expressionofhoxb2,aretinoicacidsignalingtargetinpancreatic ...transcription factor essential for...

Expression of HOXB2, a Retinoic Acid SignalingTarget in PancreaticCancer and Pancreatic Intraepithelial NeoplasiaDavendraSegara,1Andrew V. Biankin,1,2 James G. Kench,1,3 Catherine C. Langusch,1Amanda C.Dawson,1

David A. Skalicky,1David C. Gotley,4 Maxwell J. Coleman,2 Robert L. Sutherland,1and SusanM. Henshall1

Abstract Purpose: Despite significant progress in understanding the molecular pathology of pancreaticcancer and its precursor lesion: pancreatic intraepithelial neoplasia (PanIN), there remain nomolecules with proven clinical utility as prognostic or therapeutic markers. Here, we used oligo-nucleotide microarrays to interrogate mRNA expression of pancreatic cancer tissue and normalpancreas to identify novel molecular pathways dysregulated in the development and progressionof pancreatic cancer.Experimental Design: RNAwas hybridized to Affymetrix Genechip HG-U133 oligonucleotidemicroarrays. A relational database integrating data from publicly available resources was createdto identify candidate genes potentially relevant to pancreatic cancer. The protein expressionof one candidate, homeobox B2 (HOXB2), in PanIN and pancreatic cancer was assessed usingimmunohistochemistry.Results:We identified aberrant expression of several components of the retinoic acid (RA)signaling pathway (RARa, MUC4, Id-1, MMP9, uPAR, HB-EGF, HOXB6, and HOXB2), many ofwhich are known to be aberrantly expressed in pancreatic cancer and PanIN. HOXB2, a down-stream target of RA, was up-regulated 6.7-fold in pancreatic cancer compared with normal pan-creas. Immunohistochemistry revealedectopic expressionofHOXB2 in15%ofearlyPanIN lesionsand 48 of 128 (38%) pancreatic cancer specimens. Expression of HOXB2 was associated withnonresectable tumors andwas an independent predictor of poor survival in resected tumors.Conclusions:We identified aberrant expression of RA signaling components in pancreaticcancer, including HOXB2, which was expressed in a proportion of PanIN lesions. EctopicexpressionofHOXB2was associatedwith apoor prognosis for allpatientswithpancreatic cancerandwas an independent predictor of survival in patients who underwent resection.

Pancreatic cancer is the fifth leading cause of cancer deathin Western societies with a 5-year survival rate of <10% (1).Pancreatic cancer presents at an advanced stage; thus, only10% to 20% of patients are suitable for surgical treatment atthe time of presentation (1). Clinical management of thesepatients is complicated by inconsistencies in the influence ofconventional clinicopathologic variables on outcome suggesting

that some of these variables lack accuracy. In addition,preoperative assessment of some variables such as lymph nodemetastases is difficult. Whereas in other cancers assessment ofaberrations in gene expression that cosegregate with therapeu-tic response and outcome are being adopted routinely toincrease predictive power (e.g., ER and HER-2/neu in breastcancer), there remain no molecular markers of clinical utilityin pancreatic cancer. This highlights the need for theidentification of novel regulatory pathways important inpancreatic cancer that may also have diagnostic, therapeuticand prognostic utility.

There is now compelling histopathologic and molecularevidence to support the evolution of pancreatic cancer througha series of noninvasive duct lesions called pancreatic intra-epithelial neoplasia (PanIN; refs. 2, 3). Early duct lesionsdesignated PanIN-1A and PanIN-1B show minimal cytologicand architectural atypia and are associated with activating K-rasmutations (4), shortened telomeres (5), and overexpressp21WAF1/CIP1 (6). PanIN-2 lesions exhibit mild to moderatecytologic and architectural atypia and are associated withloss of p16INK4A expression (7) and cyclin D1 overexpression(6). PanIN-3 exhibits significant cytologic and architecturalatypia, manifests p53 mutations (8), and loss of DPC4/Smad4 expression (6). These molecular aberrations increasein frequency with advancing PanIN lesions through to invasivecancer.

www.aacrjournals.org Clin Cancer Res 2005;11(9) May1, 20053587

Authors’Affiliations: 1Cancer Research Program, Garvan Institute of MedicalResearch and 2Division of Surgery, St. Vincent’s Hospital, Darlinghurst, Sydney,New SouthWales, Australia; 3Institute of Clinical Pathology and Medical Research,Westmead Hospital,Westmead, New SouthWales, Australia; and 4University ofQueensland, Department of Surgery, Princess Alexandra Hospital,Woolloongabba,Queensland, AustraliaReceived 9/5/04; revised 2/1/05; accepted 2/10/05.Grant support: Royal Australasian College of Surgeons, National Health andMedical Research Council of Australia, St. Vincent’s Clinic Foundation Sydney,Cancer Council New SouthWales, R.T. Hall Trust, and Prostate Cancer Foundationof Australia (S.M. Henshall).The costs of publication of this article were defrayed in part by the payment of pagecharges.This article must therefore be hereby marked advertisement in accordancewith18 U.S.C. Section1734 solely to indicate this fact.Note:D. Segara and A. Biankin contributed equally to this work.Requests for reprints: Robert L. Sutherland, Cancer Research Program GarvanInstitute of Medical Research 384 Victoria Street, Darlinghurst. New SouthWales 2010, Australia. Phone: 61-2-9295-8322; Fax: 61-2-9295-8321; E-mail:[email protected].

F2005 American Association for Cancer Research.

