Molecular and Metabolic Subtypes Correspondence for Pancreatic Ductal Adenocarcinoma Classification

Jou nal o Clinical Medicine Re iew Molecula and Me abolic Sub ypes Co espondence o Panc ea ic Duc al Adenoca cinoma Classi ica ion Pila Espiau-Rome a , Sa ah Cou ois, Bea iz Pa ejo-Alonso and Pa icia Sancho * T ansla ional Resea ch Uni , Hospi al Uni e si a io Miguel Se e , IIS A agon, 50009 Za agoza, Spain; [email p o ec ed] (P.E.-R.); [email p o ec ed] (S.C.); [email p o ec ed] (B.P.-A.) *Co espondence: [email p o ec ed] Recei ed: 14 Oc obe 2020; Accep ed: 17 Decembe 2020; Published: 21 Decembe 2020   Abs ac : Panc ea ic duc al adenoca cinoma (PDAC), he mos common o m o panc ea ic cance , is an ex emely le hal disease due o la e diagnosis, agg essi eness and lack o e ec i e he apies. Conside ing i s in insic he e ogenei y, pa ien s a i ica ion models based on ansc ip omic and genomic signa u es, wi h pa ially o e lapping subg oups, ha e been es ablished. Besides molecula al e a ions, PDAC umou s show a s ong desmoplas ic esponse, esul ing in p o ound me abolic ep og amming in ol ing inc eased glucose and amino acid consump ion, as well as lipid sca enging and biosyn hesis. In e es ingly, ecen wo ks ha e also e ealed he exis ence o me abolic sub ypes wi h di e en ial p ognosis wi hin PDAC, which co ela ed o de ined molecula subclasses in pa ien s: lipogenic sub ype co ela ed wi h a classical/p ogeni o signa u e, while glycoly ic umou s associa ed wi h he highly agg essi e basal/squamous p o ile. Bioin o ma ic analyses ha e demons a ed ha he ep esen a i e genes o each me abolic sub ype a e up- egula ed in PDAC samples and p edic pa ien su i al. This sugges s a ela ionship be ween he gene ic signa u e, me abolic p o ile, and agg essi eness o he umou . Conside ing all his, de ining me abolic sub ypes ep esen s a clea oppo uni y o pa ien s a i ica ion conside ing umou unc ional beha iou independen ly o hei mu a ional backg ound. Keywo ds: PDAC; panc ea ic cance ; glycolysis; lipid me abolism; classi ica ion; s a i ica ion 1. In oduc ion Panc ea ic duc al adenoca cinoma (PDAC) is conside ed one o he mos agg essi e solid malignancies. I ep esen s he hi d cause o cance - ela ed dea hs in indus ialised coun ies oday [ 1 ] and i is p edic ed o become he second by 2030 [ 2 ]. Despi e p og ess in he unde s anding o he molecula and gene ic basis o his disease, i e-yea su i al a es ha e emained below 10% a e diagnosis and one-yea su i al occu s in only 28% o cases [3]. The easons o he poo p ognosis o his disease include bad accessibili y o he o gan, absence o dis inc symp omsandhigh a eo me as asis, occu inginabou 50%o pa ien s[ 4 ]. On he o he hand, he e a e no eliable bioma ke s app o ed o ea ly diagnosis [5]. Cu en ly, he only a ailable cu a i e op ion o PDAC pa ien s is su gical esec ion ollowed by adju an chemo he apy [ 4 ]. Howe e , his occu s in a mino i y o pa ien s, since 80–90% o hem a e diagnosed wi h ad anced disease when he umou is no esec able [ 6 , 7 ]. In addi ion, he he e ogenei y andplas ici yo PDAC umou slead ochemo esis ance[ 8 ]. In ac , se e alphaseIII ials o chemo he apy agen s o a ge ed he apies e ec i e in o he malignancies ha e ailed o bene i unselec ed PDAC popula ions. In addi ion, wi h he excep ion o he a e subse o misma ch epai -de icien umou s, checkpoin inhibi o s ha e ailed o show e icacy in me as a ic pa ien s. Unde such ci cums ances, he solu ion possibly lies in ea ly de ec ion and p ope classi ica ion o pa ien s [ 4 ]. Pa ien s a i ica ion has become an in aluable ool o he clinical managemen o cance pa ien s, p o iding diagnos ic and p ognos ic in o ma ion and, c ucially, guiding he apeu ic J. Clin. Med. 2020,9, 4128; doi:10.3390/jcm9124128 www.mdpi.com/jou nal/jcm J. Clin. Med. 2020,9, 4128 2 o 21 decisions, especially when a ge ed he apies o a speci ic mu a ion o subg oup a e a ailable. In ac , s a i ica ion had a key ole imp o ing su i al a es in diseases such as b eas cance . Howe e , pa ien s a i ica ion based on his omo phological o molecula ea u es in PDAC has p o ed challenging, se e ely delaying he iden i ica ion o such a ge ed he apies. This, oge he wi h he und uggable na u e o mu a ed K-RAS (al hough clinical ials a e unde way o inhibi he mu a ion K-RAS G12C ), limi s PDAC ea men o ine ec i e con en ional chemo he apy. Genomic s udies ha e e ealed sub ypes o PDAC based on hei molecula ea u es, bu di e se s udies ha e epo ed di e en classi ica ion sys ems incomple ely o e lapping. In e es ingly, inc easing e o s a e being made in di e en di ec ions. In ac , ecen da a sugges ha PDAC umou s could also be classi ied a ending o hei me abolic pheno ype. In his e iew, we desc ibe and analyse he co espondence o mos o he published molecula and me abolic signa u es o PDAC and p opose a modi ied me abolic signa u e ha could s a i y pa ien s acco ding o me abolic needs independen ly o he mu a ional load. 2. PDAC Mu a ional P o ile Many s udies ha e been conduc ed on PDAC samples o de e mine molecula abe a ions a DNA, RNA, p o ein and epigene ic le els [ 9 ]. In gene al e ms, mu a ions in indi idual genes comp ise 10 main pa hways: DNA damage epai (TP53 o BRCA2), cell cycle egula ion (Cyclin-Dependen Kinase Inhibi o 2A,CDKN2A),T ans o mingg ow h ac o -be a(TGF- β )pa hway (SMAD4), NOTCHandWNT signalling, ch oma in egula ion, Ex acellula Signalling Regula ed Kinases (ERK)-Mi ogen-Ac i a ed P o ein Kinases (MAPK) signalling (K-RAS,B-RAF), axonal guidance and RNA p ocessing [10,11]. Al oge he , K-RAS gain o unc ion, accoun ing o up o 90% o PDAC cases, and inac i a ion o he umou supp esso genes p16/CDKN2A,SMAD4 and TP53 con ibu e o PDAC o ma ion, main enance, p og ession and, ul ima ely, me as asis onse . Occasionally, he mu a ions occu in o he componen s o he pa hway in which hey a e in ol ed [ 9 , 12 , 13 ]. In addi ion, genome-wide and exome-wide analyses ha e iden i ied a long lis o less equen al e a ions in genes ela ed o axon o ien a ion o DNA damage epai , such as BRCA1 o BRCA2 [ 12 , 14 , 15 ]. Mo eo e , mos PDACs show complex pa e ns o ch omosomal ea angemen [16,17]. Some indings ha e sugges ed a linea pa hway o e olu ion om no mal cells o PDAC cells, in which no all mu a ions occu a he same ime. Telome e sho ening is conside ed he igge o panc ea ic umo igenesis by inducing gene ic ins abili y, coupled wi h a mu a ion in he K-RAS gene. Su i ing cells a e hen al e ed by loss o CDKN2A unc ion. In hi d place, he umou supp esso genes TP53 and SMAD4 a e inac i a ed, leading o in e media e s ages called panc ea ic in aduc al neoplasias (PanINs) and, ul ima ely, ull-blown PDAC [ 10 , 18 ]. Howe e , i seems ha in some cases, CDKN2A, TP53 and SMAD4 mu a ions can occu in a single e en , which is called “punc ua ed equilib ium” [ 11 ]. 