAI), transcript variant 3, mRNA. 2224.53 a 4.19E-05 -2.7 CYBRD1 cytochrome b reductase 1 2a31.1 (CYBRD1), mRNA. 222482 a 1.8SE-OS -2.4 SSBP3 Single stranded DNA binding 1p32.3 protein 3 222486 s at 5.8OE-06 -4.9 ADAMTS1 ADAM metallopeptidase with 21q21.2 thrombospondin type 1 motif, 1 (ADAMTS1), mRNA. 222488 s at 9.1OE-05 -2.4 DCTN4 dynactin 4 (p62) (DCTN4), 5q31-q32 mRNA. 222494 at 4.OOE-06 -2.8 CHES1 checkpoint Suppressor 1 14q24.3-q32.11 (CHES1), mRNA. 2225.03 s at 0.0003949 2.5 WDR41 WD repeat domain 41 5q13.3 (WDR41), mRNA. US 8,440,393 B2 109 110 TABLE 1-continued Gene expression profile. Affymetrix Parametric Fold Change Probe Set P-Value (Tumor? Normal) Gene Symbol Description Map 222533 at O.OOO1299 -2.1 CRBN cereblon (CRBN), mRNA. 3p26.2 2225.38 S. at 0.0007304 -3.2 APPL adaptor protein containing pH 3p21.1-p14.3 domain, PTB domain and leucine zipper motif 1 (APPL), mRNA. 222605 at 164E-05 -3.1 RCOR3 REST corepressor 3 (RCOR3), 1q32.3 mRNA. 222722 at 1.2SE-OS -8.7 OGN osteoglycin (osteoinductive 9q22 factor, mimecan) (OGN), transcript variant 3, mRNA. 222753 s at 0.0001335 2.1 SPCS3 signal peptidase complex 4q34.2 Subunit 3 homolog (S. cerevisiae) (SPCS3), mRNA. 222791 at 1.2OE-06 -3.4 RSBN1 round spermatid basic protein 1 1p13.2 (RSBN1), mRNA. 222834 S. at 0.0006523 -2.1 GNG12 guanine nucleotide binding 1p31.2 protein (G protein), gamma 12 (GNG12), mRNA. 222968 at S.98E-OS 2.9 C6orf248 chromosome 6 open reading 6p21.3 frame 48 (C6orf248), mRNA. 222975 S. at 9.41E-05 -2.9 CSDE1 cold shock domain containing 1p22 E1, RNA-binding (CSDE1), transcript variant 2, mRNA. 223007 s at 0.0007305 -2.7 C9orfs chromosome 9 open reading 9q31 frame 5 (C9orf5), mRNA. 223010 S. at 0.0005426 -2.2 OCLAD1 OCIA domain containing 1 4p11 (OCLAD1), mRNA. 223.01.1 s at 0.0008498 -2 OCLAD1 OCIA domain containing 1 4p11 (OCLAD1), mRNA. 223050 s at 0.0003276 2 FBXW5 F-box and WD-40 domain 9q34.3 protein 5 (FBXW5), transcript variant 1 mRNA. 223O82 at O.OOO3668 2.1 SH3KBP1 SH3-domain kinase binding Xp22.1-p21.3 protein 1 (SH3KBP1), transcript variant 1 mRNA. 223170 at O.OOO1428 -2.6 DKFZP564K1964 DKFZP564K1964 protein 17q11.2 (DKFZP564K1964), mRNA. 223189 x at 9.46E-05 -2.4 MLLS myeloid/lymphoid or mixed- 7q22.1 ineage leukemia 5 (trithorax homolog, Drosophila) (MLL5), mRNA. 223208 at O.OOO9547 2 KCTD10 potassium channel 12q24.11 etramerisation domain containing 10 (KCTD10), mRNA. 223227 at O.OOO6403 -2.2 BBS2 Bardet-Biedl syndrome 2 16q21 (BBS2), mRNA. 223263 s at 0.0003743 -2.1 FGFR1OP2 FGFR1 oncogene partner 2 12p11.23 (FGFR1OP2), mRNA. 223276 at S.1OE-OS 2.9 NID67 putative Small membrane protein 5q33.1 NID67 (NID67), mRNA. 223283 s at 0.0009337 -2.5 SDCCAG33 Serologically defined colon 18q22.3 cancer antigen 33 (SDCCAG33), mRNA. 223306 at 6.02E-OS -2.4 EBPL emopamil binding protein-like 13q12-q13 (EBPL), mRNA. 223366 at O.OOO847S -2.9 CDNA FLJ16218 fis, clone 8 CTONG3001501, highly similar to Mits musculus glucocorticoid induced gene 1 mRNA 223384 s at 0.0009265 -2.1 TRIM4 tripartite motif-containing 4 7q22-q31.1 (TRIM4), transcript variant beta, mRNA. 223395 at 9.3SE-OS -6.6 ABI3BP ABI gene family, member 3 3q12 (NESH) binding protein (ABI3BP), mRNA. 223437 at O.OOO4419 -2.2 PPARA peroxisome proliferative 22d 13.31 activated receptor, alpha (PPARA), transcript variant 3, mRNA. 223464 at O.OOO4952 2.3 OSBPLS oxysterol binding protein-like 5 11 p 15.4 (OSBPL5), transcript variant 2, mRNA. 223501 at O.OOO44O1 4.3 TNFSF13B Tumor necrosis factor (ligand) 13q32-34 Superfamily, member 13b US 8,440,393 B2 111 112 TABLE 1-continued Gene expression profile. Affymetrix Parametric Fold Change Probe Set P-Value (Tumor? Normal) Gene Symbol Description Map 223538 at O.OOO8098 2 SERF1A Small EDRK-rich factor 1A 5q12.2-q13.3 (telomeric) 223566 s at O.OOO7174 -2.7 BCOR BCL6 co-repressor (BCOR), Xp21.2-p11.4 transcript variant 1, mRNA. 223617 X at 3.01E-OS 2.6 ATAD3B ATPase family, AAA domain 1p36.33 containing 3B (ATAD3B), mRNA. 223629 at O.OOO8805 -2.4 PCDHB5 protocadherin beta 5 (PCDHB5), mRNA. 223672 at 4.8 SGIP1 SH3-domain GRB2-like 1p31.2 (endophilin) interacting protein 1 (SGIP1), mRNA. 223697 X at 4.6 C9orf64 chromosome 9 open reading frame 64 (C9orf64), mRNA. 223991 s at 2.42E-05 2.6 GALNT2 UDP-N-acetyl-alpha-D- galactosamine:polypeptide N acetylgalactosaminyltransferase 2 (GalNAc-T2) (GALNT2), mRNA. 224254 x at 2.71E-OS 5.5 TF Transferrin 3q22.1 224445 S. at 111E-05 -3 ZFYVE21 Zinc finger, FYVE domain 14q32.33 containing 21 (ZFYVE21), mRNA. 224.549 X at 6.5OE-06 8 224,598 at 3.68E-OS 2.4 MGAT4B mannosyl (alpha-1,3-)- glycoprotein beta-1,4-N- acetylglucosaminyltransferase, isoenzyme B (MGAT4B), transcript variant 1, mRNA. 224605 a. -2.2 LOC4O1152 HCV F-transactivated protein 1 4q26 (LOC4O1152), mRNA. 224612 S at 6.79E-05 2.2 DNAJCS DnaJ (Hsp40) homolog, 20q13.33 Subfamily C, member 5 224618 a O.OOO1475 2 ROD1 ROD1 regulator o 9q32 differentiation 1 (S. pombe) 224660 a O.OOO8567 -2.1 MGC14156 hypothetical protein MGC14156 4q22.1 (MGC14156), mRNA. 224664 a -2.3 C10orf104 chromosome 10 open reading 10q22.1 rame 104 (C10orf104), mRNA. 224665 a -2.2 C10orf104 chromosome 10 open reading 10q22.1 rame 104 (C10orf104), mRNA. 224667 x at 3.1 Transcribed locus 224689 a O.OOO1149 -2 MANBAL mannosidase, beta A, lysosomal 20d 11.23-q12 ike (MANBAL), transcript variant 2 mRNA. 224734 a O.OOO1511 -2.9 HMGB1 High-mobility group box 1 13q12 224741 x at O.OOO1144 -2.5 GASS Growth arrest-specific 5 1q23.3 224754 a O.OOO2O6 -2.2 SP1 Sp1 transcription factor (SP1), 12q13.1 mRNA. 224755 a. O.OOO1441 -2.2 SMBP SM-1104.4 binding protein 10q24.1 224763 a 5.3OE-06 -3.9 RPL37 ribosomal protein L37 (RPL37), 5p13 mRNA. 224780 a O.OOO8827 -2 RBM17 RNA binding motif protein 17 10p15.1 (RBM17), mRNA. 224812 a O.OOO1992 -2.3 HIBADH 3-hydroxyisobutyrate 7p15.2 dehydrogenase (HIBADH), mRNA. 224841 x at O.OOO1078 -2.6 RNU47 PREDICTED: RNA U47 small nuclear (RNU47), misc RNA. 224856 at O.OOO4439 -2.6 FKBP5 FK506 binding protein 5 6p21.3-21.2 (FKBP5), mRNA. 224893 at -2.8 DKFZPS64O863 DKFZP564JO863 protein 11 q13.1 224895 at -2.6 YAP1 Yes-associated protein 1, 65 kDa 11 q13 (YAP1), mRNA. 224901 at -3.5 SCD4 Stearoyl-CoA desaturase 5 224.950 at O.OOO9862 2.2 PTGFRN prostaglandin F2 receptor negative regulator (PTGFRN), mRNA. 224967 at 2.7 UGCG UDP-glucose ceramide glucosyltransferase 224970 at 5.88E-OS -2.5 NFIA Nuclear factor IFA 1p31.3-p31.2 225050 at 4.8SE-OS -2.5 ZNF512 Zinc finger protein 512 2p23 (ZNF512), mRNA. US 8,440,393 B2 113 114 TABLE 1-continued Gene expression profile. Affymetrix Parametric Fold Change Probe Set P-Value (Tumor? Normal) Gene Symbol Description Map 225,060 a O.OOO1496 -2.4 LRP11 ow density lipoprotein receptor- 6q25.1 related protein 11 (LRP11), mRNA. 225,078 a O.OOO2343 -3.4 EMP2 Epithelial membrane protein 2 16p13.2 225,098 a O.OOO1021 -2 ABI2 Abl interactor 2 2a33 225106 S. at 0.0005784 2.9 FLJ10826 hypothetical protein FLJ10826 16q12.2 (FLJ10826), transcript variant 2, mRNA. 225123 a 4.OOE-07 -3.1 SESN3 Sestrin 3 11q21 225.125 a. 3. SOE-OS -3.1 TMEM32 transmembrane protein 32 Xq26.3 (TMEM32), mRNA. 225132 a O.OOO1306 -2 FBXL3 F-box and leucine-rich repeat 13q22 protein 3 (FBXL3), mRNA. 225133 a 4.79E-05 -2.6 KLF3 Kruppel-like factor 3 (basic) 4p14 225147 a 1SOE-06 3.7 PSCD3 pleckstrin homology, Sect and 7p22.1 coiled-coil domains 3 (PSCD3), mRNA. 225162 a 1.4OE-05 -3.8 SH3D19 SH3 domain protein D19 4q31.3 (SH3D19), mRNA. 2251.79 a O.OOO1313 -2.1 HIP2 Huntingtin interacting protein 2 4p14 225198 a 2.56E-OS -2.7 WAPA VAMP (vesicle-associated 18p11.22 membrane protein)-associated protein A, 33 kDa (VAPA), transcript variant 2, mRNA. 225207 a O.OOO1018 -3.4 PDK4 pyruvate dehydrogenase kinase, 7q21.3-q22.1 isoenzyme 4 (PDK4), mRNA. 225219 a 2.7OE-OS -3.5 SMADS SMAD, mothers against DPP 5q31 homolog 5 (Drosophila) (SMAD5), transcript variant 3, mRNA. 225220 a O.OOO74 -2.2 Homo sapiens, clone 4 MAGE: 4249217, mRNA 225223 a 2.89E-OS -2 SMADS SMAD, mothers against DPP 5q31 homolog 5 (Drosophila) (SMAD5), transcript variant 3, mRNA. 225239 a O.OOO4432 4.3 mmunoglobulin light chain 11 variable region 225243 s at 0.0001.227 -2.6 SLMAP sarcolemma associated protein 3p21.2-p14.3 (SLMAP), mRNA. 225274 a 2.8OE-06 -3 SNRPG Small nuclear ribonucleoprotein 2p13.3 polypeptide G 225310 a O.OOO177 -2.2 RBMX RNA binding motif protein, X- Xq26.3 inked 225326 a 8.88E-OS -2 RBM27 PREDICTED: RNA binding 5 motif protein 27 (RBM27), mRNA. 225.330 a. O.OOO2697 -2.1 IGF1R insulin-like growth factor 1 15q26.3 receptor 225332 a. 5.78E-OS -2.1 KRTAP4-7 Keratin associated protein 4-7 17q12-q21 225344 a O.OOO4132 2.8 NCOA7 nuclear receptor coactivator 7 6q22.32 (NCOA7), mRNA. 225352 a O.OOO1828 -2.3 TLOC1 translocation protein 1 3q26.2 (TLOC1), mRNA. 225381 a O.OOO663S -4.3 LOC399959 PREDICTED: hypothetical 11 LOC399959 (LOC399959), mRNA. 