Cancer Prevention

Cancer Research. on February 27, 2021. © 2005 American Association forclincancerres.aacrjournals.org Downloaded from

http://clincancerres.aacrjournals.org/

During vertebrate development, retinoic acid (RA) signalingis important for the correct patterning of embryonic structures(9). Endodermal expression of pdx-1 (a homeobox-containingtranscription factor essential for pancreatic development) isinduced by RA (10) and marks a pluripotent population of cellsthat give rise to all cell types in the pancreas. RA signalingregulates pancreas exocrine lineage selection, and treatmentwith RA analogues can effect a shift from an acinar to a ductalphenotype through epithelial-mesenchymal interactions (11).Such a shift from an exocrine to a predominantly ductalphenotype is characteristic of mouse models of pancreaticcancer development. In addition, pancreatic stellate cells, whichare essential for the development of fibrosis associated withchronic pancreatitis and pancreatic cancer, store retinoids in fatdroplets, and in turn can have their function altered with RAanalogue treatment in vitro (12). The retinoid signal istransduced by two families of nuclear transcription factors:RA receptors (RAR) and retinoid X receptors, that are membersof the nuclear receptor superfamily, which in the presence ofligand heterodimerize to activate the transcription of targetgenes through RA response elements (13). Although few RAresponse elements have been identified, one of the mechanismsby which retinoids exert their effects is thought to be throughregulation of HOX gene expression (9, 14).

Homeobox genes are transcription factors with establishedroles in development and cell function. The homeobox is ahighly conserved 183-bp DNA sequence coding for a 61-amino-acid domain, the homeodomain (15). This region binds DNAelements, primarily those that contain a TAAT core motif(16). Accordingly, homeodomain containing proteins act asboth activators and repressors of transcription. Human class 1homeobox genes called HOX genes consist of 39 genesarranged in four clusters HOXA, HOXB, HOXC, and HOXDlocalized on chromosomes 7, 17, 12, and 2, respectively (17).Mammalian development requires a complex interaction ofHOX gene networks, with HOX gene expression commencingduring gastrulation and collectively controlling the identity ofvarious regions along the body axis from the hindbrain to thetail (18, 19). Aberrant expression of HOX genes has beenimplicated in the development of solid tumors including renalcarcinoma (20), colon cancer (21), ovarian carcinoma (22),and breast carcinoma (23, 24). Given the emerging importanceof developmental pathways in pancreatic cancer such as Notch(25) and sonic hedgehog (26), the role of RA signaling in earlypancreas development and evidence of RA signaling andhomeobox gene network dysregulation in carcinogenesis, wepresent data suggesting that aberrant RA signaling may beimportant in pancreatic cancer. Based on these data, weassessed HOXB2 , a RA-responsive gene, and show that ectopicexpression of HOXB2 occurs in a significant proportion ofpancreatic cancer, is detectable in a proportion of PanIN, and isassociated with a poor prognosis, supporting a potential role ofHOXB2 in the biological behavior of some pancreatic cancer.

Materials andMethods

RNA preparation and transcript profiling. Ethical approval wasobtained from five teaching hospitals (The Princess Alexandra Hospital,Brisbane, Australia, Westmead Hospital, Concord Hospital, Royal PrinceAlfred Hospital, and St. Vincent’s Hospital Campus in Sydney, Australia)for the acquisition of fresh and archival tissue and recording of

clinicopathologic data. Multiple samples of pancreatic tissue of f500mg were excised intraoperatively from 12 patients, undergoingpancreatic resection for pancreatic cancer, immediately snap frozen inliquid nitrogen and stored at �80jC, before RNA extraction. Total RNAwas isolated from 12 pancreatic cancer specimens and six macroscopi-cally and microscopically normal appearing pancreas from the samepatients (matched). Biotinylated cRNA for Affymetrix Genechip hybrid-ization was prepared through a single round of reverse transcriptionwith Superscript II (Life Technologies, Rockville, MD) followed by se-cond strand synthesis to create double stranded cDNA. After purificationthe cDNA was transcribed and labeled using a T7 polymerase (EnzoTechnologies, New York, NY) and purified (27). Hybridization cocktailswere prepared as per the Affymetrix protocol (Affymetrix, Santa Clara,CA) and quality assured on Affymetrix Test3 arrays, before hybridizationto HG-U133A and B oligonucleotide microarrays.

Data analysis. A relational database was constructed using File-

Maker Pro (FileMaker, Inc., San Francisco, CA) to facilitate multiplequeries of gene expression data generated from the above experiments

and public domain data available electronically from the Internet. Thedatabase incorporated (a) transcript profiles of pancreatic cancer and

normal pancreas from the experiments done in this study (absolute

values); (b) mathematical algorithms programmed within the databaseto generate fold change comparisons between the average expression

across all samples of pancreatic cancer to the average in normal pancreas;

(c) linear statistical analyses generated using the Affymetrix Data MiningTool Software (MAS 5.0), which included t test and Mann-Whitney U

test data for comparisons between normal pancreas and pancreaticcancer and (d) interactive molecular pathway maps were generated using

GenMAPP software (Gladstone Institutes UCSF, San Francisco, CA,

http://www.GenMAPP.org/default.html), designed to incorporate tran-script profile data into maps of known pathways including those

involved in carcinogenesis and development. Data files using Swissprotidentification numbers were uploaded into the program, and various

pathway maps available as part of the package were used to model

numerous pathways. An existing RA signaling GenMAPP was modifiedto include all molecules thought to be regulated by RA signaling and is

presented in Fig. 1. Statistical data were generated using the t test andMann-Whitney U tests to compare the average expression across samples

of pancreatic cancer to the average expression of all samples of normal

pancreas for the GenMAPP that is presented.Patient cohort. We identified a cohort of 128 patients with a diag-

nosis of pancreatic adenocarcinoma that underwent pancreatic resectionor biopsy between January 1972 and November 2001 with availablearchived tissue. This cohort represents a subset of a previously describedgroup of 348 patients (28). Archival formalin-fixed, paraffin-embeddedtissue from all 128 pancreata that were resected or biopsied were used toconstruct seven pancreatic cancer tissue arrays, which contained up to55 � 1.6 mm cores per slide. Conventional sections of 26 cases of normalpancreas from areas distal to the pancreatic cancer were used to assessgene expression in benign ductal epithelial cells and PanIN lesions.