2.1. K-RAS K-RAS is one o he h ee mu a ionally ac i a ed o ms o he Ras p o ein, oge he wi h H-RAS and N-RAS. I is he mos ep esen ed one among all cance s bea ing RAS mu a ions, and he one exclusi ely mu a ed in PDAC. Mo eo e , i is known o be he main d i e mu a ion o his disease, since K-RAS is mu a ed in >90% o he cases, hus making PDAC he mos RAS-addic ed o all cance ypes [19]. Howe e , K-RAS ac i a ion alone is no su icien o PDAC de elopmen [20,21]. Ras p o eins a e small GTPases ha ac as signalling swi ches egula ing cell p oli e a ion, su i al, di e en ia ion and in lamma ion, by ac i a ing he MAPK module. In a no mal scena io, Ras is ound in i s inac i e o m coupled o guanosine diphospha e (GDP) and, upon ac i a ion by di e se ex acellula s imuli, Ras becomes ac i e by coupling guanosine iphospha e (GTP), hus swi ching on downs eam e ec o s such as ERKs, Jun Amino-Te minal Kinases (JNKs), S ess-Ac i a ed P o ein Kinases (p38/SAPKs) and PI3K/PDK1/AKT, amongs o he s [ 22 , 23 ]. In PDAC and o he cance s bea ing oncogenic K-RAS mu a ions, his p o ein is cons i u i ely ac i e in a GTP-bound s a e. PDAC-associa ed K-RAS al e a ions mainly ha bou missense mu a ions ha depend on allele dosage o umo igenesis J. Clin. Med. 2020,9, 4128 3 o 21 and me as asis onse , and con ibu e o PDAC e olu ion om PanINs o inal cance o ma ion wi h he loss o o he key PDAC-d i ing playe s [24]. 2.2. p16/CDKN2A Along he e olu iona y cou se o PDAC onse , K-RAS mu a ion is ollowed by he occu ence o gene ic al e a ions in he umou supp esso gene CDKN2A in 95% o PDAC pa ien s [ 24 ]. Tumou supp esso genes a e impo an o es ain abe an cell p oli e a ion in he con ex o oncogenic signalling and, o ha end, hese genes ely on di e en s a egies such as cell cycle a es , apop osis o senescence induc ion. Speci ically, he gene CDKN2A, encoding o P16INK4a and P14RF p o eins, shows a dual mechanism: P16INK4a inhibi s Cyclin Dependen Kinases (CDK) 2 and 4, hus p e en ing cell cycle o p og ess in o S-phase, while P14RF igge s P53 by inhibi ing i s inhibi o , Mouse double minu e 2 homolog (MDM2), he eby inducing g ow h a es and apop osis [25]. 2.3. TP53 P53, he p o ein encoded by heTP53 gene, isknown obe he“genome gua dian”due oi sessen ial ole upon ch oma in damage. Amongs i s main unc ions, we ind cell cycle blockade and ac i a ion o DNA epai enzymes. Ul ima ely, P53 may lead he cell o a senescence s a e and/o apop osis when DNA damage is no longe epa able. Gene ic inac i a ion o TP53 is p esen in up o 75% o all PDAC cases and appea in ad anced PanINs a e loss o CDKN2A. Loss o unc ional P53 leads o an uncon olled s a e o p oli e a ion [ 25 ]. Impo an ly, mu an P53 may con ibu e o me as a ic p og ession o K-RAS-d i ing PDAC-bea ing mice, highligh ing a no el ole o his ansc ip ion ac o in la e s ages o PDAC beyond he malignan ans o ma ion o PanINs in o PDAC [26,27]. 2.4. SMAD4 SMAD4 se es as he cen al media o o TGF- β pa hway and i is known o be he ou h and, oge he wi h TP53, ul ima e d i e o PDAC ini ia ion [ 28 ]. Howe e , he ole o SMAD4 in he pa hogenesis o PDAC is complex and i s inal esul s a e con ex -dependen . On he one hand, he TGF- β pa hway is a well-accep ed epi helial- o-mesenchymal ansi ion (EMT) induce [ 29 ], and also in PDAC [ 25 ], hus highligh ing i s p o- umo igenic po en ial. Con e sely, Smad4 signalling in he KPC (LSL-K asG12D/+;LSL-T p53R172H/+;Pdx-1-C e) mouse PDAC model media es a umou supp essi e p ocess known as “le hal EMT”. In his scena io, Smad4 igge s apop osis h ough ep ession o he gas oin es inal lineage-mas e egula o Kl 5 and, consequen ly, inhibi s PDAC p og ession induced by Kl 5/Sox4. Consequen ly, he loss o SMAD4 umou supp essi e unc ion may con ibu e o panc ea ic oncogenesis h ough he TFG-βcanonical pa hway [30]. 3. PDAC Molecula Signa u es Besides pu ely mu a ional analyses, a conside able numbe o genomic and ansc ip omic s udies in pa ien samples ha e iden i ied he exis ence o PDAC sub ypes wi h p ognos ic and biological ele ance, as de ailed below. Collisson e al. [ 31 ] epo ed he i s exhaus i e ansc ip omic analysis on clinical samples, and based hei classi ica ion on gene exp ession and molecula p o iling. PDAC issues we e classi ied in o “classical”, “quasi-mesenchymal” and “exoc ine-like” (Table 1), wi h di e en p ognoses and esponses o selec ed he apies. The classical sub ype showed high exp ession o epi helial and adhesion-associa ed genes, such as he ansc ip ion ac o GATA6, and a ou able p ognosis ega ding su i al a e PDAC esec ion. On he o he hand, he quasi-mesenchymal sub ype exhibi ed high exp ession o mesenchymal-associa ed genes, was ela i ely less dependen on K-RAS han he classical one, and, impo an ly, was associa ed wi h poo p ognosis. Finally, he exoc ine-like sub ype e ealed high exp ession o diges i e exoc ine enzyme genes. Howe e , he las sub ype was no ound in cance cell lines, aising conce ns abou i s speci ici y [32]. J. Clin. Med. 2020,9, 4128 4 o 21 Table 1. Molecula Panc ea ic Duc al Adenoca cinoma (PDAC) signa u es. Au ho Type o S udy Type and Numbe o PDAC Samples Dys egula ed Pa hways and Mu a ions Ou come Collison e al. [31] T ansc ip ional 1. Clinical samples mic oa ay da ase s Mic odissec ed (n=27) GSE15471 (n=36) GSE11838 (n=107) GSE16515 (n=52) E-MEXP-950 (n=50) 2. Valida ion: Mouse cell lines (n=15) Human cell lines (n=19) Classical: (↑) Adhesion- associa ed genes (GATA6). Mo e K-RAS-dependen Good Quasi-mesenchymal: (↑) Mesenchymal associa ed genes Bad Exoc ine: (↑) Diges i e exoc ine enzyme genes Mo i e al. [33] T ansc ip ional 1. Mic oa ay da a P ima y umou (n=145) Me as a ic umou (n=61) Cell lines (n=17) Panc eas no mal samples (n=46) Dis an si e adjacen samples (n=88) 2. Valida ion P ima y umou s (n =15) PDXs (n=37) Cell lines (n=3) CAF lines (n=6) Classical: Classical Collison ((↑) adhesion-associa ed genes (GATA6)) and SMAD4 Good Basal: ( ↑ ) Genes also highly exp essed in basal umou s in bladde and b eas cance Bad No mal s oma: ( ↑ ) Panc ea ic s ella e cells, smoo h muscle ac in, imen in and desmin ma ke s Good Ac i a ed s oma: (↑) Mac ophages, umou p omo ion and ib oblas ac i a ion-associa ed genes Bad Bailey e al. [10]Mu a ional T ansc ip ional P ima y PDAC umou samples and a e acina cell ca cinoma (n=382) PDAC exomes (n=74) Squamous: Hype me hyla ion and ( ↓ ) panc ea ic endode mal cell a e genes. TP53,KDM6A and TP63∆N Bad Panc ea ic p ogeni o : (↑) Panc ea ic ea ly de elopmen genes (PDX1)Good ADEX: (↑) K-RAS ac i a ion and panc ea ic la e de elopmen and di e en ia ion genes Immunogenic: (↑) Immune supp ession and s ong immune in il a ion Zhao e al. [34]T ansc ip ional (me analysis) 1. Mic oa ay da ase s o PDAC p ima y umou samples (n=1268) TCGA (n=172) GSE79670 (n=51) TCGC PACA-AU (n=71) MTAB-1791 (n=195) ICGC a ay (n=178) GSE71729 (n=145) GSE62165 (n=118) GSE62452 (n=69) GSE57495 (n=63) GSE60980 (n=49) GSE77858 (n=46) GSE55643 (n=45) GSE15471 (n=39) L1: (↑) Me abolic genes L2: (↑) Me abolic, cell p oli e a ion and epi helium genes (CDKN2A)Bad L3: (↑) Collagen and ECM associa ed genes L4: (↑) Immune p o ile Good L5: (↑) Neu oendoc ine and insulin ela ed pa hways Good L6: (↑) Me abolic and diges i e enzyme genes Lombe k e al. [35]Epigene ic 1. PDXs (n=24) 2. Clinical samples mic oa ay da ase s GSE71729 (n=145) ICGC (n=178) TCGA (n=172) Classical: (↑) TFs in ol ed in panc ea ic de elopmen , me abolic egula o s and Ras signalling Good Basal: (↑) TF p oli e a i e and ansc ip ion nodes Bad Mau e e al. [36] T ansc ip ional Compu a ional modelling 1. P ima y PDAC umou samples (n=122) 2. Clinical samples mic oa ay da ase s GSE71729 (UNC) (n=125) ICGC (n=93) TCGA (n=127) Classical: Classical Mo i Good Basal: Basal Mo i Bad Immune- ich: (↑) immune