225387 a 3.43E-OS -3.2 TM4SF9 Tetraspanin 5 4q23 225416 a O.OOO3496 -2.2 RNF12 Ring finger protein 12 Xq13-q21 225421 a O.OOO9226 -3.1 ACY1L2 aminoacylase 1-like 2 6q15 (ACY1 L2), mRNA. 225.426 a 6.78E-OS -2.5 PPP6C Protein phosphatase 6, catalytic 9q33.3 Subunit 225480 a O.OOO2718 2.2 C1orf122 chromosome 1 open reading 1p34.3 rame 122 (C1orf122), mRNA. 225489 a 3.71E-OS -2 TMEM18 transmembrane protein 18 2p25.3 (TMEM18), mRNA. 225498 a 7.4SE-OS -2.2 CHMP4B chromatin modifying protein 4B 20d 11.22 (CHMP4B), mRNA. 2255.05 s at 1.90E-06 4.1 C20orf&1 chromosome 20 open reading 20p13 frame 81 (C20orf31), mRNA. 225509 a 101E-05 -2.9 SAP3OL Hypothetical protein LOC56757 5q33.2 2255.24 a S.OOE-OS -3.1 ANTXR2 anthrax toxin receptor 2 4q21.21 (ANTXR2), mRNA. US 8,440,393 B2 115 116 TABLE 1-continued Gene expression profile. Affymetrix Parametric Fold Change Probe Set P-Value (Tumor? Normal) Gene Symbol Description Map 225526 a O.OOO4149 -2 MKLN1. muskelin 1, intracellular 7q32 mediator containing kelch motifs (MKLN1), mRNA. 225546 a 4.45E-05 -2.7 EEF2K Eukaryotic elongation factor-2 16p12.1 kinase 225571 a O.OOO188 -4.7 LIFR leukemia inhibitory factor 5p13-p12 receptor (LIFR), mRNA. 225574 a 8.57E-OS -2.4 MGC101.98 hypothetical protein MGC10198 4q35.1 (MGC10198), mRNA. 225575 a. O.OOO3237 -4.5 LIFR leukemia inhibitory factor 5p13-p12 receptor (LIFR), mRNA. 225611 a O.OOO4119 -2.6 MAST4 Microtubule associated 5q12.3 serine threonine kinase family member 4 2256.26 a O.OOO3153 4.8 PAG1 phosphoprotein associated with 8q21.13 glycosphingolipid microdomains 1 (PAG1), mRNA. 225636 a O.OOO278 2.5 STAT2 signal transducer and activator 12q13.3 of transcription 2, 113 kDa (STAT2), mRNA. 225646 a 1. SOE-OS 4.4 CTSC cathepsin C (CTSC), transcript 11 q14.1-q14.3 variant 2 mRNA. 225647 s at 0.0009077 2.5 CTSC cathepsin C (CTSC), transcript 11 q14.1-q14.3 variant 1 mRNA. 225686 a O.OOO3333 2.2 FAM33A family with sequence similarity 17q23.2 33, member A (FAM33A), mRNA. 225698 a 3.41E-05 -3.2 TIGA1 TIGA1 (TIGA1), mRNA. 5q21-q22 225728 a O.OOO6083 -2.8 ARGBP2 Arg/Abl-interacting protein 4q35.1 ArgBP2 225793 a O.OOO103 -2.1 MGC46719 Lix1 homolog (mouse) like 1q21.1 225799 a 2.29E-OS 5.2 MGC4677 hypothetical protein MGC4677 2p11.2 (MGC4677), mRNA. 225811 a 5.44E-05 -2.1 Transcribed locus, weakly 11 similar to XP 510104.1 PREDICTED: Similar to hypothetical protein FLJ25224 Pan troglodytes 225845 at O.OOO3774 -3.4 BTBD15 BTB (POZ) domain containing 11q24.3 5 (BTBD15), mRNA. 225855 at 7.78E-05 -2.4 EPB41LS erythrocyte membrane protein 214.2 band 4.1 like 5 (EPB41L5), mRNA. 225886 at O.OOOSO86 -2.2 DDX5 RNA-binding protein 45 17q21 (RBP45), putative 225915 at 3. SOE-OS -3.4 CAB39L. Calcium binding protein 39-like 13q14.2 225939 at O.OOO1338 -2.8 EIF4E3 Eukaryotic translation initiation 3p14 actor 4E member 3 225941 at O.OOO2862 -2.2 EIF4E3 Eukaryotic translation initiation 3p14 actor 4E member 3 225946 at O.OOO3S11 -2.3 C12Orf2 Chromosome 12 open reading 12p12.3 rame 2 225947 at 5.75E-05 2.3 MYOHD1 myosin head domain containing 17q12 (MYOHD1), mRNA. 225967 s at 0.0001375 2.5 LOC2841 84 PREDICTED: hypothetical 17 LOC284.184 (LOC284.184), mRNA. 225976 at O.OOO3271 -2 BTF3L4 basic transcription factor 3-like 1p32.3 4 (BTF3L4), mRNA. 225987 at O.OOO3S03 3.3 TNFAIP9 STEAP family member 4 7q21.12 225996 at O.OOO6306 -9.3 MRNA, cDNA 2 DKFZp686N1345 (from clone DKFZp686N1345) 226017 at O.OOO3O1 2.3 CKLFSF7 chemokine-like factor 3p23 superfamily 7 (CKLFSF7), transcript variant 2, mRNA. 226020 S. at 7.89E-05 -2.2 OMA1 OMA1 homolog, Zinc 1p32.2-p32.1 metallopeptidase (S. cerevisiae) (OMA1), mRNA. 226038 at O.OOO2389 -3.2 LONRF1 LON peptidase N-terminal 8p23.1 domain and ring finger 1 (LONRF1), mRNA. 226063 at 4.OOE-06 3.3 WAV2 vav 2 oncogene (VAV2), 9q34.1 mRNA. US 8,440,393 B2 117 118 TABLE 1-continued Gene expression profile. Affymetrix Parametric Fold Change Probe Set P-Value (Tumor? Normal) Gene Symbol Description Map 226066 a O.OOOS179 -3 MITF microphthalmia-associated 3p 14.2-p14.1 transcription factor (MITF), transcript variant 5, mRNA. 226117 a 8.72E-OS -2.7 TIFA TRAF-interacting protein with a 4q25 forkhead-associated domain (TIFA), mRNA. 226120 a O.OOO1436 -2.7 TTC8 tetratricopeptide repeat domain 14q31.3 8 (TTC8), transcript variant 3, mRNA. 226180 a O.OOO9691 -2 WDR36 WD repeat domain 36 5q22.1 (WDR36), mRNA. 226184 a 3.75E-05 -3.6 FMNL2 formin-like 2 (FMNL2), 2d23.3 transcript variant 2, mRNA. 226203 a O.OOO3968 -2 CDNA clone IMAGE: 5299888 15 226223 a O.OOO2488 -2.6 PAWR PRKC, apoptosis, WT1, 12q21 regulator 226225 a. O.OOO3786 -3.2 MCC Mutated in colorectal cancers 5q21-q22 226230 a. O.OOO2S89 -2.4 KIAA1387 KIAA1387 protein 2p16.1 226280 a 3.34E-OS -2.8 BNIP2 BCL2 adenovirus E1B 19 kDa 15q22.2 interacting protein 2 226297 a 8.43E-OS -2.1 ESTs 226303 a O.OOO3656 -2.9 PGMS phosphoglucomutase 5 (PGM5), 9q13 mRNA. 226336 a O.OOO1456 -2 PPLA Peptidylprolyl isomerase A 7p13-p11.2 (cyclophilin A) 226344 a O.OOO9627 -2.9 ZMAT1 Zinc finger, matrin type 1 Xq21 (ZMAT1), transcript variant 1, mRNA. 226403 a 9.2OE-06 3.8 TMC4 transmembrane channel-like 4 19q13.42 (TMC4), mRNA. 226472 a O.OOO7652 -2 PPIL4 peptidylprolyl isomerase 6q24-q25 (cyclophilin)-like 4 (PPIL4), mRNA. 226484 a O.OOO9077 2 ZNF651 Zinc finger protein 651 3p22.1 (ZNF651), mRNA. 226499 a O.OOO9453 2.8 MGC61598 Similar to ankyrin-repeat protein 9q34.3 Nrarp 226521 s at 0.0006084 -2.1 FLJ13614 hypothetical protein FLJ13614 4q21.21-q21.23 (FLJ13614), mRNA. 226529 a 4.8SE-OS -2.3 FLJ11273 hypothetical protein FLJ11273 7p21.3 (FLJ11273), mRNA. 226541 a O.OOO1172 -2.3 FBXO3O F-box protein 30 (FBXO30), 6q24 mRNA. 226561 a O.OOO6O2S -2.1 LOC28SO86 Hypothetical protein 2a36.3 LOC28SO86 226599 a O.OOO2151 2.3 KIAA1727 KIAA1727 protein 4q31.3 (KLAA1727), mRNA. 226625 a. 7.3OE-06 -3.8 TGFBR3 Transforming growth factor, 1p33-p32 beta receptor III (betaglycan, 300 kDa) 226663 a 8.6OE-06 2.8 ANKRD10 Ankyrin repeat domain 10 13q34 226668 a 0.0003707 -2.6 WDSUB1 WD repeat, SAM and U-box 2d24.2 domain containing 1 (WDSUB1), mRNA. 226688 a O.OOO2101 -3.3 C3orf23 chromosome 3 open reading 3p21.33-p21.32 frame 23 (C3orf23), transcript variant 1 mRNA. 226695 a O.OOO1301 2.5 PRRX1 paired related homeobox 1 1q24 (PRRX1), transcript variant pmx-1b, mRNA. 226705 a O.OOO2111 -2 FGFR1 Fibroblast growth factor 8p11.2-p11.1 receptor 1 (fms-related tyrosine kinase 2, Pfeiffer syndrome) 226713 a 181E-05 -2.9 C3orf6 Chromosome 3 open reading 3 frame 6 226747 a O.OOO6894 -3.7 KIAA1344 KIAA1344 (KIAA1344), 14q22.1 mRNA. 226751 a 8.3OE-06 -2.4 C2Orf32 chromosome 2 open reading 2p14 frame 32 (C2orf32), mRNA. 226765 a O.OOO2842 2.2 SPTBN1 Spectrin, beta, non-erythrocytic 1 2p21 226777 a 5.39E-OS 4.9 ADAM12 A disintegrin and 10q26.3 metalloproteinase domain 12 (meltrin alpha) US 8,440,393 B2 119 120 TABLE 1-continued Gene expression profile. Affymetrix Parametric Fold Change Probe Set P-Value (Tumor? Normal) Gene Symbol Description Map 2268.06 s at 1.95E-05 -2.8 MRNA, cDNA 1 DKFZp686.J23256 (from clone DKFZp686.J23256) 226829 at O.OOO6675 2.9 KIAA1914 KIAA1914 (KIAA1914), 10q25.3 transcript variant 2, mRNA. 226867 at O.OOO8554 -2 C9orf SS Chromosome 9 open reading 9p22.1 frame 55 226873 at O.OOO2O67 -2.8 Transcribed locus 16 226899 at 3.1OE-06 3.1 UNCSB unc-5 homolog B (C. elegans) 10q22.1 (UNCSB), mRNA. 226909 at O.OOO4528 -2 KIAA1729 KIAA1729 protein 4p16.1 (KLAA1729), mRNA. 226911 at 1.2OE-05 3.7 FLJ39155 hypothetical protein FLJ39155 5p13.2-p13.1 (FLJ39155), transcript variant 4, mRNA. 226933 s at 5.79E-05 3.8 ID4 inhibitor of DNA binding 4, 6p22-p21 dominant negative helix-loop helix protein (ID4), mRNA. 226943 a O.OOO4664 2 MRNA, cDNA DKFZp547P055 12 (from clone DKFZp547P055) 226994 a O.OOO2644 -2.3 DNAJA2 DnaJ (Hsp40) homolog, 16q11.1-q11.2 Subfamily A, member 2 226997 a O.OOO2372 3.7 CDNA FLJ10196 fis, clone 5 HEMBA1004776 227031 a O.OOO1825 -2.4 SNX13 Sorting nexin 13 7p21.1 227070 a O.OOO3008 -2.7 GLT8D2 glycosyltransferase 8 domain 12q containing 2 (GLT8D2), mRNA. 227082 a O.OOO7234 -2 MRNA, cDNA 3 DKFZp586K1922 (from clone DKFZp586K1922) 227121 a O.OOO6088 -2.1 MRNA, cDNA 3 DKFZp586K1922 (from clone DKFZp586K1922) 227132 a. O.OOO2728 -2.4 LOCS1123 HSPCO38 protein (LOC51123), 8q22.3 mRNA. 227138 a O.OOO8198 -2.1 CRTAP cartilage associated protein 3p22.3 (CRTAP), mRNA. 227148 a O.OOO1511 -3.4 PLEKHH2 pleckstrin homology domain 2p21 containing, family H (with MyTH4 domain) member 2 (PLEKHH2), mRNA. 2271 78 a O.OOOS124 -3.9 CUGBP2 CUG triplet repeat, RNA 10p13 binding protein 2 (CUGBP2), transcript variant 2, mRNA. 227197 a O.OOO108 -2.9 SGEF Src homology 3 domain- 3q25.2 containing guanine nucleotide exchange factor 227214 a O.OOO2O58 2.1 GOPC Golgi associated PDZ and 6q21 coiled-coil motif containing 227221 a. O.OOO4345 2.1 CDNA FLJ31683 fis, clone 3 NT2RI2005.353 