For this cohort, the average age at diagnosis was 63.8 years(median, 66.5; range, 34-86; Table 1). Of the 128 patients, 76 were

from pancreatic resections, 46 intraoperative incision biopsies, and6 postmortem specimens. Median follow-up for the cohort was

7.6 months (range, 0-117 months). Eight patients were alive at thecensus date (September 21, 2002). Median disease-specific survival

was 7.25 months. For the resected group of 76 patients, 39 (51%) hadlymph node metastasis (Table 1). The mean tumor size was 31 mm.Resection margins were microscopically free of tumor in 40 patients

(53%). Poorly differentiated tumors occurred in 25 patients (33%).Median follow-up was 11.0 months with a median disease-specific

survival of 10.1 months, 1-year survival of 48.6%, and 5-year survivalof 11%. The 30-day mortality for resection was 2 (3%).

Immunohistochemistry. Pancreatic tissue microarrays were cut at4 Am, deparaffinized, and rehydrated before unmasking in targetretrieval solution (EDTA and citrate, DAKO Co., Carpinteria, CA) in amicrowave for 30 minutes. Using a DAKO autostainer, endogenous

www.aacrjournals.orgClin Cancer Res 2005;11(9) May1, 2005 3588

Cancer Prevention



peroxidase activity was quenched in 3% hydrogen peroxide in methanolfollowed by avidin/biotin and serum-free protein blocks (DAKO).Sections were incubated for 30 minutes with 1:200 anti-HOXB2 (P-20)goat polyclonal antibody (Santa Cruz Biotechnology, Santa Cruz, CA). Astreptavidin-biotin peroxidase detection system was used accordingto the manufacturer’s instructions (LSAB label + link kit; DAKO) with3,3V-diaminobenzidine as a substrate. Counterstaining was done withMayer’s hematoxylin. HOXB2-positive breast cancer was used as apositive control (24), whilst ovary was used as a negative control.Antibody specificity was confirmed using blocking peptide sc-17165 P(Santa Cruz Biotechnology), which abrogated nuclear staining forimmunohistochemistry and eliminated a specific band on Westernblotting using pancreatic cancer cell lines. In addition, mRNA expres-sion in cell lines using reverse transcription-PCR correlated with proteinexpression on Western blotting.

Immunohistochemical scoring. Up to four separate samples ofpancreas were examined per patient. Staining was assessed by twoblinded independent observers (D.S. and J.G.K.). Standardization ofscoring was achieved by comparison of scores between observers and byconferencing, where any discrepancies were resolved by consensus.Scores were given as the percentage of nuclei staining positive withinthe representative area of the tissue microarray core and the absolute

intensity of nuclear staining on a scale of 0 to 3 (0, no staining; 1,slight/weak heterogenous nuclear staining; 2, strong homogenousnuclear staining; and 3, intense homogenous nuclear staining). Thecriteria to achieve a positive score were HOX B2 nuclear intensity of >1in >20% of nuclei.

Statistical analysis. Kaplan-Meier and the Cox proportional hazardsmodel were used for univariate and multivariate analysis using Statview5.0 Software (Abacus Systems, Berkeley, CA). P < 0.05 was accepted asstatistically significant. Those factors that were prognostic on univariateanalysis were assessed in a multivariable model to identify factors thatwere independently prognostic and those that were the result ofconfounding. This analysis was done sequentially on all patients whohad available tissue (n = 128) and on a subgroup of patients whounderwent operative resection (n = 76).

Results

Transcript profiling data analysis. Whereas previous tran-script profiling studies have been limited to identifying singlegenes aberrantly expressed in pancreatic cancer (29, 30), weemployed a strategy that used GenMAPP software to identify


Fig.1. A customizedGenMAPPof RA signaling componentswith statistically significant relative expression levels in pancreatic cancer (PC) comparedwithnormal pancreas.Relative expression levels are represented as average fold change with those with statistically significant up-regulation marked red, those with statistically significantdown-regulationmarked blue (a < 0.05 on t test and/or Mann-WhitneyU test) and no change (NC) marked green for (A) RARs, (B) cellular RA-binding proteins (CRABP),(C) downstream targets of RA signaling previously described to be aberrantly expressed in pancreatic cancer and PanIN, (D) downstream targets of RA signaling aberrantlyexpressed in this study, (E) krox20-mediated regulation of HOXB2 and HOXB1expression in normal hindbrain development. Ps presented inTable 2. Abbreviations notmentioned in text: RBP4, retinoid binding protein 4; CRALBP, cellular retinaldehyde binding protein; RALDH, retinaldehyde dehydrogenase; hRADH, retinol dehydrogenasehomologue.

HOXB2 and Pancreatic Cancer



molecular pathways in which a significant proportion of genesshowed aberrant expression. Using this approach, we confirmedaberrations in molecular pathways known to be important inpancreatic cancer (transforming growth factor–h signaling, cellcycle regulation, and apoptosis; data not shown). In addition,we identified aberrant expression of a significant number ofcomponents of RA signaling (Table 2; Fig. 1). RAR-a and RAR-gwere up-regulated 2.9- and 2.2-fold, respectively, in pancreaticcancer compared with normal pancreas. Expression of asubstantial number of known RA-responsive genes was alsoaltered in pancreatic cancer, consistent with dysregulated RAsignaling activity, primarily demonstrating up-regulation ofgenes downstream of RAR-a. A substantial number of genesregulated by RA and known to be highly expressed in pancreatic

cancer and PanIN from other studies, were also up-regulated:S100 calcium binding protein P (S100P ; ref. 31; 152-fold),MUC4 mucin (ref. 32; 24.6-fold), matrix metalloproteinase 9(MMP9 ; ref. 33; 2.0-fold), Id-1 (ref. 34; 2.3-fold), urokinaseplasminogen activator receptor (uPAR ; ref. 35; 13.5-fold), andheparin-binding epidermal growth factor-like growth factor(HB-EGF ; ref. 36; 2.5-fold; Table 2). Other genes, yet to becharacterized in pancreatic cancer but thought to be regulated byRA were also aberrantly expressed, including a RA-induced G-protein–coupled receptor (26.3-fold) and RAR respondersRARRES 1 (16.5-fold) and RARRES 3 (3.8-fold; Table 2).