and in e leukin le els Good ECM- ich: (↑) ma ix ex acellula pa hways Bad Dijk e al. [37] T ansc ip ional 1. P ima y PDAC umou samples (n=90) 2. Panc ea ic cance PDXs coho (n=14) 3. PDAC Cell lines coho (n=51) Epi helial: (↑)MYC, mi ochond ial componen s and ibosome signa u e Good Mesenchymal: (↑)K-RAS, pa hways ela ed o EMT, s omal signalling and TGF-β Bad Compound panc ea ic: Simila o he mesenchymal sub ype and (↑) endoc ine pa hways Good J. Clin. Med. 2020,9, 4128 5 o 21 Table 1. Con . Au ho Type o S udy Type and Numbe o PDAC Samples Dys egula ed Pa hways and Mu a ions Ou come Chan-Seng-Yue e al. [38] Whole genome sequencing T ansc ip ional Lase cap u e mic odissec ed samples om la e-s age PDAC 1. WGS (n=314) 2. Bulk RNAseq (n=248) 3. Single-cell RNAseq (n=15) Classical A/B: (↑)SMAD4 and GATA6 al e a ions Good Basal-like A/B: (↑) EMT and TGF-β pa hways, loss o CDKN2A,TP53 mu a ions, K-RAS imbalance Bad Hyb id Mid Nicolle e al. [39] T ansc ip ional PDXs (n=76) G aded ypes be ween classical and basal based on umou di e en ia ion G ade dependan Fo each classi ica ion, ype o s udy, ype and numbe o samples, dys egula ed pa hways and mu a ions and p ognosis a e desc ibed in each column. Up- egula ed and down- egula ed pa hways a e shown as ( ↑ ) and ( ↓ ), espec i ely. CAF, Cance -Associa ed Fib oblas ; EMT, Epi helial- o-Mesenchymal T ansi ion; PDAC, Panc ea ic Duc al Adenoca cinoma; PDX, Pa ien -De i ed Xenog a ; TF, T ansc ip ion Fac o ; WGS, Whole Genome Sequencing. Mo i e al. [ 33 ] pe o med ansc ip ional analysis o classi y PDAC umou s in o wo sub ypes wi hin wo di e en issues (Table 1): umou and s oma. Based on umou cells, PDAC samples we e classi ied in o “classical” and “basal” sub ypes. The classical g oup was en iched in SMAD4 and GATA6, while he basal sub ype was molecula ly simila o basal umou s in bladde and b eas cance . Al hough classical umou s showed be e p ognosis, pa ien s wi h basal umou s esponded be e o adju an he apy. On he o he hand, he s oma was di ided in o “no mal” and “ac i a ed” sub ypes. The no mal s omal g oup showed high le els o ma ke s o panc ea ic s ella e cells, smoo h muscle ac in, imen in and desmin. Howe e , he ac i a ed s omal sub ype was de ined by a complex se o genes associa ed wi h mac ophages, umou p omo ion and ib oblas ac i a ion; his desc ibed an ac i a ed p o- umo igenic in lamma o y s omal esponse wi h poo p ognosis. Compa ing hei signa u e wi h Collisson’s [ 31 ], Mo i e al. ound an o e lap be ween bo h classical sub ypes, while genes in he quasi-mesenchymal sub ype seemed a mixed selec ion o genes om basal and s omal sub ypes (Table 2) [33]. Table 2. Molecula sub ypes co esponding o PDAC classi ica ion. Au ho s Common Sub ypes O he s Collisson e al. [31] Classical Quasi-mesenchymal Exoc ine-like Mo i e al. [33] Classical Basal-like No mal and ac i a ed s oma Bailey e al. [39] P ogeni o Squamous ADEX Immunogenic Zhao e al. [34] L1 L2 L6 L3, L4 and L5 Lombe k e al. [35] Classical Basal Mau e e al. [36] Classical Basal Immune- ich and ECM- ich Dijk e al. [37] Epi helial Mesenchymal Sec e o y Compound panc ea ic Chan-Seng-Yue e al. [ 38 ] Classical (A, B) Basal-like (A, B) Hyb id Nicolle e al. [39] F om Classical o Basal Co ela ed sub ypes a e g ouped in columns. Colou s de ine he subg oups wi h he bes (g een) o wo s ( ed) p ognosis. ECM, Ex acellula Ma ix. La e , Bailey e al. [ 10 ] de ined new molecula sub ypes o PDAC based on comp ehensi e in eg a ed genomic analysis and RNA exp ession p o iles: “squamous”, “panc ea ic p ogeni o ”, “abe an ly di e en ia ed endoc ine exoc ine” (ADEX) and “immunogenic” (Table 1). The squamous sub ype was linked o hype me hyla ion and down- egula ion o genes de e mining endode mal iden i y in panc eas, wi h poo p ognosis in pa ien s. I was also en iched in TP53 and KDM6A mu a ions and up- egula ion o he ansc ip ional ne wo k TP63 ∆ N. PDAC p ogeni o - ype umou s exp essed genes in ol ed in ea ly de elopmen o he panc eas, such as PDX1. The ADEX class o e exp essed ansc ip ional pa hways p esen in K-RAS ac i a ion and in la e s ages o panc ea ic de elopmen and di e en ia ion. Las ly, he immunogenic sub ype had panc ea ic p ogeni o ea u es, bu i was ela ed o acqui ed immune supp ession pa hways (CTLA4 and PD-1) and s ong immune in il a ion (B and T cells). When hey compa ed hei classi ica ion wi h ea lie s udies, h ee o hei J. Clin. Med. 2020,9, 4128 6 o 21 classes di ec ly o e lapped wi h Collisson’s g oups [ 31 ]: Collisson’s classical, quasi-mesenchymal and exoc ine-like sub ypes ma ched Bailey’s panc ea ic p ogeni o , squamous and ADEX g oups, espec i ely (Table 2). In addi ion, 50% o he squamous sub ype umou s we e included in Mo i ’s basal subg oup [33]. In 2018, Zhao e al. [ 34 ] conduc ed a e ospec i e me a-analysis o comple e ansc ip ome da a om pa ien s wi h PDAC (Table 1). They ob ained six di e en sub ypes g ouped in umou - (L1, L2 and L6) and s omal-speci ic (L3, L4 and L5) sub ypes. L1, L2 and L6 we e all en iched in me abolic genes, as de ailed in Sec ion 5. Besides me abolism, he sub ypes we e di e en ially en iched in o he ou es: L2 was ela ed o cell p oli e a ion and epi helium genes, such as CDKN2A; L3 had inc eased egula ion o collagen and ex acellula ma ix (ECM)-associa ed genes; L4 had an immune p o ile; L5 was in ol ed in neu oendoc ine and insulin- ela ed pa hways; L6 showed ac i i y o diges i e enzymes. Each exp ession p o ile was ela ed o clinical da a: epi helium and p oli e a ion in L2 showed he poo es p ognosis, immune and neu oendoc ine p o iles a L4 and L5 showed he bes ou comes. Conside ing he abo e s a i ica ions o PDAC, Zhao e al. ound ha L1 and L6 we e e y simila o he classical Mo i ’s sub ype, while L2 app oached hei basal sub ype (Table 2). L1, L2 and L6 we e close o he ac i a ed sub ype. In Bailey’s da ase , L1 and L4 app oxima ed he immunogenic sub ype, L2 he squamous, L3 he panc ea ic p ogeni o and L6 esembled he ADEX sub ype. Finally, L1 and L3 we e simila o Collison’s classical sub ype, L2 app oached he quasi-mesenchymal and L6 ela ed o he exoc ine-like one. Lombe k e al. [ 35 ] iden i ied epigenomic landscapes o PDAC sub ypes by pe o ming an in eg a i e analysis o genome-wide Ch oma in Immunop ecipi a ion-sequencing (ChIP-seq) on mul iple his one modi ica ions, as well as RNA-sequencing (RNA-seq) and DNA me hyla ion s udies (Table 1). They epo ed wo sub ypes: “classical” and “basal”, which co ela ed wi h clinical pa ame e s. Supe enhance mapping coupled wi h T ansc ip ion Fac o (TF) binding mo i and up- egula ion analysis showed ha classical umou s we e ela ed o TFs in ol ed in de elopmen o he panc eas, me abolic egula o s and K-RAS signalling, while basal umou s co ela ed wi h p oli e a i e and EMT TF nodes. The basal samples we e linked o a mo e agg essi e pheno ype han he classical ones. Mau e e al. [ 36 ] used Lase Cap u e Mic odissec ion (LCM) o sepa a e s omal and epi helial compa men s o PDAC bulk umou samples, and con i med using machine lea ning echniques ha s omal con amina ion may in e e e wi h PDAC classi ica ion (Table 1). In addi ion, hey desc ibed wo epi helial sub ypes o PDAC ha co ela ed wi h he classical and basal sub ypes om Mo i e al., wi h he basal sub ype ha ing a poo e p