227260 a O.OOO1678 3 Transcribed locus 1 227 273 a 4.47E-OS -2.6 Transcribed locus 10 227 278 a O.OOOS386 2.3 Transcribed locus, weakly 1 similar to XP 510104.1 PREDICTED: Similar to hypothetical protein FLJ25224 Pan troglodytes 227293 at 8.47E-OS -2.2 LNX Ligand of numb-protein X 4q12 227295 at 2.7OE-06 3.4 IKIP IKK interacting protein (IKIP), 12q23.1 transcript variant 3.1, mRNA. 227317 at O.OOO3721 2.3 LMCD1 LIM and cysteine-rich domains 3p26-p24 1 (LMCD1), mRNA. 227347 X at 2.86E-05 4 HES4 hairy and enhancer of split 4 1p36.33 (Drosophila) (HES4), mRNA. 227372 s at 0.0004102 3.4 BALAP2L1 BAI1-associated protein 2-like 1 7q21.3-q22.1 (BAIAP2L1), mRNA. 227383 at O.OOOS636 2.4 Similar to KIAA0454 protein 1q21.1 227384 s at 0.000128 2.6 Similar to KIAA0454 protein 1q21.1 227396 at O.OOO1476 2.6 Homo sapiens, clone 11 IMAGE: 4454331, mRNA 2274.07 at 9.42E-05 -2.3 FLJ90013 hypothetical protein FLJ90013 4p15.32 (FLJ90013), mRNA. 227529 S. at 0.0001731 -2.9 AKAP12 A kinase (PRKA) anchor protein 6q24-q25 (gravin) 12 US 8,440,393 B2 121 122 TABLE 1-continued Gene expression profile. Affymetrix Parametric Fold Change Probe Set P-Value (Tumor? Normal) Gene Symbol Description Map 227530 at 8.47E-OS -3 AKAP12 A kinase (PRKA) anchor protein 6q24-q25 (gravin) 12 227636 at O.OOO3391 -2.3 THAP5 THAP domain containing 5 7q31.1 227703 s at 1.29E-05 -5.9 SYTL4 synaptotagmin-like 4 Xq21.33 (granuphilin-a) (SYTL4), mRNA. 227708 at O.OOO6098 -2.1 EEF1A1 Eukaryotic translation 6q14.1 elongation factor 1 alpha 1 227719 at 4.OOE-06 -3.6 CDNA FLJ37828 fis, clone 13 BRSSN2OO6575 227728 at 6.11E-OS -2.9 PPM1A Protein phosphatase 1A 14q23.1 (formerly 2C), magnesium dependent, alpha isoform 227827 at O.OOO8248 -4.5 ARGBP2 Arg/Abl-interacting protein 4q35.1 ArgBP2 227850 X. at 0.0001134 2.9 CDC42EP5 CDC42 effector protein (Rho 19q13.42 GTPase binding) 5 (CDC42EP5), mRNA. 227866 a O.OOO7379 -2 RBM16 RNA binding motif protein 16 6q25.1-q25.3 227945 a. O.OOO3107 -2.3 TBC1D1 TBC1 (tre-2/USP6, BUB2, 4p14 codc16) domain family, member 1 (TBC1D1), mRNA. 227952 a. O.OOO1513 4.1 Full length insert cDNA clone 4 Y46GO4 227971 a 3.26E-OS -4.1 NRK Nik related kinase (NRK), Xq22.3 mRNA. 228O12 a O.OOO6848 -2 MATR3 Matrin 3 5q31.2 228.027 a 3.77E-05 -4 GPRASP2 G protein-coupled receptor Xq22.1 associated sorting protein 2 (GPRASP2), transcript variant 2, mRNA. 228030 a. O.OOOS637 3 RBM6 RNA binding motif protein 6 3p21.3 228.098 s at 0.0009902 2.2 MYLIP myosin regulatory light chain 6p23-p22.3 interacting protein (MYLIP), mRNA. 228202 a O.OOO1345 -9.3 PLN Phospholamban 6q22.1 228204 a 1.OOE-07 2.9 PSMB4 proteasome (prosome, 1q21 macropain) subunit, beta type, 4 (PSMB4), mRNA. 228253 a 1.1OE-05 3.8 LOXL3 lysyl oxidase-like 3 (LOXL3), 2p13 mRNA. 228297 a O.OOO4293 3.3 CNN3 calponin 3, acidic (CNN3), 1p22-p21 mRNA. 228310 a 8.1OE-06 -3.3 ENAH enabled homolog (Drosophila) 1q42.12 (ENAH), transcript variant 2, mRNA. 228331 a O.OOO2123 3.4 C11orf1 Chromosome 11 open reading 11 q12.1 frame 31 228333 a O.OOOS4O7 -2.2 Full length insert cDNA clone 2 YT94EO2 228335 a. O.OOO878S -3.8 CLDN11 claudin 11 (oligodendrocyte 3q26.2-q26.3 transmembrane protein) (CLDN11), mRNA. 228497 a O.OOO1644 3.2 SLC22A15 Solute carrier family 22 (organic 1p13.1 cation transporter), member 15 (SLC22A15), mRNA. 228523 a O.OOO4047 2 NANOS1 nanos homolog 1 (Drosophila) 10q26.11 (NANOS1), transcript variant 2, mRNA. 228551 a 3.9SE-OS -3.1 MGC24039 Hypothetical protein 12p11.21 MGC24039 228554 a 1.35E-05 -4.7 MRNA, cDNA 11 DKFZp586GO321 (from clone DKFZp586G0321) 228569 a O.OOO681S -2.2 PAPOLA Poly(A) polymerase alpha 14q32.31 228573 a O.OOO1917 -2.5 ANTXR2 Anthrax toxin receptor 2 4q21.21 228577 x at 0.0003076 2.8 ODF2L outer dense fiber of sperm tails 1p22.3 2-like (ODF2L), transcript variant 2 mRNA. 228579 a 7.OOE-07 5.2 KCNQ3 Potassium voltage-gated 8q24 channel, KQT-like subfamily, member 3 228785 a. O.OOO6036 2.2 ZNF281 Zinc finger protein 281 1q32.1 228805 a O.OOO3526 -3 FLJ44216 FLJ44216 protein (FLJ44216), 5q.35.2 mRNA. US 8,440,393 B2 123 124 TABLE 1-continued Gene expression profile. Affymetrix Parametric Fold Change Probe Set P-Value (Tumor? Normal) Gene Symbol Description Map 228841 at O.OOO4789 -2 CDNA FLJ32429 fis, clone 5 SKMUS2OO1014 228.850 s at 0.0005294 2.8 SLIT2 Slithomolog 2 (Drosophila) 4p15.2 228885 at 182E-05 -4 MAMDC2 MAM domain containing 2 9q21.11 (MAMDC2), mRNA. 228905 at O.OOO1087 -2.4 Transcribed locus, moderately 8 similar to XP 517655.1 PREDICTED: Similar to KIAAO825 protein Pan troglodytes 228961 a O.OOO4391 -2.1 FLJ35954 Hypothetical protein FLJ35954 5q11.2 22.9085 a O.OOO3123 2.5 LRRC3B eucine rich repeat containing 3p24 3B (LRRC3B), mRNA. 229114 a O.OOO2683 -2 GAB1 GRB2-associated binding 4q31.21 protein 1 229119 S at 0.0001491 -2.2 TTC19 Hypothetical protein 17p12 LOC1251SO 229129 a O.OOO6968 -2.2 HNRPD Heterogeneous nuclear 4q21.1-q21.2 ribonucleoprotein D (AU-rich element RNA binding protein 1, 37 kDa) 229130 a. O.OOO2O67 -3 LOC285535 Hypothetical protein 4p16.1 LOC285535 229145 a. 4.45E-05 -4.2 C10orf104 chromosome 10 open reading 10q22.1 rame 104 (C10orf104), mRNA. 229160 a O.OOO3SS2 -6.5 MUM1L1 melanoma associated antigen Xq22.3 (mutated) 1-like 1 (MUM1L1), mRNA. 229200 a O.OOO156 2.3 Hypothetical LOC400813 1q44 229204 a O.OOO3OO7 2.2 HP1-BP74 Heterochromatin protein 1, 1p36.12 binding protein 3 229218 a O.OOO3816 3.3 COL1A2 Collagen, type I, alpha 2 7q22.1 229221 a O.OOO6672 4.1 CD44 CD44 antigen (homing function 11 p.13 and Indian blood group system) 229287 a O.OOO7653 -2.8 Full-length cDNA clone 14 CSODKO1OYA20 of HeLa cells Cot 25-normalized of Homo Sapiens (human) 229299 a O.OOO1103 -2.2 FLJ30596 hypothetical protein FLJ30596 5p13.2 (FLJ30596), mRNA. 229.308 a 2.94E-05 -5.4 Transcribed locus 18 229319 a O.OOO1521 -2.4 Homo sapiens, clone 6 MAGE: 4105966, mRNA 229331 a 0.0007515 2.3 SPATA18 spermatogenesis associated 18 4q11 homolog (rat) (SPATA18), mRNA. 229339 a O.OOO2425 -5 MYOCD Myocardin 7p11.2 229354 a O.OOOS/8 3.3 PDCD6 Aryl-hydrocarbon receptor 5pter-p15.2 repressor 2294.31 a O.OOO216 -2.1 RFXAP regulatory factor X-associated 3a14 protein (RFXAP), mRNA. 2294.83 a O.OOO2012 2.5 UBE2H Ubiquitin-conjugating enzyme 7q32 E2H (UBC8 homolog, yeast) 229515 a. O.OOO2226 -2.2 PAWR PRKC, apoptosis, WT1 2d21 regulator 229520 S. at 0.0001174 2.3 C14orf118 Chromosome 14 open reading 4q22.1-q24.3 frame 118 229531 a O.OOO7976 -2.2 Mitochondrial carrier triple Xq22.2 repeat 6 229553 a O.OOO8476 2 PGM2L1 phosphoglucomutase 2-like 1 1a13.4 (PGM2L1), mRNA. 2295.80 a O.OOO8828 -6.5 Transcribed locus 3 229638 a O.OOO8539 2.9 IRX3 iroquois homeobox protein 3 6d12.2 (IRX3), mRNA. 229642 a. O.OOO2914 4.2 ARHGEF7 Rho guanine nucleotide 3d34 exchange factor (GEF) 7 (ARHGEF7), transcript variant 2, mRNA. 229665 a. O.OOO1871 2.2 CSTF3 Hypothetical protein 1p13 LOC283267 229711 S. at 0.0003052 2.1 CDNA FLJ37519 fis, clone 2 BRCAN2OO4699 229748 x at 0.0001962 2.7 LOC285.458 Hypothetical protein 4 LOC285.458 US 8,440,393 B2 125 126 TABLE 1-continued Gene expression profile. Affymetrix Parametric Fold Change Probe Set P-Value (Tumor? Normal) Gene Symbol Description Map 229795 a O.OOO2666 3.4 Transcribed locus 12 229801 a O.OOO1036 3.4 C10orfaf chromosome 10 open reading 10p14 rame 47 (C10orfA7), mRNA. 2298.30 a. O.OOO4695 3.9 PDGFA Platelet-derived growth factor 7p22 alpha polypeptide 229844 a O.OOO1747 -2.2 Transcribed locus 3 229891 x at 1.34E-05 -2.9 KIAA1704 KIAA1704 13q13-q14 229969 a O.OOO3327 -2.5 Transcribed locus, moderately 6 similar to XP 50823.0.1 PREDICTED: zinc finger protein 195 Pan troglodytes 229994 a O.OOO2738 -3 MRNA, cDNA 1 DKFZp686.J23256 (from clone DKFZp686.J23256) 230000 a O.OOO4792 2.7 C17orf27 Chromosome 17 open reading 17q25.3 rame 27 230030 a. 0.0003377 -5.2 HS6ST2 heparan sulfate 6-O- Xq26.2 Sulfotransferase 2 (HS6ST2), mRNA. 23.0068. S. at 8.34E-05 -3.2 PEG3 paternally expressed 3 (PEG3), 19q13.4 mRNA. 230077 a O.OOO1004 3.9 TFRC Transferrin receptor (p90, 3q29 CD71) 230081 a S.OOE-07 -6.1 PLCXD3 phosphatidylinositol-specific 5p13.1 phospholipase C, X domain containing 3 (PLCXD3), mRNA. 230130 a. 9.27E-OS 3 SLIT2 Slithomolog 2 (Drosophila) 4p15.2 23O141 a O.OOO2O42 -2 ARID4A AT rich interactive domain 4A 14q23.1 (RBP1-like) 230174 a O.OOO8233 -2.1 LYPLAL1 Lysophospholipase-like 1 1q41 230178 s at 9.98E-05 -2.4 STATIP1 Signal transducer and activator 18q12.2 of transcription 3 interacting protein 1 230270 a 7.57E-05 2.9 ESTs 230333 a O.OOO8148 2.5 SAT Spermidine?spermine N1- Xp22.1 acetyltransferase 23O336 a O.OOO8645 2.7 Transcribed locus 4 230369 a 0.0005567 -2.1 GPR161 G protein-coupled receptor 161 1q24.2 230387 a O.OOO1569 3.1 ATP2C1 ATPase, Ca++ transporting, type 3q22.1 2C, member 1 230440 a 2.81E-05 3.3 ZNF469 PREDICTED: zinc finger 16 protein 469 (ZNF469), mRNA. 230561 S. at 0.0001454 -2.9 FLU23861 hypothetical protein FLJ23861 2a34 (FLJ23861), mRNA. 230574 a O.OOO80S6 2.9 Hypothetical LOC388480 18q21.33 230746 s at 1.5OE-06 13.3 STC1 stanniocalcin 1 (STC1), mRNA. 8p21-p11.2 230758 a O.OOO64 -2.6 Transcribed locus X 230793 a 7.34E-OS -3.1 LRRC16 leucine rich repeat containing 16 6p22.2 (LRRC16), mRNA. 230850 a O.OOO1807 3 FMNL3 Formin-like 3 12q13.12 230885 a O.OOO1271 2.2 SPG7 Spastic paraplegia 7, paraplegin 16q24.3 (pure and complicated autosomal recessive) 230958. S. at 4.15E-05 -2.1 MRNA, cDNA 1 DKFZp686.J23256 (from clone DKFZp686.J23256) 231130 at O.OOO9273 -2 FKBP7 FK506 binding protein 7 2a31.2 231183 s at 0.000129 4.3 JAG1 jagged 1 (Alagile syndrome) 20p12.1-p11.23 (JAG1), mRNA. 231202 at O.OOO2962 3.1 FLJ38508 Aldehyde dehydrogenase 1 12q23.3 family, member L2 231411 at O.OOO1224 3.2 LHFP lipoma HMGIC fusion partner 13q12 (LHFP), mRNA. 231597 X at 0.0001876 15.9 ESTs, Weakly similar to T47135 hypothetical protein DKFZp761L0812.1 H. sapiens 231806 s at 0.0009603 2.1 STK36 serine threonine kinase 36 2a35 (fused homolog, Drosophila) (STK36), mRNA. 231825 X at 8.13E-05 3.6 ATF7IP Activating transcription factor 7 12p13.1 interacting protein 231882 at O.OOO2463 2.7 CDNA FLJ10674 fis, clone 22 US 8,440,393 B2 127 128 TABLE 1-continued Gene expression profile. Affymetrix Parametric Fold Change Probe Set P-Value (Tumor? Normal) Gene Symbol Description Map 231886 at O.OOO3613 3 Similar to hypothetical protein 3q29 LOC2847O1 232034 at O.OOO816S 3.9 LOC2O3274 Hypothetical protein 9q21.11 LOC2O3274 232145 at O.OOO2S39 2.1 LOC38.8969 hypothetical LOC38.8969 2p11.2 (LOC38.8969), mRNA. 232150 at 2.29E-OS 3.1 C20orf18 Chromosome 20 open reading 20p13 rame 18 232169 X at 0.0002893 3.1 NDUFS8 NADH dehydrogenase 11 q13 (ubiquinone) Fe—S protein 8, 23 kDa (NADH-coenzyme Q reductase) 232174 a 7.OOE-06 4.1 EXT1 Exostoses (multiple) 1 8q24.11-q24.13 232180 a 8.56E-OS 3.2 UGP2 UDP-glucose 2p14-p13 pyrophosphorylase 2 232215 X at 0.0001855 3.4 FLJ11029 Hypothetical protein FLJ11029 17q23.2 232254 a 8.7SE-OS 2.5 FBXO25 F-box protein 25 8p23.3 232266 X at 0.0001426 3.9 CDC2LS Cell division cycle 2-like 5 7p13 (cholinesterase-related cell division controller) 232304 a 2.6OE-06 4 PELI1 Pellino homolog 1 (Drosophila) 2p13.3 232347 X at 0.0001636 2.6 CBR4 Carbonic reductase 4 4q32.3 232406 a O.OOO1583 2.9 JAG1 agged 1 (Alagile syndrome) 20p12.1-p11.23 232458 a 1.67E-OS 4.1 COL3A1 Collagen, type III, alpha 1 2a31 (Ehlers-Danlos syndrome type V, autosomal dominant) 232516 X at 7.53E-05 3.3 YAP YY1 associated protein 1 1q22 232530 a. O.OOO6114 2.8 PLD1 Phospholipase D1, 3q26 phophatidylcholine-specific 232538 a O.OOO8.933 2.4 CDNA: FLJ23573 fis, clone 16 LNG12S2O 232541 a O.OOO7593 3.5 EGFR Epidermal growth factor 7p12 receptor (erythroblastic eukemia viral (v-erb-b) oncogene homolog, avian) 232617 a 8.18E-OS 3.2 CTSS cathepsin S (CTSS), mRNA. 1q21 232653 a O.OOO473 3.5 Homo sapiens cDNA FLJ14044 is, clone HEMBA1006124 232702 a O.OOO6941 3 RABGAPIL RAB GTPase activating protein 1q24 -like 232797 a O.OOO8213 2.3 ITGAV integrin, alpha V (vitronectin 2a31-q32 receptor, alpha polypeptide, antigen CD51) 232814 x at 0.0005021 2.8 C14orf153 Chromosome 14 open reading 14q32.32-q32.33 rame 153 232889 at O.OOO3899 2.6 CDNA clone IMAGE:5576908 5 232952 at O.OOO1405 2.7 DDEF1 HSPCO54 protein 8q24.1-q24.2 233041 x at 0.0004112 3 BTBD9 BTB (POZ) domain containing 9 6p21 233180 at 1.17E-OS 2.8 RNF152 Ring finger protein 152 18q21.33 233274 at 3.OOE-07 3.6 NCK1 NCK adaptor protein 1 3q21 233319 X at 0.0001543 3.1 PHACTR4 Phosphatase and actin regulator 4 1p35.3 233330 S. at 4.59E-05 S.6 Similar to Ribosome biogenesis 9q13 protein BMS1 homolog 233406 at O.OOO2846 2.9 KIAAO256 KIAAO256 gene product 15q21.1 233496 s at 0.0003586 -2 CFL2 Cofilin 2 (muscle) 14q12 233702 x at 0.0007629 3.1 CDNA: FLJ20946 fis, clone 7 ADSEO1819 233849 s at 0.0005935 -2.3 ARHGAPS Rho GTPase activating protein 5 14q12 (ARHGAPS), transcript variant 2, mRNA. 233912 X at 1.73E-05 4.4 ELMOD2 ELMO domain containing 2 4q31.21 234.192 s at 0.0001567 -3.1 GKAP1 G kinase anchoring protein 1 9q21.32 (GKAP1), mRNA. 234.339 S at 6.34E-05 -2.4 GLTSCR2 glioma tumor Suppressor 19q13.3 candidate region gene 2 (GLTSCR2), mRNA. 234464 S at 4.21E-05 3 EME1 essential meiotic endonuclease 1 17q21.33 homolog 1 (S. pombe) (EME1), mRNA. 234512 X at 2.4OE-06 -3.1 LOC4421.59 PREDICTED: similar to Rp17a 6 protein (LOC442159), mRNA. 234562 X at 1.80E-06 8.6 CKLFSF8 Chemokine-like factor super 3p23 family 8 US 8,440,393 B2 129 130 TABLE 1-continued Gene expression profile. Affymetrix Parametric Fold Change Probe Set P-Value (Tumor? Normal) Gene Symbol Description Map 234578 at 7.26E-OS 4.9 MRNA, cDNA 1 DKFZp434E1812 (from clone DKFZp434E1812) 234675 X at 6.52E-05 4.3 CDNA: FLJ23566 fis, clone 14 LNG1088O 234.723 X at 1.5OE-06 9.4 CDNA: FLJ21228 fis, clone 7 COLOO739 234753 x at 0.0002577 5.7 234762 X at 8.75E-05 4.1 NLN Neurolysin (metallopeptidase 5q12.3 M3 family) 234788 X at 0.0001541 3.3 FLJ13611 Hypothetical protein FLJ13611 5q12.3 234873 X at 0.000607 -2 234981 X at 9.85E-05 4.4 LOC134147 Similar to mouse 5p15.2 2310016AO9Rik gene 234985 at 9.3OE-06 2.9 LOC143458 Hypothetical protein 11 p.13 LOC143458 234998 at O.OOO9467 -2.7 CDNA clone IMAGE: 5313062 15 235.005 at O.OOO3344 -2.3 MGC4562 Hypothetical protein MGC4562 15q22.31 235061 at O.OOO2578 -2.9 PPM1 K protein phosphatase 1 K (PP2C 4q22.1 domain containing) (PPM1K), mRNA. 235072 S at 6.66E-05 -3 Transcribed locus 6 235122 at S.42E-OS 3.1 CDNA clone IMAGE: 6254.031 1 235151 at O.0004926 -2.1 LOC283357 Hypothetical protein 12p13.33 LOC283357 235204 at 1.87E-OS 3.1 ENTPD7 Ectonucleoside triphosphate 10 diphosphohydrolase 7 235205 at O.OOO12S2 3.5 LOC346887 PREDICTED: Similar to Solute 8 carrier family 16 (monocarboxylic acid transporters), member 14 (LOC346887), mRNA. 235278 a O.OOO2619 -3.6 C20orf133 chromosome 20 open reading 20p12.1 rame 133 (C20orf133), transcript variant 2, mRNA. 235309 a p < 1e–07 -4.9 CDNA clone IMAGE: 4140029 16 235327 x at 0.0002751 2.9 UBXD4 UBX domain containing 4 2p23.3 (UBXD4), mRNA. 235343 a 3.96E-OS 4.2 FLJ12505 Hypothetical protein FLJ12505 1q32.3 235374 a 3.04E-05 2.7 MDH1 Malate dehydrogenase 1, NAD 2p13.3 (soluble) 235412 a O.OOO2124 3.2 ARHGEF7 Rho guanine nucleotide 13q34 exchange factor (GEF) 7 235433 a O.OOO3304 -2.2 SATL1 Spermidine?spermine N1-acetyl Xq21.1 transferase-like 1 235556 a O.OOO1833 -2 Transcribed locus, weakly 5 similar to NP 703324.1 glutamic acid-rich protein (garp) Plasmodium falciparum 3D7 235.601 a O.OOO3888 2.7 ESTs 235612 a 2.90E-OS -3.1 Transcribed locus, moderately 1 similar to NP 858931.1. NES1 nitrogen fixation 1 isoform b precursor; cysteine desulfurase: nitrogen-fixing bacteria S-like protein; nitrogen fixation 1 (S. cerevisiae, homolog) Homo Sapiens 235628 X at 0.0004866 2.4 Hypothetical protein 5q23.2 LOC133926 235693 at O.OOO3764 2.9 Transcribed locus 5 235725 at O.OOO4132 -2.2 Transcribed locus 14 235927 at 6.41E-OS 2.4 XPO1 Exportin 1 (CRM1 homolog, 2p16 yeast) 235944 at O.OOO4.031 3.3 HMCN1 hemicentin 1 (HMCN1), 1q25.3-q31.1 mRNA. 236249 at 1.54E-OS 2.7 IKIP KK interacting protein (IKIP), 12q23.1 transcript variant 1, mRNA. 236251 at 8.16E-OS 2.7 ITGAV integrin, alpha V (vitronectin 2a31-q32 receptor, alpha polypeptide, antigen CD51) US 8,440,393 B2 131 132 TABLE 1-continued Gene expression profile. Affymetrix Parametric Fold Change Probe Set P-Value (Tumor? Normal) Gene Symbol Description Map 236678 at O.OOOS987 2.8 JAG1 Jagged 1 (Alagile syndrome) 20p12.1-p11.23 236715 x at 0.0002206 2.5 UACA uveal autoantigen with coiled- 15q22-q24 coil domains and ankyrin repeats (UACA), transcript variant 1, mRNA. 236829 at 6.57E-OS 2.9 TMF1 TATA element modulatory 3p21-p12 factor 1 236883 at O.OOO3O76 2.6 ESTs 236936 at O.OOO7927 -2.4 Transcribed locus 8 236966 at O.OOO47S7 2.6 TXNDC6 thioredoxin domain containing 6 3q22.3 (TXNDC6), mRNA. 236974 at O.OOO3599 2.5 CCNI Cyclin I 4q21.1 237206 at O.OOO3483 -5.9 MYOCD Myocardin 17p11.2 237333 at O.OOO8796 -2 SYNCOILIN Syncoilin, intermediate filament 1 1p34.3-p33 2374.75 X at 0.0001562 4.3 SEPP1 Selenoprotein P. plasma, 1 5q31 237494 at 2.03E-OS 3.2 Transcribed locus 15 237868 X at 0.000358 3 ESTs, Weakly similar to ALUF HUMAN ALU CLASSF WARNING ENTRY !!! H. sapiens 238.026 at O.OOO790S -2 RPL3SA Ribosomal protein L35a 3q29-qter 238142 at O.OOO678 2.4 LOC2S3982 Hypothetical protein 16q11.2 LOC2S3982 238183 at O.OOO1644 3.7 ESTs 238273 at O.OOO7281 2 Full-length cDNA clone 7 CSODBOOSYG10 of Neuroblastoma Cot 10 normalized of Homo sapiens (human) 238327 a 6.09E-OS 2.6 Similar to MGC52679 protein 22d 13.33 238478 a O.OOO183 -3.7 BNC2 Basonuclin 2 9p22.3-p22.2 238584 a O.OOO21.96 3.3 IQCA IQ motif containing with AAA 2a37.2-q37.3 domain 238613 a 8.97E-OS -2.7 ZAK sterile alpha