Studies of hindbrain development have provided the greatestinsights into the mechanism of RA signaling. RA-dependentlineage restriction in rhombomeres 3 and 5 is marked by


Table1. Clinicopathologic and outcome data for all patients in the cohort

VariableWhole cohort

no. (%)Median

survival (mo) P (log-rank)Resected

cohort no. (%)Median

survival (mo) P (log-rank)

Sex 128 76Male 72 (56) 45 (59)Female 56 (44) 31(41)

Age (y) 128 76Mean 63.8 61Median 66.5 65Range 34-86 34-83

Treatment 128Resection 76 (59) 11Operative biopsy 46 (36) 3.9 <0.0001No operative intervention 6 (5)Outcome 128 76

Follow-up (mo) 0-117 0.2-117Median 7.6 1130-dmortality 2 (3)Death frompancreatic cancer 114(89) 63 (83)Death from other cause 2 (2) 2 (3)Alive 8 (6) 8 (11)Lost to follow-up 4 (3) 3 (4)

Stage 127I 27 (21)II 13 (10) 13.7III 70 (55)IV 17 (13) 6.4 <0.0001

Differentiation 127Well 11 (9) 7 (9)Moderate 68 (53) 8.9 44 (58) 12.2Poor 48 (38) 5 0.0152 25 (33) 8.6 0.0582

Tumor size (mm)V20 15 (20) 17.1>20 61 (80) 9.7 0.0375

MarginsClear 40 (53) 14.5Involved 36 (47) 8.5 0.0014

Lymphnode statusPositive 39 (51) 9.2Negative 35 (46) 13.8 0.0235

HOXB2 expression 128 76Positive 48 (38) 5 16 (21) 6.75Negative 80 (72) 9.9 <0.0001 60 (79) 14.0 <0.0001

Cancer Prevention



krox20 and HOXB2 expression. RA, through an as yetunknown mechanism that may involve CEBPh (37), resultsin increased krox20 expression, which in turn increasesHOXB2 expression by directly binding promoter elements ofHOXB2 (38). krox20 also suppresses HOXB1 expression. Theexpression profile in the present study is consistent withactivity of this pathway of HOXB2 regulation (Fig. 1). Inaddition, the variant promyelocytic leukemia fusion proteinPLZF-RARA also regulates HOXB2 expression through asimilar mechanism and is thought to be important inpromyelocytic leukemia development and resistance to RAtherapy (39). For these reasons, HOXB2 , a previouslyuncharacterized gene in pancreatic cancer, which showed a6.7-fold increase (P < 0.001) compared with normal pancreas,was selected for further study.

HOXB2 expression in pancreatic cancer and pancreatic intra-epithelial neoplasia. Representative examples of HOXB2immunostaining are shown in Fig. 2. Nuclear expression wasidentified in 48 of 128 cancers (38%). When HOXB2expression was present within the tumor, >80% of the nucleistained positively. HOXB2 expression was detected in thehistologically normal pancreatic ducts of 2 of 26 (8%) patients,in 1 of 24 (4%) PanIN-1A lesions, 3 of 20 (15%) PanIN-1B, 3 of10 (30%) PanIN-2, and 1 of 4 (25%) PanIN-3 lesions, showingthat HOXB2 expression occurs in PanIN and may play a role inthe evolution of PanIN.

HOXB2 expression in the whole cohort was associated with apoor outcome (median survival, 5 versus 9.9 months; log-rankP < 0.0001; Fig. 3A). In addition, operative resection (P <0.0001), low-stage (P < 0.0001), and non–poorly differentiated


Table 2. Components ofRA signalingwith differential expressionbetweenpancreatic cancer andnormalpancreas onAffymetrix U133microarrays

Probe set Unigene cluster Gene name Fold change P

204351_at Hs.2962 S100 (calcium-binding protein P) 152 0.001203108_at Hs.194691 GPCR (RA-induced 3) 26.3 0.007217109_at Hs.198267 MUC4 24.6 0.001206392_s_at Hs.82547 RAR responder1 (RARRES1) 16.5 0.004205366_s_at Hs.98428 HOXB6 14.4 0.009211924_s_at Hs.179657 UPAR 13.5 0.002202859_x_at Hs.624 Interleukin 8 (IL8) 12.5 0.001205453_at Hs.2733 HOXB2 6.7 0.001219799_s_at Hs.179608 Retinol dehydrogenase homologue (hRADH) 4.7 0.016203596_s_at Hs.27610 RA- and IFN-inducible protein (IFT5) 4.2 0.010204070_at Hs.17466 RAR responder 3 (RARRES 3) 3.8 0.005228601_at Hs.93574 HOXD3 3.4 0.024205249_at Hs.1359 Krox20 (EGR2) 3.2 0.008231936_at Hs.40408 HOXC9 3.2 0.009213844_at Hs.37034 HOXA5 3.0 0.035203749_s_at Hs.250505 RAR-a 2.9 0.007201042_at Hs.512708 TGM2 2.9 0.001202510_s_at Hs.101382 TNFAIP2 2.9 0.004204420_at Hs.283565 FOSL1 (FOS-like antigen-1) 2.8 0.005205601_s_at Hs.22554 HOXB5 2.7 0.003206858_s_at Hs.820 HOXC6 2.6 0.016202575_at Hs183650 Cellular RA-binding protein 2 (CRABP2) 2.6 0.01838037_at Hs.799 HB-EGF 2.5 0.0092214782_at Novel gene similar to retinaldehyde-binding