ognosis han he classical (Table 2). Mo eo e , hey iden i ied wo s omal sub ypes, an “immune- ich” g oup cha ac e ised by high immune and in e leukin le els, and an “ECM- ich” g oup, associa ed wi h ex acellula ma ix pa hways. The ECM- ich sub ype appea ed o ha e dec eased su i al compa ed wi h he immune- ich one. Finally, hey ound an associa ion o he epi helial and s omal sub ypes in which he basal epi helium sub ype and he ECM- ich s oma we e linked. In 2020, Dijk e al. [ 37 ] conduc ed an unsupe ised consensus clus e ing and iden i ied ou molecula sub ypes: “sec e o y”, “epi helial”, “compound panc ea ic” and “mesenchymal” (Table 1). The sec e o y sub ype showed en ichmen in endoc ine and exoc ine pa hways o he panc eas. Tumou s o he epi helial sub ype we e cha ac e ised by up- egula ion o he MYC oncogene and high exp ession o mi ochond ial componen s and ibosome signa u e. The mesenchymal sub ype had high le els o K-RAS ansc ip ion and was en iched in pa hways ela ed o EMT, s omal signalling and TGF- β . The compound panc ea ic sub ype p esen ed simila cha ac e is ics o he mesenchymal sub ype bu , in addi ion, was en iched in endoc ine unc ions. The sec e o y and mesenchymal sub ypes p esen ed wo se p ognosis han he epi helial and compound panc ea ic ones. Dijk e al. ound an in e connec ion be ween hese sub ypes and hose desc ibed by Collison, Bailey and Mo i (Table 2). Fo example, he sec e o y sub ype co ela ed signi ican ly wi h Collison’s exoc ine-like, Bailey’s ADEX, and Mo i ’s basal sub ypes. The epi helial sub ype sha ed cha ac e is ics wi h he Collison’s and Mo i ’s classical sub ypes, bu also wi h he Bailey ´ s panc ea ic p ogeni o and squamous J. Clin. Med. 2020,9, 4128 7 o 21 sub ypes. The compound panc ea ic g oup was simila o Collison’s exoc ine, Bailey’s ADEX and Mo i ’s classical. Las ly, he mesenchymal sub ype co ela ed wi h Collison’s quasi-mesenchymal, Bailey’s squamous and Mo i ’s basal. Chan-Seng-Yue e al. [ 38 ] pe o med whole genome sequencing and bo h bulk and single-cell RNAseq analyses on lase cap u e mic odissec ed umou s om mo e han 200 la e-s age pa ien s. They iden i ied h ee majo sub ypes: “classical”, “basal-like” and “hyb id” (Table 1). Al hough he classical and basal-like sub ypes we e ai ly o e lapped wi h he p e ious classi ica ions by Mo i , Bailey and Collison (Table 2), hey ound ha hese sub ypes could be subdi ided in o A and B subclasses wi h di e ences in hei esponse o chemo he apy, agg essi eness and disease s age. Fo example, basal-like A umou s a e highly chemo esis an and ea u e a high squamous signa u e en iched in me as a ic umou s, while basal-like B umou s co espond o a low squamous signa u e p esen in esec able umou s. Impo an ly, hey ound ha classical and basal-like p og ams co-exis wi hin a umou and demons a ed ha molecula sub ypes a e linked o a speci ic copy numbe abe a ions in genes such as mu an K-RAS (basal-like sub ype) and GATA6 (classical sub ype). Finally, Nicolle e al. [ 39 ] p oposed a molecula g adien classi ica ion o s ablish a PDAC ansc ip omic signa u e ha could be e ep esen he exis ence o in e media e cellula pheno ypes be ween classical and basal-like sub ypes (Table 1). In his s udy, hey g aded Pa ien -De i ed Xenog a s (PDXs) acco ding o di e en molecula le els o di e en ia ion and con i med ha he highe exp ession o genes linked o he classical PDAC sub ype (GATA6) was co ela ed wi h inc eased di e en ia ion o PDX samples, while lowe exp ession o genes was linked o he basal-like sub ype. They also ound ha he highe he di e en ia ion o he umou , he be e he p ognosis and esponse o chemo he apy. In e es ingly, al hough pa ial o e lap amongs he di e en signa u es and molecula sub ypes can be ound in hese s udies, ecen wo ks using single-cell ansc ip omics demons a ed ha se e al sub ypes can co-exis wi hin a umou [ 38 , 40 ], u he unde lining he high le el o he e ogenei y p esen in PDAC umou s. Undoub edly, mo e e o s in his di ec ion a e equi ed o u he dissec he complexi y o hese umou s, as well as o deciphe he in e play o he di e en subpopula ions du ing disease p og ession. 4. Me abolic Rep og amming in PDAC A he his ological le el, one o he mos no able cha ac e is ics o PDAC is i s dense s oma, being up o 90% o he umou olume. I s main ea u es a e ex ensi e ib osis, lack o ascula isa ion, hypoxia and immune in il a ion. Hypoxia is associa ed wi h inc eased cance cell p oli e a ion, su i al, EMT, in asi eness and me as asis [41]. The lack o ascula isa ion no only causes hypoxia in he umou , bu also causes me abolic s ess due o nu ien dep i a ion. As a esul , umou cells unde go he so-called “me abolic ep og amming”, an upda ed hallma k o cance [ 32 , 42 ]. Cance cells inc ease nu ien acquisi ion along wi h enhanced low h oughanabolic pa hways. Thisleads oinc easedglycolysisandglucose anspo , highglu amine consump ion, lipid and amino acid biosyn hesis and main enance o edox homeos asis. In addi ion, ecycling o cellula componen s also occu s h ough au ophagy, which deg ades mac omolecula complexes and o ganelles in o indi idual me aboli es [41,43–45]. 4.1. Wa bu g Pheno ype The majo example o me abolic ep og amming is highe glucose consump ion. The ole o glucose me abolism in cance was well de ined by O o Wa bu g back in he la e 1920s, leading o be conside ed one o he hallma ks o cance [ 46 ]. Glycoly ic low is p ecisely con olled o ul il apid p oli e a i e and syn he ic needs. Unlike no mal cells, umou cells ha e high le els o glycolysis, e en in he p esence o oxygen and educed mi ochond ial unc ion, leading o a s a e called “ae obic glycolysis”, also called he “Wa bu g e ec ”. On he o he hand, he “ e e se Wa bu g e ec ” desc ibes a wo-compa men model in which cance cells induce ae obic glycolysis in he s omal cells, whose glycolysis end-p oduc s a e ans e ed o he cance cells o eed mi ochond ial oxida i e J. Clin. Med. 2020,9, 4128 8 o 21 phospho yla ion (OXPHOS). This allows umou s o espond o a ia ions in nu ien a ailabili y and o op imise cell p oli e a ion and g ow h [ 47 ]. In e es ingly, his wo-compa men model can also be applied conside ing he unc ional he e ogenei y o cance cells in PDAC: glycoly ic di e en ia ed umou cells could p o ide subs a es o oxida i e cance s em cells (CSCs) [48]. A hypo hesis o me abolic ewi ing owa ds enhanced glycolysis o e he educ ion o mi ochond ial oxida ion as a sou ce o ATP in PDAC is he p esence o a dense desmoplas ic s oma, which basically impedes neo ascula isa ion. This c ea es a hypoxic mic oen i onmen in which oxygen and nu ien s a e limi ed [ 49 ]. This s a e c ea es a posi i e eedback loop by which, on he one hand, panc ea ic cance cells eel a selec i e p essu e unde he hypoxic and nu ien sho age whe e only he mos agg essi e popula ions will emain. On he o he hand, unde such s ess-d i en si ua ions, panc ea ic cance cells a e o ced o modi y hei me abolism in o de o cope wi h hei bioene ge ic demands o PDAC p og ession, expansion and dissemina ion h ough he blood essels owa ds less sca ce en i onmen s [ 50 ]. Indeed, a ecen s udy demons a ed ha glycolysis and hypoxia signa u es co ela e in PDAC and ha P olyl 4-Hyd oxylase subuni Alpha 1 (P4HA1), a c i ical enzyme in ol ed in collagen syn hesis, con ols