motif and leucine 2d24.2 Zipper containing kinase AZK (ZAK), transcript variant 2, mRNA. 238642 a. O.OOO3128 3 LOC338692 Ankyrin repeat domain 13 11 q13.2 amily, member D 238673 a 6.02E-OS 4.7 Transcribed locus 8 238714 a O.OOO1825 2.5 RAB12, member RAS oncogene 18p11.22 amily 238719 a O.OOO4981 -2.5 PPP2CA Protein phosphatase 2 (formerly 5q31.1 2A), catalytic Subunit, alpha isoform 238852 a O.OOO3O11 2.3 ESTs 238878 a O.OOO6945 -6.2 ARX Aristaless related homeobox Xp22.1-p21.3 239227 a 9.04E-05 3 EXT1 Exostoses (multiple) 1 8q24.11-q24.13 239246 a O.OOO2103 3 FARP1 FERM, RhoGEF (ARHGEF) 13q32.2 and pleckstrin domain protein 1 (chondrocyte-derived) 239258 a O.OOO9901 3.1 RHOQ Rashomolog gene family, 2p2 member Q 2392.62 a. 3.72E-OS -3.8 CDNA FLJ26242 fis, clone 11 DMCOO770 239264 a O.OOO3642 2.2 SEC8L1 SEC8-like 1 (S. cerevisiae) 7q3 2393.67 a O.OOO1649 3.1 BDNF brain-derived neurotrophic 11 p.13 actor (BDNF), transcript variant 6, mRNA. 239516 a O.OOO1834 2.7 LYPLAL1 Lysophospholipase-like 1 1q4 239540 a O.OOO9106 2.6 GTF3C1 General transcription factor IIIC, 16p12 polypeptide 1, alpha 220 kDa 239748 x at 0.0003286 3.7 OCLA ovarian carcinoma 4p1 immunoreactive antigen 239806 a O.OOO2S2 4.4 Transcribed locus 2 239848 a O.OOOS816 -2.7 GA17 Dendritic cell protein 11 p.13 240216 a O.OOO1639 2.6 ZBTB20 Zinc finger and BTB domain 3q13.2 containing 20 240421 x at 0.0001971 3.7 CDNA clone IMAGE: 5268630 4 240655 a. 3.01E-OS 4.3 ALCAM Activated leukocyte cell 3q13.1 adhesion molecule US 8,440,393 B2 133 134 TABLE 1-continued Gene expression profile. Affymetrix Parametric Fold Change Probe Set P-Value (Tumor? Normal) Gene Symbol Description Map 240795 at O.OOO1514 2.3 CDNA clone IMAGE: 528.8566 5 241.223 X at 0.0002281 3.2 ESTs, Weakly similar to ALU1 HUMAN ALU SUBFAMILY J SEQUENCE CONTAMINATION WARNING ENTRY H. Sapiens 24.1268 X at 0.0008446 3.5 SAMHD1 SAM domain and HD domain 1 20pter-q12 241.303 X at 0.0007123 2.9 ESTs 24.1387 at O.OOO812S 2.7 PTK2 PTK2 protein tyrosine kinase 2 8q24-qter 241421 at O.OOOSO12 2.6 Transcribed locus 1 241435 at 1.S2E-OS 4.4 ETS1 V-ets erythroblastosis virus E26 11q23.3 oncogene homolog 1 (avian) 241617 X at 0.000334 2.6 ESTs, Weakly similar to 81.0024C cytochrome oxidase I H. Sapiens 241627 X at 0.0005548 2.9 FLJ10357 Hypothetical protein FLJ10357 14q11.2 241632 X at 3.7OE-06 3.3 ESTs 241686 X at 3.28E-05 4.4 ESTs, Weakly similar to hypothetical protein FLJ20378 Homo sapiens H. sapiens 24.1718 X at 7.37E-05 3.4 ESTs 241727 x at 0.0003166 2.3 DHFRL1 dihydrofolate reductase-like 1 3q11.2 (DHFRL1), mRNA. 24.1769 at O.OOO4802 2.5 ITGAV integrin, alpha V (vitronectin 2a31-q32 receptor, alpha polypeptide, antigen CD51) 241809 at O.OOO1618 2.5 LOC284.465 Hypothetical protein 1p13.2 LOC284.465 242029 at 9.6OE-06 3.7 FNDC3B Fibronectin type III domain 3q26.31 containing 3B 242051 at 6.2OE-06 -3.7 Transcribed locus X 242100 at 2.4OE-06 4.8 CSS3 chondroitin sulfate synthase 3 5q23.3 (CSS3), mRNA. 242171 at O.OOO7535 3.1 ESTs, Weakly similar to ALU1 HUMAN ALU SUBFAMILY J SEQUENCE CONTAMINATION WARNING ENTRY H. Sapiens 242233 at 7.63E-OS 2.4 KIAA1219 KIAA1219 protein 20d 11.23 242240 at O.OOO1413 3 PTK2 PTK2 protein tyrosine kinase 2 8q24-qter 242329 at 8.94E-OS 3.1 LOC4O1317 PREDICTED: hypothetical 7 LOC4O1317 (LOC4O1317), mRNA. 242363 at 9. OSE-OS -2.7 DNCI2 Dynein, cytoplasmic, 2a31.1 intermediate polypeptide 2 242364 X at 0.0001.237 3.1 EVER1 Epidermodysplasia 17q25.3 verruciformis 1 242369 x at 7.2OE-06 4.1 NCOA2 Nuclear receptor coactivator 2 8q13.3 242398 X at 3.00E-06 6.1 MEPSO WD repeat domain 77 1p13.2 242405 at 4.O8E-OS 2 MAML2 Mastermind-like 2 (Drosophila) 11q21 242488 at 2.11E-05 -4.7 CDNA FLJ38396 fis, clone 1 FEBRA2OO7957 242500 at O.OOO7278 2.6 Transcribed locus 6 242546 at 2.7OE-06 5.4 LOC44O156 14q11.1 2425.78 x at 0.0002453 4 SLC22A3 Solute carrier family 22 6q26-q27 (extraneuronal monoamine transporter), member 3 242862 X at 6.29E-05 5.2 ESTs 242999 at 0.0007732 2.3 ARHGEF7 Rho guanine nucleotide 13q34 exchange factor (GEF) 7 (ARHGEF7), transcript variant 2, mRNA. 243 g at 9.74E-OS 2.1 MAP4 microtubule-associated protein 4 3p21 (MAP4), transcript variant 2, mRNA. 243006 at 1.1OE-06 4.7 FYN FYN oncogene related to SRC, 6q21 FGR, YES 243147 X at 7.12E-05 5.7 ESTs, Weakly similar to RMS1 HUMAN REGULATOR OF MITOTICSPINDLE ASSEMBLY 1 H. sapiens US 8,440,393 B2 135 136 TABLE 1-continued Gene expression profile. Affymetrix Parametric Fold Change Probe Set P-Value (Tumor? Normal) Gene Symbol Description Map 243249 at 7.21E-OS 2.4 ESTs, Weakly similar to hypothetical protein FLJ20378 Homo sapiens H. sapiens 243305 at 2.9 KIAA1340 Kelch domain containing 5 12p11.22 243442 X at 3.7 ESTs, Weakly similar to ALU1 HUMAN ALU SUBFAMILY J SEQUENCE CONTAMINATION WARNING ENTRY H. Sapiens 243846 x at O.OOO7887 3 FLJ32810 Hypothetical protein FLJ32810 11q22.1 243915 a. O.OOO1482 2.9 ESTs, Weakly similar to 2109260A B cell growth factor H. Sapiens 244007 a O.OOO373 -2 Transcribed locus 244022 a. O.OOO7683 2 FNDC3B Fibronectin type III domain 3q26.31 containing 3B 244050 a O.OOO1908 -2.6 LOC4O1494 similar to RIKEN 4933428IO3 9p21.3 (LOC4O1494), mRNA. 244188 a O.OOO4372 3.1 FL21924 Hypothetical protein FLJ21924 1p13 244193 a O.OOO3257 2.1 FLJ13236 Hypothetical protein FLJ13236 2d 13.12 244197 x at 2.3OE-OS 3.3 CNOT2 CCR4-NOT transcription 2a15 complex, Subunit 2 244457 a 8.66E-OS 3.1 ITPR2 nositol 14,5-triphosphate 2p11 receptor, type 2 244503 a 2.5 ESTs 244633 a 2.5 PIAS2 Protein inhibitor of activated STAT, 2 244648 a O.OOO3403 3.5 FLJ10996 Hypothetical protein FLJ10996 244745 a. O.OOO6739 -2.7 RERG RAS-like, estrogen-regulated, growth inhibitor (RERG), mRNA. 31874 a 8.21E-OS 2.9 GAS2L1 Growth arrest-specific 2 like 1 33323 r at O.OOO1426 S.1 SFN Stratifin 38069 a O.OOO31.97 2 CLCN7 chloride channel 7 (CLCN7), mRNA. 38671 a O.OOO3864 2.4 PLXND1 plexin D1 (PLXND1), mRNA. 39549 a O.OOO8832 2.5 NPAS2 neuronal PAS domain protein 2 (NPAS2), mRNA. 39891 a O.OOOSS41 2.2 DKFZp547K1113 Hypothetical protein DKFZp547K1113 40524 a 2.8 PTPN21 protein tyrosine phosphatase, non-receptor type 21 (PTPN21), mRNA. 41856 a O.OOO8627 2.3 UNCSB Unc-5 homolog B (C. elegans) 44783 s at 6.8SE-OS 3 HEY1 hairy fenhancer-of-split related with YRPW motif 1 (HEY1), mRNA. 46665 a. S.11E-OS 2.4 SEMA4C Sema domain, immunoglobulin 2d 11.2 domain (Ig), transmembrane domain (TM) and short cytoplasmic domain, (semaphorin) 4C (SEMA4C), mRNA. 47550 a 3.4OE-05 4.4 LZTS1 leucine Zipper, putative tumor 8p22 Suppressor 1 (LZTS1), mRNA. 50376 a O.OOO154 2.1 ZNF444 Zinc finger protein 444 19q13.43 (ZNF444), mRNA. 52255 S at 1.1SE-OS 6.1 COLSA3 collagen, type V, alpha 3 19p 13.2 (COL5A3), mRNA. 55583 a O.OOO1374 3.2 DOCK6 dedicator of cytokinesis 6 19p 13.2 (DOCK6), mRNA. 57539 a O.OOO6736 2.1 ZGPAT Zinc finger, CCCH-type with G 20d 13.3 patch domain (ZGPAT), transcript variant 3, mRNA. 57703 a O.OOO1957 2.4 SENP5 SUMO1, sentrin specific peptidase 5 (SENP5), mRNA. 57739 a O.OOO4289 2.2 DND1 Dead end homolog 1 (zebrafish) 59433 a 5.44E-05 2.2 Transcribed locus 61734 a 3.36E-OS 2.7 RCN3 Reticulocal bin 3, EF-hand 19q13.33 calcium binding domain US 8,440,393 B2 137 138 As indicated in Table 1, 652 genes were up-regulated at least 6-fold in tumor endothelium (Table 2), with 7 being 22-fold in ovarian cancer endothelium samples. Of the 652 elevated more than 10-fold and 2 being elevated more than genes that were up-regulated 22-fold, 35 genes were elevated 28-fold. TABLE 2 Genes up-regulated by 26-fold in the tumor associated endothelium. Fold chromosomal Gene ID Gene Description difference location Function 25975 EGFL6 EGF-like-domain, 36.8 Xp22 May regulate cell cycle and Multiple 6 oncogenesis 7130 TNFAIP6 Tumor necrosis factor, 29.1 2d23.3 Anti-inflammatory and Alpha-induced protein 6 chondroprotective effect 7291 TWIST1 Basic helix-loop-helix 19.0 7p21.2 inhibits chondrogenesis (bHLH) transcription factor 6781 STC1 Stanniocalcin 1 13.3 8p21-p11.2 Regulates calcium phosphate homeostasis, and cell metabolism 8452S HOP Homeodomain-only 13.1 4q11-q12 Transcriptional repressor. Protein, transcript Modulates serum response variant 2 actor-dependent cardiac specific gene expression and cardiac development 1462 CSPG2 Chondroitin sulfate 10.4 5q14.3 Extracellular matrix proteoglycan 2 component of the vitreous gel. (versican) Anti-cell adhesive. 57125 Plexin domain 10.2 17q21.1 Unknown containing 1 6696 SPP1 Secreted 9.5 4q21-q25 Expressed during phosphoprotein 1 embryogenesis, wound (Osteopontin, bone healing, and tumorigenesis. sialoprotein, Regulates the assembly of Early T-lymphocyte heterotypic fibers composed of activation1) both type 1 and type V collagen. 