protein (sRABP)2.4 0.036

208937_s_at Hs.75424 Id-1 (inhibitor of DNA binding1) 2.3 0.039201505_at Hs.82124 Laminin b1 2.2 0.026204118_s_at Hs.1497 RAR-c 2.2 0.049212501_at Hs.99029 CEBP b (CCAATenhancer-binding protein h) 2.1 0.005203936_s_at Hs.151738 MMP9 2.0 0.008221701_s_at Hs.24553 STRA6 1.9 0.030202449_s_at Hs.20084 Retinoid X receptor a (RXR-a) 0.4 0.005231906_at Hs.301963 HOXD8 0.6 0.010202882_x_at Hs.106346 RA-repressible protein (RARG-1) 0.6 0.045207914_x_at Hs.336963 Even-skipped homeobox1 (EVX1) 0.5 0.016208224_at Hs.99992 HOXB1 0.4 0.010205883_at Hs.37096 Zinc finger protein145 (ZNF145, PLZF) 0.3 0.002203423_at Hs.101850 Cellular RA-binding protein1 (CRABP1) 0.2 0.001209496_at Hs.37682 RAR responder 2 (RARRES 2) 0.1 0.001





Fig. 2. Photomicrographs of HOXB2nuclear staining (magnification,�200). A , negative HOXB2expression in normal pancreatic duct.B, positive nuclear staining inPanIN-1B lesion. C, positive nuclearstaining in PanIN-3 lesion. D, positivenuclear staining in pancreatic cancer.E, negative nuclear staining inPanIN-1A lesion. F, negative nuclearstaining in pancreatic cancer. G,HOXB2 staining in pancreatic cancerwithout blocking peptide. H, serialsection of (G) stained withanti-HOXB2 antibody after incubationwith blocking peptide sc-17165 P.

Cancer Prevention



tumors (P = 0.0152) were associated with significantlyimproved survival using Kaplan-Meier analysis. However,multivariate analysis identified resection and stage as the onlyindependent prognostic factors when modeled together withdegree of differentiation and HOXB2 status (Table 3A). WhereasHOXB2 expression was identified in 32 of 52 (62%) unresectedtumors, it was present in only 16 of 76 (21%) resectedpancreatic cancers. Hence, HOXB2 expression was associatedwith nonresectable tumors (m2; P < 0.0001) and consequentlywas not an independent prognostic factor. Operative resectiondid not benefit those patients whose tumors expressedHOXB2 (log-rank P = 0.37; Fig. 3B) but was beneficial tothose patients who did not express HOXB2 (median survival,

14.0 versus 3.7 months; log-rank P < 0.0001; Fig. 3C).Survival for patients with tumors that were HOXB2 negativeand who underwent resection was significantly longer thansurvival in all other groups (14 versus 4.3 months; log-rankP < 0.0001; Fig. 3D). Hence, in this cohort, lack of HOXB2expression cosegregated with operative resectability, with onlythose who were HOXB2 negative having a survival advantagefrom operative resection.

Survival analysis of patients that underwent operativeresection identified decreased survival associated with HOXB2nuclear expression (median survival, 6.75 versus 14.0 months;log-rank P < 0.0001; Fig. 3E). Kaplan-Meier analyses identifiedclear margin status (P = 0.0014), tumor size of V20 mm


Fig. 3. Kaplan-Meier survival curves for (A)HOXB2 nuclear expression in thewholecohort. Effect of resection on prognosis inthe following subgroups: (B) HOXB2positive, (C) HOXB2 negative, (D) allpatients stratified for HOXB2 status andresection. Kaplan-Meier survival curves for76 patients who underwent surgicalresection: (E) HOXB2 nuclear expression,(F) margin status, (G) tumor size, and (H)lymphnode involvement.




(P = 0.0375), and no lymph node involvement (P = 0.0235) asbeing associated with a survival advantage (Fig. 3F-H). Degreeof differentiation was not associated with a survival advantage(P = 0.0582). HOXB2 expression and involved surgical marginswere independent prognostic factors when modeled against allcombinations of HOXB2 expression, involved surgical margin,lymph node involvement, and tumor size in the subgroup ofpatients who underwent surgical resection (Table 3B-D).

Discussion

Expression profiling identified differential expression of asignificant number of RA signaling pathway components anddownstream responders in pancreatic cancer compared withnormal pancreas. These included some genes that are known tobe associated with pancreatic cancer: MUC4, MMP9, Id-1 ,uPAR , HB-EGF , and S100P , as well as novel candidates.Although there is substantial evidence implicating aberrantretinoid signaling in carcinogenesis (e.g., acute promyelocyticleukemia; ref. 40), the mechanisms by which retinoid targetgenes exert these effects remains to be elucidated. Here wepresent evidence, implicating RA signaling in pancreatic cancerand show that a RA-responsive homeodomain transcriptionfactor, HOXB2, is ectopically expressed in a significantproportion of pancreatic cancer, with a profound associationwith tumor progression. HOXB2, which is not normallyexpressed in the pancreas at any stage during development oradult life, was expressed in 38% of pancreatic cancers andseemed to occur during the development of a proportion ofPanIN. Ectopic HOXB2 expression was associated with non-resectable tumors and was an independent prognostic factor inresected tumors when modeled with known clinicopathologicprognostic factors. In addition, only those patients that wereHOXB2 negative obtained a survival advantage with operativeresection.