glycolysis h ough HIF1 α s abilisa ion [ 51 ]. Mo eo e , gene exp ession a ays o me as a ic PDAC e ealed a glycolysis-based signa u e cha ac e ised by inc eased exp ession o many glycoly ic enzymes [52], highligh ing he impo ance o glycoly ic me abolism in PDAC p og ession. In ac , Liu e al. ecen ly demons a ed ha he EMT- ela ed gene SNAIL was able o induce a mig a o y pheno ype in PDAC cell lines by p omo ing mesenchymal- ela ed genes exp ession along wi h enhanced glucose up ake and lac a e p oduc ion [53]. O e all, he glycoly ic s a e is cha ac e ised by an inc eased exp ession o glycoly ic enzymes and glucose and lac a e anspo e s, such as Glucose T anspo e 1 (GLUT1), and Monoca boxyla e T anspo e s 1 and 4 (MCT1, MCT4) [ 54 , 55 ]. Speci ically, he o e exp ession o hese memb ane anspo e s leads o an enhanced glucose sca enging om he hypo ascula ised umou mic oen i onmen , which esul s in inc eased glucose a ailabili y in he cance cell as well as a be e balance o he glucose pa hway in o de o keep glycolysis a high a e. On he one hand, GLUT1 is an ATP-independen glucose anspo e ha enables glucose ans e ence om a high-g adien ex acellula compa men o low-g adien cy oplasmic compa men . I s exp ession dosage has been epo ed o be associa ed wi h PDAC p og ession om low- o high-g ade panc ea ic p eneoplas ic lesionswhencompa ed ono mal panc eas[ 54 ]. On heo he hand, MCT1 andMCT4a e p o on-coupled sympo anspo e s wi h highe a ini y o lac a e e lux. These anspo e s implica ed in glucose homeos asis a e equi ed o cance cells o neu alise in acellula acidi ica ion due o he inc eased glycoly ic a e and ha e been epo ed o be o e exp essed in PDAC. Kong e al. showed ha inhibi ion o knockdown o MCT esul ed in an inhibi ed lac a e lux. In e es ingly, hey also demons a ed ha hese anspo e s a e implica ed in PDAC cell lines in asi eness, hus highligh ing once again he implica ions o glucose homeos asis in panc ea ic cance [56]. Impo an ly, he up- egula ion o mos o hese genes is media ed by PDAC d i e mu a ions on K-RAS and TP53. As e iewed by B yan e al. [ 57 ], oncogenic K-RAS enhances he exp ession o many glycoly ic enzymes such as GLUT1, Hexokinase 1 and 2 (HK1, HK2) and Lac a e Dehyd ogenase A (LDHA), hus inc easing glycoly ic lux. This me abolic ep og amming owa ds glycolysis con ibu es o an enhanced su i al o glycoly ic PDAC cell lines in he p esence o low le els o glucose. Mo eo e , ano he s udy based on ansc ip ome and me abolome analyses showed ha mu an K-RAS in ad anced PDAC mouse models is necessa y o an enhanced glucose up ake [ 57 ]. This s udy also e ealed ha abe an K-RAS is implica ed in glucose me abolism in e media es channelling in o di e en anabolic pa hways, such as he hexosamine biosyn hesis pa hway (HBP) and pen ose phospha e pa hway (PPP), hus p o ing ha glucose me abolism is necessa y o uel anabolic b anches o PDAC me abolism o p o ide he cance cells wi h building blocks o i s inc eased p oli e a ion demands [ 58 ]. Mo eo e , he enhanced exp ession o he glycoly ic enzymes was demons a ed o be ela ed o bad PDAC p ognosis, in asi eness and me as ases onse [59]. J. Clin. Med. 2020,9, 4128 9 o 21 4.2. Lipid Me abolism in PDAC PDAC umou s a e also highly dependen on lipid me abolism [ 60 , 61 ] and, in ac , a high a die was shown o suppo umou g ow h in mu ine models [ 60 ]. On he one hand, a y acids (FA) can be p o ided exogenously by he abso p ion o ex acellula lipids ( om die , li e syn hesis o adipose issue). Fo example, cance -associa ed adipocy es can p o ide adipokines and lipids o cance cells [ 62 ], inc easing panc ea ic cance cell agg essi eness [ 63 ]. The exogenous FA up ake equi es he p esence o he anspo e CD36 and FA-binding p o eins (FABPs). CD36 can also in luence gemci abine esis ance in PDAC, by egula ing an i-apop osis p o eins [ 64 , 65 ]. Unsu p isingly, PDAC pa ien s wi h high CD36 exp ession ha e lowe o e all su i al and ecu ence- ee su i al a es han pa ien s wi h low exp ession. In his con ex , CD36 could be conside ed as an un a ou able p ognosis ac o and he use o an i-CD36 s a egies in associa ion wi h con en ional chemo he apies could ep esen a p omising he apeu ic app oach [64,66]. On he o he hand, panc ea ic cance cells can syn hesise de no o lipids h ough he lipogenesis p ocess, p oducing mo e han 90% o he iacylglyce ol-FA. This p ocess uses he mi ochond ial ci a e p oduced om he ica boxylic acid (TCA) cycle uelled by glucose and/o glu amine as ca bon sou ces [ 67 ]. Lipogenic enzymes a e o en o e exp essed in PDAC cells; o example Fa y Acid Syn hase (FASN) is pa icula ly p ominen and associa ed wi h poo p ognosis [ 61 , 68 – 70 ]. The pha macological inhibi ion o his enzyme educes s emness ea u es and gemci abine esis ance in panc ea ic cance cells [ 71 ]. The p oduced iacylglyce ol molecules a e s o ed in lipid d ople s and a co ela ion was es ablished be ween he accumula ion o lipid d ople s and umou p og ession and agg essi eness [72]. Likewise, an ele a ed lipid syn hesis co ela es wi h CSC p ope ies and su i al in di e en ypes o cance [ 72 ]. Thus, cance cells acqui e FA h ough ei he lipid up ake o de no o lipogenesis, and ac i a e in acellula lipolysis o mobilise he FA s ocks. FA sus ain h ee equi emen s o PDAC de elopmen and cance cells in gene al: cell memb ane o ma ion, biosyn hesis o signalling molecules and lipid-de i ed messenge s, and ene gy p oduc ion. Fi s o all, lipid syn hesis is an impo an equi emen o highly p oli e a i e cance cells o sus ain memb ane o ma ion [ 72 ]. F om de no o syn hesis, sa u a ed o monounsa u a ed FA modula e memb ane luidi y and o m mo e dense memb ane laye s ha may educe he up ake o d ugs and con ibu e o he apy- esis ance [ 73 ]. Mo eo e , lipids a e implica ed in signal ansduc ion in wo di e en ways: by building lipid a s modula ing p o ein ec ui men s and in e ac ions, as well as by o ma ion o lipidic signalling molecules. This is he case o phospha idylinosi ol-3,4,5- isphospha e [PI(3,4,5)P3], able o ac i a e he p o ein kinase B/AKT and s imula e cell p oli e a ion and su i al [ 73 ]. Finally, FA ep esen an impo an sou ce o ene gy in non-glycoly ic umou s, using he mi ochond ial β -oxida ion ( a y acid oxida ion, FAO) o p oduce ATP [ 72 , 73 ]. In ha con ex , Luo e al. demons a ed ha he use o e omoxi , an inhibi o o FAO by blocking he en ance o FA in he mi ochond ia ia Ca ni ine Palmi oyl T ans e ase 1A (CPT1A), can es o e he sensi i i y o panc ea ic CSCs o gemci abine by inducing an ene gy c isis in hose cells [ 74 ]. These da a sugges ha CPT1A is an impo an ac o o he cance me abolism ep og amming and could ep esen an a ac i e he apeu ic a ge , and highligh how his p ocess is impo an o he cance cell o supply ATP unde ene gy s ess. Addi ionally, PDAC cells a e also highly dependen on choles e ol, as i con ibu es, o example, o he o ma ion o lipid a s, hus modula ing he ec ui men o key oncogenes ecep o such as he Epide mal G ow h Fac o (EGF)- ecep o and egula ing su i al pa hways. Cance cells can inc ease hei con en h ough ei he syn hesis ( h ough me alona e pa hway) o endocy osis media ed by low-densi y lipop o eins (LDL) and LDL Recep o (LDLR). In PDAC pa ien s, choles e ol biosyn hesis is associa ed wi h a mo e di e en ia ed pheno ype (classical sub ype), while high LDLR exp ession co ela es wi h a highe isk o umou ecu ence. In e es ingly, inhibi ion o choles e ol syn hesis induced a mesenchymal pheno ype [ 75 ] while blocking choles e ol up ake ia he knock-down o LDLR was able o sensi ise PDAC cells o chemo he apy [76,77]. Taken oge he , all hese epo s p o e he po en ial o lipid me abolism a ge ing in o de o sensi ise PDAC cells o chemo he apy. Howe e , some an i umou igenic e ec s o speci ic FA like J. Clin. Med. 2020,9, 4128 16 o 21 7. Conclusions PDAC is a he e ogeneous disease ha can be classi ied, acco ding o i s me abolic needs, in o glycoly ic and lipogenic sub ypes wi h di e en p ognosis. Indeed, he glycoly ic sub ype would be mo e agg essi e and esis an o con en ional chemo he apy han he lipogenic one. This me abolic s a i ica ion co ela es wi h p e ious molecula classi ica ion sys ems, as hey show sub ypes sha ing simila cha ac e is ics and p ognosis. In his con ex , classi ica ion in o me abolic sub ypes may be be e placed o clinical use, since i could p o ide in o ma ion on he unc ional pheno ype o he umou , co ela ing o agg essi eness, chemo esis ance and me as a ic abili ies independen ly om i s mu a ional s a e. The in o ma ion collec ed in his e iew s ongly sugges s ha pa ien s a i ica ion based on me abolic ea u es may bea p ognos ic alue and guide he apeu ic decisions in he u u e, iden i ying a subg oup o pa ien s wi h poo p ognosis ha may be eligible o pe sonalised ea men s designed acco ding o me abolic ulne abili ies. Au ho Con ibu ions: Concep ualisa ion, P.E.-R. and P.S.; da a analysis and in es iga ion P.E.-R.; w i ing—o iginal d a p epa a ion, P.E.-R., S.C. and B.P.-A.; w i ing— e iew and edi ing and supe ision and unding acquisi ion, P.S. All au ho s ha e ead and ag eed o he published e sion o he manusc ip . Funding: This esea ch was unded by he Ins i u o de Salud Ca los III h ough he Miguel Se e P og am (CP16/00121) and Fondo de In es igaciones Sani a ias (PI17/00082) (bo h co- inanced by Eu opean unds (FSE: “El FSE in ie e en u u u o” and FEDER: “Una mane a de hace Eu opa,” espec i ely) and he Wo ldwide Cance Resea ch Cha i y oge he wi h Fundaci ó n Cien í ica Asociaci ó n Española con a el C á nce (FCAECC) (19-0250). Acknowledgmen s: We wan o hank Alba Royo and Lau a Sancho o p oo eading he manusc ip . Con lic s o In e es : The au ho s decla e no con lic o in e es . Re e ences 1. Fouche , E.D.; Ghigo, C.; Chouaib, S.; Galon, J.; Io anna, J.; Oli e, D. Panc ea ic Duc al Adenoca cinoma: A S ong Imbalance o Good and Bad Immunological Cops in he Tumo Mic oen i onmen . F on . Immunol. 2018,9, 1044. [C ossRe ] 2. Oppong, K.W.; Bekkali, N.L.H. Panc ea ic duc al adenoca cinoma epidemiology and isk assessmen : Could we p e en ? Possibili y o an ea ly diagnosis. Endosc. Ul asound 2017,6, 58–S61. [C ossRe ] [PubMed] 3. Hu , C.; T amon ano, A.C.; Dowling, E.C.; B ooks, G.A.; Jeon, A.; B ugge, W.R.; Gazelle, G.S.; Kong, C.Y.; Pandha ipande, P.V. Ea ly Panc ea ic Duc al Adenoca cinoma Su i al Is Dependen on Size. Panc eas 2016 , 45, 1062–1066. [C ossRe ] [PubMed] 4. Kakka , A.; Choudhu i, J.; Mukhe jee, I. The sub ypes o panc ea ic duc al adenoca cinomas. SANAMED 2016,11, 239. [C ossRe ] 5. Le, N.; Sund, M.; Vinci, A.; Beye , G.; Ja ed, M.A.; K ug, S.; Neessee, A.; Schobe , M.; In o ma ion, P.E.K.F.C. P ognos ic and p edic i e ma ke s in panc ea ic adenoca cinoma. Dig. Li e Dis. 2016 ,48, 223–230. [C ossRe ] 6. Zei ouni, D.; Pylaye a-Gup a, Y.; De , C.J.; B yan , K.L. KRAS Mu an Panc ea ic Cance : No Lone Pa h o an E ec i e T ea men . Cance s 2016,8, 45. [C ossRe ] 7. O h, M.; Me zge , P.; Ge um, S.; Maye le, J.; Schneide , G.; Belka, C.; Schnu , M.; Laube , K. Panc ea ic duc al adenoca cinoma: Biological hallma ks, cu en s a us, and u u e pe spec i es o combined modali y ea men app oaches. Radia . Oncol. 2019,14, 1–20. [C ossRe ] 8. Adamska, A.; Domenichini, A.; Falasca, M. Panc ea ic Duc al Adenoca cinoma: Cu en and E ol ing The apies. In . J. Mol. Sci. 2017,18, 1338. [C ossRe ] 9. Pu, N.; Chen, Q.; Gao, S.; Liu, G.; Zhu, Y.; Yin, L.; Hu, H.; Wei, L.; Wu, Y.; Maeda, S.; e al. Gene ic landscape o p ognos ic alue in panc ea ic duc al adenoca cinoma mic oen i onmen . Ann. T ansl. Med. 2019 ,7, 645. [C ossRe ] 10. Bailey, P.; Ini ia i e, A.P.C.G.; Chang, D.K.; Nones, K.; Johns, A.L.; Pa ch, A.-M.; Ging as, M.-C.; Mille , D.K.; Ch is , A.N.; B uxne , T.J.C.; e al. Genomic analyses iden i y molecula sub ypes o panc ea ic cance . Na u e 2016,531, 47–52. [C ossRe ] 11. Pihlak, R.; Wea e , J.M.J.; Valle, J.W.; McNama a, M.G. Ad ances in Molecula P o iling and Ca ego isa ion o Panc ea ic Adenoca cinoma and he Implica ions o The apy. Cance s 2018 ,10, 17. [C ossRe ] [PubMed] J. Clin. Med. 2020,9, 4128 17 o 21 12. Waddell, N.; Pajic, M.; Pa ch, A.-M.; Chang, D.K.; Kassahn, K.S.; Bailey, P.; Johns, A.L.; Mille , D.; Nones, K.; Quek, K.; e al. Whole genomes ede ine he mu a ional landscape o panc ea ic cance . Na u e 2015 , 518, 495–501. [C ossRe ] [PubMed] 13. Bou ne , B.; Musca i, F.; Buscail, C.; Assena , E.; Ba he , M.; Hammel, P.; Sel es, J.; Guimbaud, R.; Co delie , P.; Buscail, L. KRAS G12D Mu a ion Sub ype Is A P ognos ic Fac o o Ad anced Panc ea ic Adenoca cinoma. Clin. T ansl. Gas oen e ol. 2016,7, e157. [C ossRe ] 14. Sahin, I.H.; A Iacobuzio-Donahue, C.; O’Reilly, E.M. Molecula signa u e o panc ea ic adenoca cinoma: An insigh om geno ype o pheno ype and challenges o a ge ed he apy. Expe Opin. The . Ta ge s 2015 , 20, 341–359. [C ossRe ] 15. Robe s, N.J.; No is, A.L.; Pe e sen, G.M.; Bondy, M.L.; E B and, R.; Gallinge , S.; Robe s, N.J.; Olson, S.H.; Rus gi, A.K.; Schwa z, A.G.; e al. Whole Genome Sequencing De ines he Gene ic He e ogenei y o Familial Panc ea ic Cance . Cance Disco . 2016,6, 166–175. [C ossRe ] 16. No a, F.; Chan-Seng-Yue, M.; Lemi e, M.; Li, Y.; Wilson, G.W.; Conno , A.A.; Den oche, R.E.; Liang, S.-B.; B own, A.M.K.; Kim, J.C.; e al. A enewed model o panc ea ic cance e olu ion based on genomic ea angemen pa e ns. Na u e 2016,538, 378–382. [C ossRe ] 17. Raphael, B.; H uban, R.H.; Agui e, A.J.; Mo i , R.A.; Yeh, J.J.; S ewa , C.; Robe son, A.G.; Che niack, A.D.; Gup a, M.; Ge z, G.; e al. In eg a ed Genomic Cha ac e iza ion o Panc ea ic Duc al Adenoca cinoma. Cance Cell 2017,32, 185–203.e13. [C ossRe ] 18. Rice, A.; He n á ndez, A.E.D.R. The Mu a ional Landscape o Panc ea ic and Li e Cance s, as Rep esen ed by Ci cula ing Tumo DNA. F on . Oncol. 2019,9. [C ossRe ] 19. Wa e s, A.M.; De , C.J. KRAS: The C i ical D i e and The apeu ic Ta ge o Panc ea ic Cance . Cold Sp ing Ha b. Pe spec . Med. 2018,8, a031435. [C ossRe ] 20. I , J.P.M.; Wang, S.C.; Heb ok, M. KRAS, Hedgehog, Wn and he wis ed de elopmen al biology o panc ea ic duc al adenoca cinoma. Na . Re . Cance 2010,10, 683–695. [C ossRe ] 21. Kanda, M.; Ma haei, H.; Wu, J.; Hong, S.; Yu, J.; Bo ges, M.; H uban, R.H.; Mai a, A.; Kinzle , K.; Vogels ein, B.; e al. P esence o Soma ic Mu a ions in Mos Ea ly-S age Panc ea ic In aepi helial Neoplasia. Gas oen e ol. 2012,142, 730–733.e9. [C ossRe ] [PubMed] 22. Mo ison, D.K. MAP Kinase Pa hways. Cold Sp ing Ha b. Pe spec . Biol. 2012 ,4, a011254. [C ossRe ] [PubMed] 23. Ese , S.; Schnieke, A.; Schneide , G.; Sau , D. Oncogenic KRAS signalling in panc ea ic cance . B . J. Cance 2014,111, 817–822. [C ossRe ] [PubMed] 24. Muelle , S.; Englei ne , T.; Ma esch, R.; Zukowska, M.; Lange, S.; Kal enbache , T.; Konukiewi z, B.; Öllinge , R.; Zwiebel, M.; S ong, A.; e al. E olu iona y ou es and KRAS dosage de ine panc ea ic cance pheno ypes. Na . Cell Biol. 2018,554, 62–68. [C ossRe ] 25. G an , T.J.; Hua, K.; Singh, A. Molecula Pa hogenesis o Panc ea ic Cance . In P og ess in Molecula Biology and T ansla ional Science; Else ie BV: Ams e dam, The Ne he lands, 2016; Volume 144, pp. 241–275. 