4318 MMP9 Matrix metallopeptidase 9 9.4 20c.11.2-q13.1 Breakdown of extracellular (gelatinase B, 92 kDa matrix. Plays a role in type IV collagenase) angiogenesis and tumor invasion 3937 Lymphocyte cytosolic 8.7 5q33.1-qter Promotes T-cell protein 2 (SH2 domain development and containing Leukocyte activation protein of 76 kDa) 1521.89 Chemokine-like factor 8.6 3p23 Regulates EGF-induced Superfamily 8 signaling. Regulates cell proliferation S366 PMAIP1 Phorbol-12-myristate 8.5 18q21.32 Unknown 13-acetate-induced protein 1 24147 Four jointed box 1 7.7 11p13 in drosophila, a downstream (drosophila) arget of the Notch signaling pathway, regulates cell growth and differentiation. Not known in human 8038 ADAM12 ADAM metallo 7.6 10q26.3 Critical for tumor peptidase domain 12 development. Involved in cell (meltrin alpha) cell and cell-matrix interactions. 9636 GIP2 Interferon, alpha 6.9 1p36.33 Unclear, may be related to inducible protein (clone regulation of cell proliferation IFI-15K) and differentiation 25878 MXRAS Matrix-remodeling 6.9 Xp22.33 Matrix remodeling associated 5 1123 Chimerin 6.9 2a31-q32.1 Rho GTPase activating protein (chimaerin) 1 3310 HSPA6 Heat shock 70 KDa 6.8 1923 Involved in protein protein 6 (HSP70B) conformational interactions 11211 Frizzled homolog 10 6.7 12q24.33 Receptor for the wingless (Drosophila) Type MMTV integration site family. 10631 POSTN Periostin, osteoblast 6.7 13q13.3 Promotes integrin-dependent specific factor cell adhesion and motility, involved in extracellular matrix deposition 85.477 SCIN Scinderin 6.6 7p21.3 Ca' dependent actin filament Severing protein, regulates cortical actin network dynamics US 8,440,393 B2 139 140 TABLE 2-continued Genes up-regulated by 26-fold in the tumor associated endothelium. Entrez Fold chromosomal Gene ID Gene Description difference location Function 27242 TNFRSF21 Tumor necrosis factor 6.6 6p21.1-12.2 Unclear; maybe related to Receptor Superfamily, activation of NF-kappaB and Member 21 MAPK8/JNK, induces cell apoptosis, involved in inflammation and immune regulation. 25891 Regeneration associated 6.2 11p13 Unknown H2123 muscle protease, transcript variant 2 4582 MUC1 Mucin 1, 6.2 Regulates cell aggregation, transmembrane, adhesion transcript variant 4 79084 MEPSO WD repeat domain 77, 6.1 1p13.2 involved in the methylation WDR77 and assembly of spliceosomal SnRNAs Sm proteins 50509 COLSA3 Collagen, type V, alpha 3 6.1 19p 13.2 Extracellular protein, associated with formation of fibrils, and some connective issue pathology Such as inflammation, cancer and atherosclerosis RPS11 Ribosomal protein S11 6.1 19q13.3 nvolved in the recognition of ermination codons. CENTA2 Centaurin, alpha 2 6.1 17q11.2 A phosphatidylinositide binding protein present in the dense membrane fractions of cell extracts 2295 FOXF2 Forkhead box F2 6.O 6p25.3 Regulates cell proliferation and Survival, associated with BMP and Wnt signaling

TABLE 3 Genes down-regulated in tumor associated endothelium. Entrez Chromosomal Gene ID Gene Description Fold difference location Function 5350 PLN Phospholamban O.108 6q22.1 Inhibits sarcoplasmic reticulum Ca(2+)- ATPase activity 6401 SELE Selectin E, O112 1q22-q25 Cell surface lycoprotein. endothelial Inhibits cell adhesion. Adhesion molecule 1 Early marker of inflammation 96.87 GREB1 GREB1 protein, O-116 2p25.1 Transcription factor, transcript variant a inhibits cell proliferation 4969 OGN Osteoglycin O.147 9q22 Regulates collagen osteoinductive fibrillogenesis Factor, mimecan), transcript variant 3 2S890 AB13BP ABI gene family, O.152 3q12 May play a role in cell member 3 (NESH) motility by regulating binding protein NESH function 90.161 HS6ST2 Heparn sulfate 6- O.153 Xq26.2 Plays a role in growth O-sulfotransferase 2 factor signaling, cell adhesion, and enzymatic catalysis. Maybe involved in vascularization by mediating FGF signaling 139221 MUM1L1 Melanoma associated 0.155 Xq22.3 Encodes tumor specific antigen (mutated) 1- antigens like 1 4129 MAOB Monoamine oxidase B O.156 Xp11.23 Regulates neurotransmitters in central nervous system 9452 ITM2A Integral membrane O.156 Xq13.3-Xq21.2 Transmembrane protein. protein 2A Marker of early stage of endochondral Ossification US 8,440,393 B2 141 142 TABLE 3-continued Genes down-regulated in tumor associated endothelium. Entrez Chromosomal Gene ID Gene Description Fold difference location Function 17O3O2 ARX Aristaless related O.162 Xp22.1-p21.3 Organ development. homeobox Bifunctional transcription factor 10659 CUGBP2 CUG triplet repeat, O.163 10p13 Binds and stabilizes RNA binding Protein COX2 mRNA, inhibits 2, transcript variant 2 its translation 5577 PRKAR2B Protein kinase, O.163 7q22 Encodes a regulatory cAMP-dependent subunit RII beta of regulatory, type II, human cAMP-dependent beta protein kinase A 345557 LCXD3 Phosphatidylinositol- O.16S 5p13.1 Quantitatively specific solubilizes phospholipase C, X AChE from purified domain containing 3 synaptic plasma membranes and intact synaptosomes of Torpedo ocellata electric Organ

Multiple genes encoding proteins involved in extracellular monoamine oxidase B (MAOB), a gene responsible for matrix function, such as collagens, TNFAIP6, ADAMTS4. 25 detoxification and degradation of monoamines was decreased MMP9, MMP11, had increased expression in tumor vascu by 6.4-fold in the tumor endothelial cells (Grimsby et al., Nat. lature compared with normal ovarian vasculature. The C. Genet. 17: 206-210, 1997). Decorin, a small multi-functional integrin (vitronectin receptor) was elevated 2.5-fold in tumor proteoglycan with anti-angiogenic properties, was decreased endothelium. Several transcription factors were upregulated by 4.8-fold (Sulochana et al., J. Biol. Chem. 280: 27935 in the ovarian cancer vasculature. For example, HEYL was 30 27948, 2005). Several other genes with potential anti-angio increased 3-fold. In addition, several transcription factors genic or anti-proliferative roles such as Fibulin-5 (FBLN-5) were identified including E2F transcription factor 3 (E2F3; and checkpoint suppressor 1 (CHES1) were down-regulated plays a role in cell proliferation) (Black, Proc Natl Acad Sci by 4.5-fold and 4.3-fold, respectively (Albig and Schiemann, U.S.A. 102: 15948-15953, 2005); runt-related transcription DNA Cell Biol. 23: 367-379, 2004; and Scott and Plon, Gene factor 1 (RUNX1; plays a role in angiogenesis) (Iwatsuki et 35 359: 119-126, 2005). These findings indicate that tumor al. Oncogene 24: 1129-1137, 2005), signal transducer and endothelial cell and non-tumor endothelial cell isolates pos activator of transcription 2 (STAT2; role in cellular prolifera sess distinct expression profiles. tion) (Gomez and Reich, J. Immunol. 170:5373-5381, 2003), the SNAIL-related zinc-finger transcription factor, SLUG Example 4 (SNAI2) (Perez-Mancera et al., Oncogene 24: 3073-3082, 40 2005), and Twist1 (Mironchik et al. Cancer Res.65: 10801 Identification of Tumor Endothelial Markers 10809, 2005). These genes were elevated 2-18 fold in the ovarian cancer vasculature relative to normal ovarian endot This example provides specific tumor endothelial cell helial cells. markers. Additional genes were identified as being overexpressed in 45 Differentially regulated genes expressed in tumor-associ ovarian tumor endothelial cells that had previously been ated endothelium were identified by comparing tumor-asso reported to be overexpressed in tumor cells. For example, ciated endothelium versus normal endothelium with tumor epidermal growth factor receptor (EGFR) expression was associated epithelial cells versus ovarian Surface epithelium increased by 3.5-fold in the tumor endothelium. EGFR is (OSE). The expression profile of microdissected papillary known to be overexpressed in ovarian carcinomas and is 50 serous ovarian cancers using the same microarray methods predictive of poor outcome (Berchuck et al., Am. J. Obstet. has previously been reported (Bonome et al. Cancer Res.65: Gynecol. 164: 669-674, 1991). Similarly, non-receptor 10602-10612, 2005). The current disclosed list of differen kinases such as focal adhesion kinase (FAK or PTK2: 3.1- tially expressed genes in tumor-associated endothelial cells fold increase) and Fyn (4.7 fold increase), which are play was compared with the gene list identified for laser microdis functional roles for tumor cells were detected. Genes that are sected tumor-epithelial cells. A total of 534 differentially overexpressed on both tumor cells and tumor-associated 55 regulated genes were uniquely altered (up- or down-regu endothelial cells are targets for anti-vascular therapy due to lated) in the endothelial cells. The 28 genes with the greatest the ability to target both the epithelial and stromal compart level of increase in ovarian tumor endothelial cells are listed mentS. in Table 4, of which 23 genes had a 26-fold increase in In addition to the 652 genes that were identified as being expression. A complete listing of the 534 differentially regu up-regulated in ovarian tumor endothelial cells, 497 genes 60 lated genes is provided in Table 5. Further, FIG. 2 illustrates were down-regulated 22-fold in ovarian cancer endothelium protein expression levels detected in ovarian endothelial cells samples (Table 1). FIG. 1 illustrates the fold changes following staining of samples with immunofluorescently-la observed in the relative expression levels between microarray beled PTK2, Fyn, MMP-9, 32-arrestin, Jagged 1 and data and real-time quantitative RT-PCR data from the pro PLXDC1, respectively. angiogenic gene signature provided in Table 1. Of the 497 65 These findings identify tumor endothelial cell specific genes that were down-regulated 22-fold, 17 genes were genes that can be used as biomarkers and potential targets for decreased at least 6-fold (as provide in Table 3). For example, treatment of ovarian cancer. US 8,440,393 B2 143 144 TABLE 4 Genes Specifically regulated in tumor-endothelium