Numerous lines of evidence from separate studies suggestthe importance of individual RA signaling components inpancreas development and pancreatic cancer evolution. RAregulates early instructive signals from lateral plate mesodermthat is essential for specification of endoderm towards a

pancreatic fate (41). RARs also regulate exocrine pancreaticdevelopment at later stages (42), by modulating lineageselection favoring ductal rather than acinar differentiationprimarily through RAR-a (10, 11). Reactivation of develop-mental pathways, specifically those that regulate exocrine celllineage, have been implicated previously in the early develop-ment of pancreatic cancer (25) and other pathways thatdetermine duct cell versus acinar cell differentiation involvingRA signaling, may also be important. In addition, forcedexpression of cellular retinol binding protein (CRABP1), amediator of RA signaling, in transgenic mice results in thedevelopment of poorly differentiated pancreatic cancer (43),further supporting a role of aberrant RA signaling in pancreaticcancer evolution. The transcript profile data presented heresuggests the RA signaling pathway has a role in pancreaticcancer, specifically, a number of RA-responsive genes known tobe important in pancreatic cancer and PanIN developmentwere aberrantly expressed in this study: MUC4 mucin isoverexpressed in a significant proportion of pancreatic cancer(44) and PanIN (32) and can be induced through RAR-aactivation (45); similarly, MMP9 is expressed in pancreaticcancer (33) and is up-regulated by RA treatment (12) as isuPAR , HB-EGF , and p21WAF1/CIP1 (46). Id-1, which antago-nizes basic helix loop helix proteins, inhibits differentiationand can enhance cell proliferation is overexpressed in PanINlesions (34) and is also RA responsive (47). In the presentstudy, HOXB2 expression was also detected in PanIN lesions.RA seems to exert its effect on exocrine lineage selectiontowards a ductal phenotype during development throughlaminin-h1 (11), which was up-regulated in the present study.As was the putative tumor suppressor gene RARRES 3 (48).Stra6 , a gene whose function is yet to be determined respondsto RA, is up-regulated in colon cancer (49) and was up-regulated in this study as was transglutaminase 2 (TGM2) andtumor necrosis factor a–induced protein 2 (TNFAIP2), bothalso RA-responsive genes (47). There is clear evidencesupporting RA regulation of HOXB2 expression from studiesof hindbrain patterning and branchial arch development (38);however, the mechanism by which RA signaling imparts itseffects on the HOX network and cellular function is poorly


Table 3. Multivariate analysis for clinicopathologic variables andHOXB2 expression in thewhole cohort and resectedpancreatic cancer

Variable Hazards ratio (95% confidence interval) P

A.Whole cohort (n = 127) Stage III/IV versus I/II 2.30 (1.44-3.69) 0.005Resection 0.43 (0.26-0.72) 0.0013HOXB2 expression 1.56 (0.94-2.57) 0.085

B. Resected (n = 74) HOXB2 expression 2.90 (1.51-5.57) 0.0014Margin involvement 1.89 (1.02-3.48) 0.0428Lymphnode involvement 1.30 (0.71-2.40) 0.3981

C. Resected (n = 76) HOXB2 expression 2.82 (1.48-5.40) 0.0017Margin involvement 2.04 (1.17-3.53) 0.0115Tumor size, >20 mm 1.48 (0.75-2.90) 0.2567

D. Resected (n = 74) HOXB2 expression 2.69 (1.39-5.20) 0.0032Margin involvement 1.75 (0.94-3.25) 0.0777Lymphnode involvement 1.34 (0.73-2.46) 0.3525Tumor size, >20 mm 1.49 (0.76-2.94) 0.2474

Cancer Prevention



understood. There were, however, some inconsistencies in thedata where downstream targets of RA were down-regulatedsuch as HOXD8. Presumably, other mechanisms can alsoregulate the expression levels of these transcripts other thanthrough RA signaling alone. Validation of these data withmanipulation of RA signaling is required to further investigatea putative functional role in pancreatic cancer; however, thestrength of data from the literature and evidence presentedhere makes a strong case for an important role in pancreaticcancer.

Multivariate analysis identified HOXB2 expression as anindependent predictor of survival in the subgroup of patientsthat underwent pancreatic resection. Although HOXB2 expres-sion was not identified as an independent predictor of survivalin the whole cohort due to its association with resection, lack ofHOXB2 expression combined with surgical resection conferred asignificant survival advantage. Because all known prognosticindicators in pancreatic cancer, such as tumor size, resectionmargins, and lymph node status can only be determined postresection, HOXB2 expression has potential utility as a prognos-tic indicator in pancreatic cancer, especially because it seems tohave a profound independent influence on survival, with theadvantage that it can be assessed using biopsy techniqueswithout resection. We have previously identified that loss ofDPC4/Smad4 expression is associated with poor outcome inpancreatic cancer (28). However, HOXB2 may have more

efficacy as a marker of prognosis in pancreatic cancer, as ectopicexpression of a protein is a more reliable indicator than loss ofexpression. Although pancreatic resection offers the best chanceof cure in patients with pancreatic cancer, it is a procedure whichcarries significant morbidity and mortality. The development ofa reliable preoperative assessment of HOXB2 status would be animportant addition to a physician’s limited diagnostic arma-mentarium in this disease and may be used, together withcurrent clinicopathologic variables of disease outcome, todetermine a patient’s suitability for operative resection.

In conclusion, gene expression profiling of pancreatic cancerhas suggested that RA signaling is a potentially importantregulatory pathway in pancreatic cancer evolution. Ectopicexpression of HOXB2 in pancreatic cancer is a frequentoccurrence, an event which manifests in the development ofPanIN in a proportion of cases, and is possibly a consequence ofaberrant RA signaling. Current prognostic factors for pancreaticcancer remain poorly defined, depend upon examination of theresected pancreas, and cannot be accurately determined preop-eratively. Assessment of HOXB2 expression may provide analternative method for determining the suitability for resectionand the prognosis of patients with pancreatic cancer. Furtherstudy to determine the effects of ectopic HOXB2 expression andother components of the HOX transcriptional network, itsrelationship to RA signaling, and clinical utility in pancreaticadenocarcinoma is required.