26. Mo on, J.P.; Timpson, P.; Ka im, S.A.; Ridgway, R.A.; A hineos, D.; Doyle, B.; Jamieson, N.B.; Oien, K.A.; Lowy, A.M.; B un on, V.G.; e al. Mu an p53 d i es me as asis and o e comes g ow h a es /senescence in panc ea ic cance . P oc. Na l. Acad. Sci. USA 2010,107, 246–251. [C ossRe ] 27. Weissmuelle , S.; Manchado, E.; Sabo owski, M.; Mo is, J.P.; Wagenblas , E.; Da is, C.A.; Moon, S.-H.; P is e , N.T.; Tschaha ganeh, D.F.; Ki zing, T.; e al. Mu an p53 D i es Panc ea ic Cance Me as asis h ough Cell-Au onomous PDGF Recep o βSignaling. Cell 2014,157, 382–394. [C ossRe ] 28. Zhao, M.; Mish a, L.; Deng, C.X. The ole o TGF- β /SMAD4 signaling in cance . In . J. Biol. Sci. 2018 ,14, 111–123. [C ossRe ] 29. Hao, Y.; Bake , D.; Dijke, P.T. TGF- β -Media ed Epi helial-Mesenchymal T ansi ion and Cance Me as asis. In . J. Mol. Sci. 2019,20, 2767. [C ossRe ] 30. Da id, C.J.; Huang, Y.-H.; Chen, M.; Su, J.; Zou, Y.; Ba deesy, N.; Iacobuzio-Donahue, C.A.; Massagu é , J. TGF-βTumo Supp ession h ough a Le hal EMT. Cell 2016,164, 1015–1030. [C ossRe ] 31. Collisson, E.A.; Sadanandam, A.; Olson, P.; Gibb, W.J.; T ui , M.; Gu, S.; Cooc, J.; Weinkle, J.; Kim, G.E.; Jakkula, L.; e al. Sub ypes o panc ea ic duc al adenoca cinoma and hei di e ing esponses o he apy. Na . Med. 2011,17, 500–503. [C ossRe ] 32. Veens a, V.L.; Ga cia-Ga ijo, A.; Van Laa ho en, H.W.M.; Bijlsma, M.F. Ex acellula In luences: Molecula Subclasses and he Mic oen i onmen in Panc ea ic Cance . Cance s 2018,10, 34. [C ossRe ] [PubMed] J. Clin. Med. 2020,9, 4128 18 o 21 33. Mo i , R.A.; Ma aya i, R.; Fla e, E.L.; Volma , K.E.; Loeza, S.G.H.; Hoadley, K.A.; Rashid, N.U.; Williams, L.A.; Ea on, S.C.; Chung, A.H.; e al. Vi ual mic odissec ion iden i ies dis inc umo - and s oma-speci ic sub ypes o panc ea ic duc al adenoca cinoma. Na . Gene . 2015,47, 1168–1178. [C ossRe ] [PubMed] 34. Zhao, L.; Zhao, H.; Yan, H. Gene exp ession p o iling o 1200 panc ea ic duc al adenoca cinoma e eals no el sub ypes. BMC Cance 2018,18, 1–13. [C ossRe ] [PubMed] 35. Lombe k, G.; Blum, Y.; Nicolle, R.; Nai , A.; Gaonka , K.S.; Ma isa, L.; Ma hison, A.; Sun, Z.; Yan, H.; Ela ouci, N.; e al. Dis inc epigene ic landscapes unde lie he pa hobiology o panc ea ic cance sub ypes. Na . Commun. 2018,9, 1–10. [C ossRe ] [PubMed] 36. Mau e , H.C.; Holms om, S.R.; He, J.; Laise, P.; Su, T.; Ahmed, A.; Hibshoosh, H.; A Chabo , J.; E Obe s ein, P.; Sepul eda, A.R.; e al. Expe imen al mic odissec ion enables unc ional ha monisa ion o panc ea ic cance sub ypes. Gu 2019,68, 1034–1043. [C ossRe ] [PubMed] 37. Dijk, F.; Veens a, V.L.; Soe , E.C.; Dings, M.P.; Zhao, L.; Hal we k, J.B.; Hooije , G.K.; Damho e , H.; Ma zano, M.; S eins, A.; e al. Unsupe ised class disco e y in panc ea ic duc al adenoca cinoma e eals cell-in insic mesenchymal ea u es and high conco dance be ween exis ing classi ica ion sys ems. Sci. Rep. 2020,10, 337. [C ossRe ] [PubMed] 38. Chan-Seng-Yue, M.; Kim, J.C.; Wilson, G.W.; Ng, K.; Figue oa, E.F.; O’Kane, G.M.; Conno , A.A.; Den oche, R.E.; G an , R.C.; McLeod, J.; e al. T ansc ip ion pheno ypes o panc ea ic cance a e d i en by genomic e en s du ing umo e olu ion. Na . Gene . 2020,52, 231–240. [C ossRe ] 39. Nicolle, R.; Blum, Y.; Duconseil, P.; Vanb ugghe, C.; B andone, N.; Poiza , F.; Roques, J.; Bigonne , M.; Gaye , O.; Rubis, M.; e al. Es ablishmen o a panc ea ic adenoca cinoma molecula g adien (PAMG) ha p edic s he clinical ou come o panc ea ic cance . EBioMedicine 2020,57, 102858. [C ossRe ] 40. Juiz, N.; Elkaou a i, A.; Bigonne , M.; Gaye , O.; Roques, J.; Nicolle, R.; Io anna, J.; Duse i, N. Basal-like and classical cells coexis in panc ea ic cance e ealed by single-cell analysis on biopsy-de i ed panc ea ic cance o ganoids om he classical sub ype. FASEB J. 2020,34, 12214–12228. [C ossRe ] 41. Kampho s , J.J.; C oss, J.R.; Fan, J.; De S anchina, E.; Ma hew, R.; Whi e, E.P.; Thompson, C.B.; Rabinowi z, J.D. Hypoxic and Ras- ans o med cells suppo g ow h by sca enging unsa u a ed a y acids om lysophospholipids. P oc. Na l. Acad. Sci. USA 2013,110, 8882–8887. [C ossRe ] 42. Hanahan, D.; Weinbe g, R.A. Hallma ks o cance : The nex gene a ion. Cell 2011 ,144, 646–674. [C ossRe ] [PubMed] 43. Ying, H.; Dey, P.; Yao, W.; Kimmelman, A.C.; D ae a, G.F.; Mai a, A.; Depinho, R.A. Gene ics and biology o panc ea ic duc al adenoca cinoma. Genes De . 2016,30, 355–385. [C ossRe ] [PubMed] 44. Chen, J.-Q.; Russo, J. Dys egula ion o glucose anspo , glycolysis, TCA cycle and glu aminolysis by oncogenes and umo supp esso s in cance cells. Biochim. e Biophys. Ac a (BBA) Bioene g. 2012 ,1826, 370–384. [C ossRe ] [PubMed] 45. Galluzzi, L.; Kepp, O.; Heiden, M.G.V.; K oeme , G. Me abolic a ge s o cance he apy. Na . Re . D ug Disco . 2013,12, 829–846. [C ossRe ] 46. Wa bu g, O.; Wind, F.; Negelein, E. THE METABOLISM OF TUMORS IN THE BODY. J. Gen. Physiol. 2004 , 8, 519–530. [C ossRe ] 47. Wilde, L.; Roche, M.; Domingo-Vidal, M.; Tanson, K.; Philp, N.; Cu y, J.; Ma inez-Ou schoo n, U.E. Me abolic coupling and he Re e se Wa bu g E ec in cance : Implica ions o no el bioma ke and an icance agen de elopmen . Semin. Oncol. 2017,44, 198–203. [C ossRe ] 48. Sancho, P.; Bu gos-Ramos, E.; Ta e a, A.; Khei , T.B.; Jagus , P.; Schoenhals, M.; Ba neda, D.; Selle s, K.; Campos-Oli as, R.; G aña, O.; e al. MYC/PGC-1 α Balance De e mines he Me abolic Pheno ype and Plas ici y o Panc ea ic Cance S em Cells. Cell Me ab. 2015,22, 590–605. [C ossRe ] 49. Koong, A.; Meh a, V.K.; Le, Q.T.; Fishe , G.A.; Te is, D.J.; B own, J.; Bas idas, A.J.; Vie a, M. Panc ea ic umo s show high le els o hypoxia. In . J. Radia . Oncol. 2000,48, 919–922. [C ossRe ] 50. Spi ak-K oizman,T.R.; Hos e e , G.; Posne , R.; Aziz,M.; Hu, C.; Demeu e, M.J.; Von Ho , D.; Hingo ani, S.R.; Palculic , T.B.; Izzo, J.; e al. Hypoxia T igge s Hedgehog-Media ed Tumo -S omal In e ac ions in Panc ea ic Cance . Cance Res. 2013,73, 3235–3247. [C ossRe ] 51. Cao, X.; Cao, Y.; Li, W.; Zhang, H.; Zhu, Z. P4HA1/HIF1 α eedback loop d i es he glycoly ic and malignan pheno ypes o panc ea ic cance . Biochem. Biophys. Res. Commun. 2019,516, 606–612. [C ossRe ] J. Clin. Med. 2020,9, 4128 19 o 21 52. Chaika, N.V.; Yu, F.; Pu ohi , V.; Mehla, K.; Lazenby, A.J.; DiMaio, D.; Ande son, J.M.; Yeh, J.J.; Johnson, K.R.; Hollingswo h, M.A.; e al. Di e en ial Exp ession o Me abolic Genes in Tumo and S omal Componen s o P ima y and Me as a ic Loci in Panc ea ic Adenoca cinoma. PLoS ONE 2012 ,7, e32996. [C ossRe ] [PubMed] 53. Liu,M.; Hancock, S.E.; Sul ani, G.; Wilkins,B.P.; Ding,E.; Osbo ne,B.; Quek,L.