Entrez Fold Chromosomal Gene ID Gene Description difference location Function 25975 EGFL6 EGF-like-domain, multiple 6 36.848 Xp22 May regulate cell cycle (EGFL6) and oncogenesis 7130 TNFAIP6 Tumor necrosis factor, alpha 29.062 2d23.3 Anti-inflammatory and induced protein 6 (TNFAIP6) chondroprotective effect 7291 TWIST1 Twist homolog 1 18.969 7p21.2 inhibits chondrogenesis (acrocephalosyndactyly 3: Saethre-Chotzen syndrome) (Drosophila) (TWIST1) 6781 STC1 Stanniocalcin 1 (STC1) 13.326 8p21-p11.2 Regulates calcium phosphate homeostasis, and cell metabolism 84525 HOP Homeodomain-only protein 13.144 4q11-q12 Transcriptional (HOP), transcript variant 2 repressor. Modulates serum response actor-dependent cardiac-specific gene expression and cardiac development 1462 CSPG2 Chondroitin sulfate 10.355 5q14.3 Extracellular matrix proteoglycan 2 (versican) component of the (CSPG2) vitreous gel. Anti-cell adhesive. 57125 Plexin domain containing 1 10.215 17q21.1 Unknown (PLXDC1) 4318 MMP9 Matrix metallopeptidase 9 9.389 20q11.2-q13.1 Breakdown of (gelatinase B, 92 kDa extracellular gelatinase, 92 kDa type IV matrix. Plays a role in collagenase) (MMP9) angiogenesis and tumor invasion 3937 LCP2 Lymphocyte cytosolic protein 8.744 5q33.1-qter Promotes T-cell 2 (SH2 domain containing development and leukocyte protein of 76 kDa) activation (LCP2) S366 PMAIP1 Phorbol-12-myristate-13 8.543 18q21.32 Unknown acetate-induced protein 1 (PMAIP1) 8038 ADAM12 ADAM metallopeptidase 7.605 10q26.3 Involved in cell-cell and domain 12 (meltrin alpha) cell-matrix interactions (ADAM12), transcript variant 1 25878 MXRAS Matrix-remodeling associated 6.865 Xp22.33 Matrix remodeling 5 (MXRA5) 1123 CHN1 Chimerin (chimaerin) 1 6.857 2a31-q32.1 Rho GTPase activating (CHN1) protein 3310 HSPA6 Heat shock 70 kDa protein 6 6.76 1d23 Involved in protein conformational interactions 10631 POSTN Periostin, osteoblast specific 6.732 13q13.3 Promotes integrin factor (POSTN) dependent cell adhesion and motility, involved in extracellular matrix deposition 11211 Frizzled homolog 10 6.701 12q24.33 Receptor for the (Drosophila) (FZD10) wingless type MMTV integration site family 27242 TNFRSF21 Tumor necrosis factor 6.649 6p21.1-12.2 Activates NF-kappaB receptor Superfamily, member and MAPK8/JNK, 21 (TNFRSF21) induces cell apoptosis, involved in inflammation and immune regulation. 25891 Regeneration associated 6.199 11p13 Unknown muscle protease, transcript variant 2 79084 MEPSO WD repeat domain 77 6.144. 1 p13.2 Involved in the methylation and assembly of spliceosomal SnRNAs Sm proteins 50509 COLSA3 Collagen, type V, alpha 3 6.118 19p 13.2 Extracellular protein, (COL5A3) associated with formation of fibrils, and Some connective tissue pathology Such as inflammation, cancer and atherosclerosis US 8,440,393 B2 145 146 TABLE 4-continued Genes specifically regulated in tumor-endothelium

Entrez Fold Chromosomal Gene ID Gene Description difference location Function 62OS RPS11 Ribosomal protein S11 6.095 19q13.3 Involved in the (RPS11) recognition of termination codons SS8O3 CENTA2 Centaurin, alpha 2 (CENTA2) 6.09 17q11.2 A. phosphatidylinositide binding protein present in the dense membrane fractions of cell extracts 9.0161 HS6ST2 Heparan sulfate 6-O- 0.153 Xq26.2 Plays a role in growth sulfotransferase 2 (HS6ST2) factor signaling, cell adhesion, and enzymatic catalysis. Maybe involved in vascularization by mediating FGF signaling 4969 OGN Osteoglycin (osteoinductive 0.147 9q22 Regulates collagen factor, mimecan) (OGN), fibrillogenesis transcript variant 3 96.87 GREB1 GREB1 protein, transcript 0.116 2p25.1 Transcription factor; variant a inhibits cell proliferation 6401 SELE Selectin E (endothelial 0.112 1d22-q25 Cell surface lycoprotein. adhesion molecule 1) (SELE) Inhibits cell adhesion. Early marker of inflammation 5350 PLN Phospholamban 0.108 6q22.1 Inhibits sarcoplasmic reticulum Ca(2+)- ATPase activity

TABLE 5 Genes Specifically regulated in tumor endothelium. Fold difference Parametric of geom means Gene p-value (Tumor? Normal) Probe set Description UG cluster symbol Map 29.062 206026 s at tumor necrosis factor, HS.437322 TNFAIP6 2d23.3 alpha-induced protein 6 (TNFAIP6), mRNA. O.274 213803 at Karyopherin (importin) HS.S32793 KPNB1 17q21.32 beta 1 O.2OS 235309 at CDNA clone HS.526499 16 MAGE: 4140029 2.85 228204 at proteasome (prosome, HS.89S45 PSMB4 1q21 macropain) subunit, beta type, 4 (PSMB4), mRNA. 13.144 211597 s at homeodomain-only HS.121443 HOP 4q11-q12 protein (HOP), transcript variant 2, mRNA. 3.116 213848 at Dual specificity HS.3843 DUSP7 3p21 phosphatase 7 3.599 233274 at NCK adaptor protein 1 HS.477693 NCK1 3q21 O.266 212653 s at EH domain binding Hs.271667 EHBP1 2p15 protein 1 (EHBP1), mRNA. 4e-007 10.215 214081 at plexin domain containing HS.12SO36 PLXDC1 17q21.1 (PLXDC1), mRNA. 4e-007 0.27 213364 S at sorting nexin 1 (SNX1), HS.188634 SNX1 15q22.31 transcript variant 2, mRNA. 4e-007 36.848 219454 at EGF-like-domain, HS-12844 EGFL6 Xp22 multiple 6 (EGFL6), mRNA. 4e-007 18.969 213943 at twist homolog 1 HS.66744 TWIST1 7p21.2 (acrocephaloSyndactyly 3: Saethre-Chotzen syndrome) (Drosophila) (TWIST1), mRNA. US 8,440,393 B2 147 148 TABLE 5-continued Genes Specifically regulated in tumor endothelium. Fold difference Parametric of geom means Gene p-value (Tumor? Normal) Probe set Description UG cluster symbol Map 4e-007 0.327 2251.23 at Sestrin 3 HS.12O633 SESN3 11q21 Se-OO7 O.16S 230O81 at phosphatidylinositol- HS14S4O4 PLCXD3 5p13.1 specific phospholipase C, X domain containing 3 (PLCXD3), mRNA. 7e-OO7 5.249 228579 at Potassium voltage-gated Hs.374023 KCNQ3 8q24 channel, KQT-like Subfamily, member 3 8e-007 6.199 213661 at regeneration associated Hs.55044 DKFZP586H2123 11p13 muscle protease (DKFZP586H2123), transcript variant 2, mRNA. 9e-OO7 6.701 219764 at frizzled homolog 10 S.31664 FZD10 12q24.33 (Drosophila) (FZD10), mRNA. 9e-OO7 3.38 212044 s at ribosomal protein L27a S.S23463 RPL27A 11p 15 (RPL27A), mRNA. 1e-OO6 O42 213574 s at Karyopherin (importin) S.S32793 KPNB1 17q21.32 beta 1 .1e-006 7.684 219700 at plexin domain containing S.12SO36 PLXDC1 17q21.1 1 (PLXDC1), mRNA. .2e-006 O.294 222791 at round spermatid basic S.48628S RSBN1 1p13.2 protein 1 (RSBN1), mRNA. .2e-006 O.232 221589 s at Aldehyde dehydrogenase S.293970 ALDH6A1 14q24.3 6 family, member A1 .2e-006 7.724. 204285 s at phorbol-12-myristate-13- HS.96 PMAIP1 18q21.32 acetate-induced protein 1 (PMAIP1), mRNA. 4e-006 10.546 204595 s at stanniocalcin 1 (STC1), S.25590 STC1 8p21-p11.2 mRNA. Se-OO6 3.695 225147 at pleckstrin homology, S.487479 PSCD3 7p22.1 Sec7 and coiled-coil domains 3 (PSCD3), mRNA. Se-OO6 9.368 234723 x at CDNA: FLJ21228 fis, HS.30671 6 7 clone COLOO739 Se-OO6 O.374 212498 at Membrane-associated HS.432862 MARCH- 5p15.2 ring finger (C3HC4) 6 VI Se-OO6 13.326 230746 s at stanniocalcin 1 (STC1), HS.25590 STC1 8p21-p11.2 mRNA. 6e-OO6 6.09 21935.8 s at centaurin, alpha 2 HS.S14063 CENTA2 17q11.2 (CENTA2), mRNA. 8e-006 O.346 201425 at aldehyde dehydrogenase HS.436437 ALDH2 12q24.2 2 family (mitochondrial) (ALDH2), nuclear gene encoding mitochondrial protein, mRNA. 19e-OO6 4.094 2255.05 s at chromosome 20 open HS.29341 C20orf&1 20p13 reading frame 81 (C20orf&1), mRNA. 2.1e-006 O.284 219939 s at cold shock domain HS.6985S CSDE1 1p22 containing E1, RNA binding (CSDE1), transcript variant 2, mRNA. 2.2e-OO6 4.136 22105.9 s at coactosin-like 1 HS289092 COTL1 16q24.1 (Dictyostelium) (COTL1), mRNA. 2.2e-OO6 6.649 21.8856 at tumor necrosis factor HS.443577 TNFRSF21 6p21.1-12.2 receptor Superfamily, member 21 (TNFRSF21), mRNA. 2.4e-006 3.826 205068 s at Rho GTPase activating HS.293593 ARHGAP26 5q31 protein 26 (ARHGAP26), mRNA. 2.4e-006 4.807 242100 at chondroitin sulfate HS.213137 CSS3 5q23.3 synthase 3 (CSS3), mRNA. 2.4e-006 S.981 206377 at forkhead box F2 HS.484423 FOXF2 6p25.3 (FOXF2), mRNA.