References1. Yeo CJ, Cameron JL, Sohn TA, et al. Six hundredfifty consecutive pancreaticoduodenectomies in the1990s: pathology, complications, and outcomes.Ann Surg 1997;226:248^57; discussion 257^60.

2. Hruban RH, Adsay NV, Albores-Saavedra J, et al.Pancreatic intraepithelial neoplasia: a new nomencla-ture and classification system for pancreatic ductlesions. AmJSurg Pathol 2001;25:579^86.

3. Hruban RH, Takaori K, Klimstra DS, et al. An illus-trated consensus on the classification of pancreaticintraepithelial neoplasia (PanIN) and intraductalpapillary mucinous neoplasms (IPMNs). Am J SurgPathol 2004;28:977^87.

4. Moskaluk CA, Hruban RH, Kern SE. p16 and K-rasgene mutations in the intraductal precursors of humanpancreatic adenocarcinoma. Cancer Res 1997;57:2140^3.

5. van Heek NT, Meeker AK, Kern SE, et al. Telomereshortening is nearly universal inpancreatic intraepithe-lial neoplasia. AmJPathol 2002;161:1541^7.

6. Biankin AV, Kench JG, MoreyAL, et al. Overexpres-sion of p21WAF1/CIP1is an early event in the develop-ment of pancreatic intraepithelial neoplasia. CancerRes 2001;61:8830^7.

7.Wilentz RE, Geradts J, Maynard R, et al. Inactivationof the p16 (INK4A) tumor-suppressor gene in pan-creatic duct lesions: loss of intranuclear expression.Cancer Res 1998;58:4740^4.

8. DiGiuseppe JA, Hruban RH, Goodman SN, et al.Overexpression of p53 protein in adenocarcinoma ofthe pancreas. AmJClin Pathol1994;101:684^8.

9. OosterveenT, vanVliet P, Deschamps J, Meijlink F.The direct context of a hox retinoic acid response ele-ment is crucial for its activity. J Biol Chem 2003;278:24103^7.

10.Tulachan SS, Doi R, Kawaguchi Y, et al. All-transretinoic acid induces differentiation of ducts andendocrine cells by mesenchymal/epithelial inter-actions in embryonic pancreas. Diabetes 2003;52:76^84.

11. Kobayashi H, SpildeTL, Bhatia AM, et al. Retinoidsignaling controls mouse pancreatic exocrine lineage

selection through epithelial-mesenchymal interac-tions. Gastroenterology 2002;123:1331^40.

12. Jaster R, Hilgendorf I, Fitzner B, et al. Regulation ofpancreatic stellate cell function in vitro : biological andmolecular effects of all-trans retinoic acid. BiochemPharmacol 2003;66:633^41.

13.Mangelsdorf DJ, Evans RM.The RXR heterodimersand orphan receptors. Cell1995;83:841^50.

14. OosterveenT, Niederreither K, Dolle P, Chambon P,Meijlink F, DeschampsJ. Retinoids regulate the anteri-or expressionboundaries of 5VHoxb genes inposteriorhindbrain. EMBOJ 2003;22:262^9.

15. Duboule D, Morata G. Colinearity and functionalhierarchy among genes of the homeotic complexes.Trends Genet 1994;10:358^64.

16. Abate-Shen C. Deregulated homeobox gene ex-pression in cancer: cause or consequence? Nat RevCancer 2002;2:777^85.

17. Apiou F, Flagiello D, Cillo C, Malfoy B, Poupon MF,Dutrillaux B. Fine mapping of human HOX gene clus-ters. Cytogenet Cell Genet1996;73:114^5.

18. Graham A, Papalopulu N, Krumlauf R. The murineand Drosophila homeobox gene complexes havecommon features of organization and expression. Cell1989;57:367^78.

19. Cillo C, Faiella A, CantileM, Boncinelli E. Homeoboxgenes and cancer. Exp Cell Res1999;248:1^9.

20. Cillo C, Barba P, Freschi G, Bucciarelli G, Magli MC,Boncinelli E. HOX gene expression in normal and neo-plastichumankidney. IntJCancer1992;51:892^7.

21. De Vita G, Barba P, Odartchenko N, et al. Expres-sion of homeobox-containing genes in primary andmetastatic colorectal cancer. Eur J Cancer 1993;29A:887^93.

22. Naora H,YangYQ, Montz FJ, Seidman JD, KurmanRJ, RodenRB. A serologically identified tumor antigenencoded by a homeobox gene promotes growth ofovarian epithelial cells. Proc Natl Acad Sci U S A2001;98:4060^5.

23. LewisMT. Homeobox genes inmammary gland de-velopment and neoplasia. Breast Cancer Res 2000;2:158^69.

24. Cantile M, Pettinato G, Procino A, Feliciello I,Cindolo L, Cillo C. In vivo expression of the wholeHOX gene network in human breast cancer. Eur JCancer 2003;39:257^64.

25. MiyamotoY, Maitra A, Ghosh B, et al. Notch medi-atesTGF a-induced changes in epithelial differentia-tion during pancreatic tumorigenesis. Cancer Cell2003;3:565^76.

26.Thayer SP, diMaglianoMP, Heiser PW, et al.Hedge-hog is an early and late mediator of pancreatic cancertumorigenesis. Nature 2003;425:851^6.

27. Baugh LR, Hill AA, Brown EL, Hunter CP.Quantitative analysis of mRNA amplification byin vitro transcription. Nucleic Acids Res 2001;29:e29.

28. Biankin AV, Morey AL, Lee C-S, et al. DPC4/Smad4 expression and outcome in pancreaticductal adenocarcinoma. J Clin Oncol 2002;20:4531^42.