US 8,440,393 B2 151 152 TABLE 5-continued Genes Specifically regulated in tumor endothelium. Fold difference Parametric of geom means Gene p-value (Tumor? Normal) Probe set Description UG cluster symbol Map 7.2e-OO6 4.12 242369 X at Nuclear receptor S.446678 NCOA2 8q13.3 coactivator 2 7.3e-OO6 O.26 226625 at Transforming growth S.482390 TGFBR3 1p33-p32 factor, beta receptor III (betaglycan, 300 kDa) 7.7e-OO6 O.245 219511 S at Synuclein, alpha S.426463 SNCAIP 5q23.1-q23.3 interacting protein (synphilin) (SNCAIP), mRNA. 7.8e-006 4.059 222449 at transmembrane, prostate S.S171SS TMEPAI 20d 13.31-q13.33 androgen induced RNA (TMEPAI), transcript variant 3, mRNA. 8e-006 2.237 212351 at eukaryotic translation S.283551 EIF2BS 3q27.1 initiation factor 2B, subunit 5 epsilon, 82 kDa (EIF2B5), mRNA. 8.1e-OO6 3.836 219634 at carbohydrate (chondroitin S.17569 CHST11 12q 4) Sulfotransferase 11 (CHST11), mRNA. 8.3e-OO6 O418 226751 at chromosome 2 open S.212885 C2Orf32 2p14 reading frame 32 (C2orf32), mRNA. 8.3e-OO6 5.037 209081 s at collagen, type XVIII, S.S17356 COL18A1 21q22.3 alpha 1 (COL18A1), transcript variant 2, mRNA. 9e-OO6 2.362 205812 s at transmembrane emp24 S.279929 TMED9 5q35.3 protein transport domain containing 9 (TMED9), mRNA. 9.3e-OO6 2.862 234985 at Hypothetical protein S.20586S LOC143458 11p13 LOC143458 9.6e-OO6 3.652 242029 at Fibronectin type III S.15943O FNDC3B 3q26.31 domain containing 3B 9.6e-OO6 O.386 211666 X at ribosomal protein L3 S.119598 RPL3 22q13 (RPL3), mRNA. 9.8e-006 4.815 2O2465 at procollagen C S.202097 PCOLCE 7q22 endopeptidase enhancer (PCOLCE), mRNA. O6e-OOS O.303 203799 at CD302 antigen (CD302), S.130014 CD3O2 2q24.2 mRNA. 1e-OOS 3.787 228253 at ysyl oxidase-like 3 S.469045 LOXL3 2p13 (LOXL3), mRNA. 1e-OOS 3.674 209685 s at protein kinase C, beta 1 S.4603SS PRKCB1 16p11.2 (PRKCB1), transcript variant 2, mRNA. 11e-OOS O.33 224445 s at Zinc finger, FYVE S.S4.9192 ZFYVE21 14q32.33 domain containing 21 (ZFYVE21), mRNA. 13e-OOS 3.466 222379 at Potassium voltage-gated S.348522 KCNE4 2a36.3 channel, Isk-related family, member 4 15e-OOS 6.118 52255 S at collagen, type V, alpha 3 S.23S368 COLSA3 19p 13.2 (COL5A3), mRNA. 17e-OOS 2.8.21 233180 at Ring finger protein 152 S.465316 RNF152 18q21.33 18e-005 O.232 212224 at aldehyde dehydrogenase S.76392 ALDH1A1 9q21.13 1 family, member A1 (ALDH1A1), mRNA. 2e-OOS 6.09S 213350 at ribosomal protein S11 S.433529 RPS11 19q13.3 (RPS11), mRNA. 2e-OOS 3.729 226911 at hypothetical protein S.201O3 FLJ3915S 5p13.2-p13.1 FLJ39155 (FLJ39155), transcript variant 4, mRNA. 21e-OOS 2.975 219102 at reticulocalbin 3, EF-hand S.439184 RCN3 19q13.33 calcium binding domain (RCN3), mRNA. 25e-OOS O.115 222722 at osteoglycin S.109439 OGN 9q22 (Osteoinductive factor, mimecan) (OGN), transcript variant 3, mRNA. US 8,440,393 B2 153 154 TABLE 5-continued Genes Specifically regulated in tumor endothelium. Fold difference Parametric of geom means Gene p-value (Tumor? Normal) Probe set Description UG cluster symbol 1.29e-005 O.17 227703 s at synaptotagmin-like 4 HS.S22O54 SYTL4 Xq21.33 (granuphilin-a) (SYTL4), mRNA. 1.3e-OOS O46 211988 at SWI/SNF related, matrix SMARCE1 17q21.2 associated, actin dependent regulator of chromatin, Subfamily e, member 1 (SMARCE1), mRNA. 131e-005 O.238 204793 at G protein-coupled S.S22730 GPRASP1 Xq22.1 receptor associated Sorting protein 1 (GPRASP1), mRNA. 1.34e-005 O.339 229891 x at KIAA1704 S.SOf 922 KIAA1704 13q13-q14 1.35e-005 O.215 228554 at MRNA, cDNA S.32405 11 DKFZp586GO321 (from clone DKFZp586G0321) 1.38e-005 O.299 210950 s at airnesyl-diphosphate S.S46253 FDFT1 8p23.1-p22 airnesyltransferase 1 (FDFT1), mRNA. 1.38e-005 4.593 220014 at mesenchymal stem cell S.157461 LOCS1334 5q23.1 protein DSC54 (LOC51334), mRNA. 14e-OOS O.262 225162 at SH3 domain protein D19 S.519018 SH3D19 (SH3D19), mRNA. 141e-005 O.387 212408 at orsin A interacting S.496459 TOR1AIP1 protein 1 (TOR1AIP1), mRNA. 141e-005 O446 201054 at heterogeneous nuclear S.96996 HNRPAO ribonucleoprotein AO (HNRPAO), mRNA. 147e-005 4235 204639 at adenosine deaminase S.407135 ADA 20d 12-q13.11 (ADA), mRNA. 1.5e-OOS 4.351 225646 at cathepsin C (CTSC), transcript variant 2, CTSC 11 q14.1-q14.3 mRNA. 151e-005 3.657 20996.9 s at signal transducer and S.470943 STAT1 2a32.2 activator of transcription 1, 91 kDa (STAT1), transcript variant beta, mRNA. 54e-OOS 2.663 236249 at IKK interacting protein IKIP 12q23.1 (IKIP), transcript variant 1, mRNA. 58e-005 3.274 218804 at transmembrane protein S.S.O.3074 TMEM16A 11q13.3 16A (TMEM16A), mRNA. 59e-OOS O427 221988 at Hypothetical protein S.356467 MGC2747 19p 13.11 MGC2747 59e-OOS 2.779 204786 s at interferon (alpha, beta S.549042 IFNAR2 21q22.11 and omega) receptor 2 (IFNAR2), transcript variant 1 mRNA. 61e-OOS O413 200023 s at eukaryotic translation 11p 15.4 initiation factor 3, Subunit 5 epsilon, 47 kDa (EIF3S5), mRNA. 62e-OOS 4.278 201596 x at keratin 18 (KRT18), KRT18 12q13 transcript variant 2, mRNA. 64e-OOS O.32 222605 at REST corepressor 3 S.356399 RCOR3 1q32.3 (RCOR3), mRNA. 64e-OOS O.371 209733 at Hypothetical protein S.348844 LOC286440 Xq22.3 LOC286440 67e-OOS 4.144 232458 at Collagen, type III, alpha S.443625 COL3A1 1 (Ehlers-Danlos syndrome type IV, autosomal dominant) 7e-OOS O.282 224901 at Stearoyl-CoA desaturase 5 S.379191 SCD4 4q21.3 73e-OOS 4.385 233912 x at ELMO domain S.4501(OS ELMOD2 4q31.21 containing 2 76e-OOS 5.792 220301 at chromosome 18 open S.280781 18q22.1 reading frame 14 (C18orf14), mRNA. US 8,440,393 B2 155 156 TABLE 5-continued Genes specifically regulated in tumor endothelium. Fold difference Parametric of geom means Gene p-value (Tumor? Normal) Probe set Description UG cluster symbol Map 1.76e-005 O.388 207170 S at LETM1 domain HS.370457 LETMD1 12q13.12 containing 1 (LETMD1), transcript variant 3, mRNA. 182e-005 O418 201076 at NHP2 non-histone HS.1822SS NHP2L1 22q13.2-q13.31 chromosome protein 2 like 1 (S. cerevisiae) (NHP2L1), transcript variant 2, mRNA. 182e-005 O.252 2288.85 at MAM domain containing S.127386 MAMDC2 9q21.11 2 (MAMDC2), mRNA. 185e-005 O411 212609 s at V-akt murine thymoma S.498.292 AKT3 viral oncogene homolog 3 (protein kinase B, gamma) 186e-005 4564 221558 s at lymphoid enhancer S.SS5947 LEF1 binding factor 1 (LEF1), mRNA. 187e-005 3.054 235204 at Ectonucleoside S.369424 ENTPD7 10 triphosphate diphosphohydrolase 7 19e-OOS 3.283 202820 at aryl hydrocarbon receptor S.171189 AHR (AHR), mRNA. 1.91e-005 4.787 1565823 at septin 7 (SEPT7), S.191346 386O2 7p14.3-p14.1 transcript variant 2 mRNA. 193e-005 O.342 221726 at ribosomal protein L22 S.S15329 RPL22 1p36.3-p36.2 (RPL22), mRNA. 1.93e-OOS O485 202029 X at ribosomal protein L38 S.380953 RPL38 17q23-q25 (RPL38), mRNA. 194e-005 O.258 209513 s at hydroxysteroid S.S9486 HSDL2 dehydrogenase like 2 (HSDL2), mRNA. 195e-005 O.356 2268.06 s at MRNA, cDNA DKFZp686.J23256 (from clone DKFZp686.J23256) 2.01e-005 4.146 1558048 x at Unknown 2.03e-005 3.183 237494 at Transcribed locus HS.174934 15 2.09e-005 O.306 2014.32 at catalase (CAT), mRNA. HSSO23O2 CAT 11p13 2.11e-005 3.698 203878. S. at matrix metallopeptidase HS143751 MMP11 22q11.23 1 (stromelysin 3) (MMP11), mRNA. 2.11e-005 2.481 212323 s at vacuolar protein sorting HS.439381 WPS13D 1p36.22-p36.21 3D (yeast) (VPS13D), transcript variant 2, mRNA. 2.11e-005 O.215 242488 at CDNA FLJ38396 fis, clone FEBRA2007957 214e-OOS 219092 s at chromosome 9 open S.16603 C9orf12 9q21.33-q22.31 reading frame 12 (C9orf12), mRNA. 215e-OOS 2.239 217118 s at chromosome 22 open S.369682 C22Orf 22q13.31 reading frame 9 (C22orf), transcript variant 2, mRNA. 2.29e-005 0.44 214097 at Ribosomal protein S21 S190968 RPS21 20d 13.3 2.29e-005 S.232 225799 at hypothetical protein S.446688. MGC4677 2p11.2 MGC4677 (MGC4677), mRNA. 2.3e-OOS 3.258 244197 x at CCR4-NOT transcription S.1333 SO CNOT2 12q15 complex, Subunit 2 2.37e-005 7.605 202952 s at ADAM metallopeptidase S.386.283 ADAM12 10q26.3 domain 12 (meltrin alpha) (ADAM12), transcript variant 1, mRNA. 2.4e-OOS O462 201030 x at lactate dehydrogenase B HS.446149 LDHB 12p12.2-p12.1 (LDHB), mRNA.