29. Iacobuzio-Donahue CA, Ashfaq R, Maitra A, et al.Highly expressed genes in pancreatic ductal adeno-carcinoma: a comprehensive characterization andcomparison of the transcript profiles obtained fromthree major technologies. Cancer Res 2003;63:8614^22.

30. Logsdon CD, Simeone DM, Binkley C, et al. Mole-cular profiling of pancreatic adenocarcinoma andchronic pancreatitis identifies multiple genes differen-tially regulated in pancreatic cancer. Cancer Res2003;63:2649^57.

31. Crnogorac-JurcevicT, Missiaglia E, Blaveri E, et al.Molecular alterations in pancreatic carcinoma: expres-sion profiling shows that dysregulated expression ofS100 genes is highly prevalent. J Pathol 2003;201:63^74.

32. Swartz MJ, Batra SK, Varshney GC, et al. MUC4expression increases progressively in pancreaticintraepithelial neoplasia. Am J Clin Pathol 2002;117:791^6.

33. GongYL, Xu GM, HuangWD, Chen LB. Expressionof matrix metalloproteinases and the tissue inhibitorsof metalloproteinases and their local invasiveness




and metastasis in Chinese human pancreatic cancer.JSurg Oncol 2000;73:95^9.

34. Maruyama H, Kleeff J,Wildi S, et al. Id-1 and Id-2are overexpressed in pancreatic cancer and in dys-plastic lesions in chronic pancreatitis. Am J Pathol1999;155:815^22.

35. Cantero D, Friess H, Deflorin J, et al. Enhanced ex-pressionof urokinase plasminogen activator and its re-ceptor in pancreatic carcinoma. Br J Cancer 1997;75:388^95.

36. Ito Y, Higashiyama S, Takeda T, Yamamoto Y,Wakasa KI, Matsuura N. Expression of heparin-binding epidermal growth factor-like growth factorin pancreatic adenocarcinoma. Int J GastrointestCancer 2001;29:47^52.

37. Duprez E,Wagner K, Koch H,Tenen DG. C/EBPb: amajor PML-RARA responsive gene in retinoic acid-induced differentiation of APL cells. EMBO J 2003;22:5806^16.

38. Gavalas A. ArRAnging the hindbrain.TrendsNeuro-sci 2002;25:62^4.

39. Ivins S, Pemberton K, Guidez F, Howell L,

Krumlauf R, Zelent A. Regulation of Hoxb2 byAPL-associated PLZF protein. Oncogene 2003;22:3685^97.

40. Melnick A, Licht JD. Deconstructing a disease:RARa, its fusion partners, and their role in the patho-genesis of acute promyelocytic leukemia. Blood1999;93:3167^215.

41. Kumar M, Jordan N, Melton DA, Grapin-Botton A.Signals from lateral plate mesoderm instruct endo-derm towards a pancreatic fate. Dev Biol 2003;259:109^22.

42. Kadison A, Kim J, Maldonado T, et al. Retinoidsignaling directs secondary lineage selection inpancreatic organogenesis. J Pediatr Surg 2001;36:1150^6.

43. GiguereV. Retinoic acid receptors and cellular reti-noid binding proteins: complex interplay in retinoidsignaling. Endocr Rev1994;15:61^79.

44. Balague C, Gambus G, Carrato C, et al. Altered ex-pression of MUC2,MUC4, andMUC5mucin genes inpancreas tissues and cancer cell lines. Gastroentero-logy1994;106:1054^61.

45. Choudhury A, Singh RK, Moniaux N, El-MetwallyTH, Aubert JP, Batra SK. Retinoic acid-dependenttransforming growth factor-h 2-mediated inductionof MUC4 mucin expression in human pancreatic tu-mor cells follows retinoic acid receptor-a signalingpathway. J Biol Chem 2000;275:33929^36.

46. Liu T-X, Zhang J-W, Tao J, et al. Gene expressionnetworks underlying retinoic acid-induced differentia-tion of acute promyelocytic leukemia cells. Blood2000;96:1496^504.

47.Ma Y, Koza-Taylor PH, DiMattia DA, et al. Micro-array analysis uncovers retinoid targets in humanbronchial epithelial cells. Oncogene 2003;22:4924^32.

48. DiSepio D, Ghosn C, Eckert RL, et al. Identificationand characterization of a retinoid-induced class IItumor suppressor/growth regulatory gene. Proc NatlAcad Sci US A1998;95:14811^5.

49. SzetoW, JiangW,Tice DA, et al. Overexpression ofthe retinoic acid-responsive gene Stra6 in humancancers and its synergistic induction by Wnt-1 andretinoic acid. Cancer Res 2001;61:4197^205.

Cancer Prevention



2005;11:3587-3596. Clin Cancer Res Davendra Segara, Andrew V. Biankin, James G. Kench, et al. Pancreatic Cancer and Pancreatic Intraepithelial NeoplasiaExpression of HOXB2, a Retinoic Acid Signaling Target in

Updated version

http://clincancerres.aacrjournals.org/content/11/9/3587

Access the most recent version of this article at:

Cited articles

http://clincancerres.aacrjournals.org/content/11/9/3587.full#ref-list-1

This article cites 47 articles, 16 of which you can access for free at:

Citing articles

http://clincancerres.aacrjournals.org/content/11/9/3587.full#related-urls

This article has been cited by 19 HighWire-hosted articles. Access the articles at:

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

Subscriptions

Reprints and

[email protected] at

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

Permissions

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

.http://clincancerres.aacrjournals.org/content/11/9/3587To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/content/11/9/3587.full#ref-list-1

http://clincancerres.aacrjournals.org/content/11/9/3587.full#related-urls

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

mailto:[email protected]



expressionofhoxb2,aretinoicacidsignalingtargetinpancreatic ...transcription factor essential for...

Documents