-E.; Tu ne ,N.Snail-O e exp ession Induces Epi helial-mesenchymal T ansi ion and Me abolic Rep og amming in Human Panc ea ic Duc al Adenoca cinoma and Non- umo igenic Duc al Cells. J. Clin. Med. 2019,8, 822. [C ossRe ] [PubMed] 54. Came on, M.E.; Yako enko, A.; T e ino, J.G. Glucose and Lac a e T anspo in Panc ea ic Cance : Glycoly ic Me abolism Re isi ed. J. Oncol. 2018,2018, 1–6. [C ossRe ] [PubMed] 55. Bas u k, O.; Singh, R.; Kaygusuz, E.; Balci, S.; Du sun, N.; Culhaci, N.; Adsay, N.V. GLUT-1 Exp ession in Panc ea ic Neoplasia. Panc eas 2011,40, 187–192. [C ossRe ] [PubMed] 56. Kong, S.C.; Nøh -Nielsen, A.; Zeebe g, K.; Reshkin, S.J.; Ho mann, E.K.; No ak, I.; Pede sen, S.F. Monoca boxyla e T anspo e s MCT1 and MCT4 Regula e Mig a ion and In asion o Panc ea ic Duc al Adenoca cinoma Cells. Panc eas 2016,45, 1036–1047. [C ossRe ] 57. B yan , K.L.; Mancias, J.D.; Kimmelman, A.C.; De , C.J. KRAS: Feeding panc ea ic cance p oli e a ion. T ends Biochem. Sci. 2014,39, 91–100. [C ossRe ] 58. Ying, H.; Kimmelman, A.C.; Lyssio is, C.A.; Hua, S.; Chu, G.C.; Fle che -Sananikone, E.; Locasale, J.W.; Son, J.; Zhang, H.; Colo , J.L.; e al. Oncogenic K as Main ains Panc ea ic Tumo s h ough Regula ion o Anabolic Glucose Me abolism. Cell 2012,149, 656–670. [C ossRe ] 59. Tian, G.; Li, G.; Liu, P.; Wang, Z.; Li, N. Glycolysis-Based Genes Associa ed wi h he Clinical Ou come o Panc ea ic Duc al Adenoca cinoma Iden i ied by The Cance Genome A las Da a Analysis. DNA Cell Biol. 2020,39, 417–427. [C ossRe ] 60. Liang, C.; Qin, Y.; Zhang, B.; Ji, S.; Shi, S.; Xu, W.; Liu, J.; Xiang, J.; Liang, D.; Hu, Q.; e al. Ene gy sou ces iden i y me abolic pheno ypes in panc ea ic cance . Ac a Biochim. e Biophys. Sin. 2016 ,48, 969–979. [C ossRe ] 61. Swie czynski, J.; Hebanowska, A.; Sledzinski, T. Role o abno mal lipid me abolism in de elopmen , p og ession, diagnosis and he apy o panc ea ic cance . Wo ld J. Gas oen e ol. 2014 ,20, 2279–2303. [C ossRe ] 62. Lengyel, E.; Makowski, L.; DiGio anni, J.; Kolonin, M.G. Cance as a Ma e o Fa : The C oss alk be ween Adipose Tissue and Tumo s. T ends Cance 2018,4, 374–384. [C ossRe ] [PubMed] 63. Cai, Z.; Liang, Y.; Xing, C.; Wang, H.; Hu, P.; Li, J.; Huang, H.; Wang, W.; Jiang, C. Cance -associa ed adipocy es exhibi dis inc pheno ypes and acili a e umo p og ession in panc ea ic cance . Oncol. Rep. 2019,42, 2537–2549. [C ossRe ] [PubMed] 64. Kubo, M.; Go oh, K.; Eguchi, H.; Kobayashi, S.; Iwagami, Y.; Tomima u, Y.; Aki a, H.; Asaoka, T.; Noda, T.; Takeda, Y.; e al. Impac o CD36onChemo esis ance inPanc ea ic Duc alAdenoca cinoma. Ann. Su g. Oncol. 2019,27, 610–619. [C ossRe ] [PubMed] 65. Kubo, M.; Eguchi, H. ASO Au ho Re lec ions: Regula ion o Chemo esis ance in Panc ea ic Duc al Adenoca cinoma by Sca enge Recep o CD36. Ann. Su g. Oncol. 2019,27, 620–621. [C ossRe ] [PubMed] 66. Tanase, C.; Gheo ghisan-Gala eanu, A.A.; Popescu, I.D.; Mihai, S.; Cod ici, E.; Albulescu, R.; Hinescu, M.E. CD36 and CD97 in Panc ea ic Cance e sus O he Malignancies. In . J. Mol. Sci. 2020 ,21, 5656. [C ossRe ] 67. Qin, C.; Yang, G.; Yang, J.; Ren, B.; Wang, H.; Chen, G.; Zhao, F.; You, L.; Wang, W.; Zhao, Y. Me abolism o panc ea ic cance : Pa ing he way o be e an icance s a egies. Mol. Cance 2020,19, 1–19. [C ossRe ] 68. Sunami, Y.; Rebelo, A.; Klee , J. Lipid Me abolism and Lipid D ople s in Panc ea ic Cance and S ella e Cells. Cance s 2017,10, 3. [C ossRe ] 69. Cohen, R.; Neuzille , C.; Tije as-Raballand, A.; Fai e, S.; De G amon , A.; Raymond, E. Ta ge ing cance cell me abolism in panc ea ic adenoca cinoma. Onco a ge 2015,6, 16832–16847. [C ossRe ] 70. Bian, Y.; Yu, Y.; Wang, S.; Li, L. Up- egula ion o a y acid syn hase induced by EGFR/ERK ac i a ion p omo es umo g ow h in panc ea ic cance . Biochem. Biophys. Res. Commun. 2015 ,463, 612–617. [C ossRe ] 71. Tad os, S.B.; Shukla, S.K.; King, R.J.; Gunda, V.; Ve nucci, E.; Ab ego, J.; Chaika, N.V.; Lyudmyla, B.; Lazenby, A.J.; Be im, L.; e al. De No oLipid Syn hesis Facili a es Gemci abine Resis ance h ough Endoplasmic Re iculum S ess in Panc ea ic Cance . Cance Res. 2017,77, 5503–5517. [C ossRe ] 72. Jagus , P.; De Lux á n-Delgado, B.; Pa ejo-Alonso, B.; Sancho, P. Me abolism-Based The apeu ic S a egies Ta ge ing Cance S em Cells. F on . Pha macol. 2019,10, 203. [C ossRe ] [PubMed] J. Clin. Med. 2020,9, 4128 20 o 21 73. Zaidi, N.; Lupien, L.; Kuemme le, N.B.; Kinlaw, W.B.; Swinnen, J.V.; Smans, K. Lipogenesis and lipolysis: The pa hways exploi ed by he cance cells o acqui e a y acids. P og. Lipid Res. 2013 ,52, 585–589. [C ossRe ] [PubMed] 74. Luo,J.; Hong,Y.; Tao,X.; Wei,X.; Zhang,L.; Li,Q.Anindispensable oleo CPT-1a osu i ecance cellsdu ing ene gy s ess h ough ewi ing cance me abolism. Tumo Biol. 2016 ,37, 15795–15804. [C ossRe ] [PubMed] 75. Gabi o a-Co nell, L.; Su umbaye a, A.; Pe i, S.; F anco-Ba aza, J.; Res i o, D.; Wei z, N.; Ogie , C.; Goldman, A.; Ha man, T.R.; F ancescone, R.; e al. Choles e ol Pa hway Inhibi ion Induces TGF- β Signaling o P omo e Basal Di e en ia ion in Panc ea ic Cance . Cance Cell 2020 ,38, 567–583.e11. [C ossRe ] [PubMed] 76. Vasseu , S.; Guillaumond, F. LDL Recep o : An open ou e o eed panc ea ic umo cells. Mol. Cell. Oncol. 2015,3, e1033586. [C ossRe ] 77. Guillaumond, F.; Bidau , G.; Ouaissi, M.; Se ais, S.; Goui and, V.; Oli a es, O.; Lac, S.; Bo ge, L.; Roques, J.; Gaye , O.; e al. Choles e ol up ake dis up ion, in associa ion wi h chemo he apy, is a p omising combined me abolic he apy o panc ea ic adenoca cinoma. P oc. Na l. Acad. Sci. USA 2015 ,112, 2473–2478. [C ossRe ] 78. Zhang, G.; He, P.; Tan, H.; Budhu, A.; Gaedcke, J.; Ghadimi, B.M.; Ried, T.; Y an is, H.G.; Lee, D.H.; Mai a, A.; e al. In eg a ion o Me abolomics and T ansc ip omics Re ealed a Fa y Acid Ne wo k Exe ing G ow h Inhibi o y E ec s in Human Panc ea ic Cance . Clin. Cance Res. 2013,19, 4983–4993. [C ossRe ] 79. Lee, J.H.; Cho, Y.-R.; Kim, J.H.; Kim, J.; Nam, H.Y.; Kim, S.W.; Son, J. B anched-chain amino acids sus ain panc ea ic cance g ow h by egula ing lipid me abolism. Exp. Mol. Med. 2019,51, 1–11. [C ossRe ] 80. Maye s, J.R.; Wu, C.; Clish, C.B.; K a , P.; To ence, M.E.; Fiske, B.P.; Yuan, C.; Bao, Y.; Townsend, M.K.; Two oge , S.S.; e al. Ele a ion o ci cula ing b anched-chain amino acids is an ea ly e en in human panc ea ic adenoca cinoma de elopmen . Na . Med. 2014,20, 1193–1198. [C ossRe ] 81. Mae in, S.; Elpe in,J.M.; Lo shaw,E.; Sendle , M.; Speakman, S.D.; Takaku a, K.; Reiche , B.M.; Ma enino a,O.A.; G ippo, P.J.; Maye le, J.; e al. Roles o au ophagy and me abolism in panc ea ic cance cell adap a ion o en i onmen al challenges. Am. J. Physiol. Li e Physiol. 2017,313, G524–G536. [C ossRe ] 82. Tumas, J.; Baski o a, I.; Pe enas, T.; No kuniene, J.; S upas, K.; Sileikis, A. Towa ds a Pe sonalized App oach in Panc ea ic Cance Diagnos ics Th ough Plasma Amino Acid Analysis. An icance . Res. 2019 ,39, 2035–2042. [C ossRe ] [PubMed] 83. Holms om, S.R.; Oli e, K.P. P o ein b eakdown p ecedes panc ea ic umo de elopmen . Na . Med. 2014 , 20, 1097–1099. [C ossRe ] [PubMed] 84. Sousa, C.M.; Biancu , D.E.; Wang, X.; Halb ook, C.J.; She man, M.H.; Zhang, L.; K eme , D.; Hwang, R.F.; Wi kiewicz, A.K.; Ying, H.; e al. Panc ea ic s ella e cells suppo umou me abolism h ough au ophagic alanine sec e ion. Na . Cell Biol. 2016,536, 479–483. [C ossRe ] [PubMed] 85. Pa ke , S.J.; Amendola, C.R.; Hollinshead, K.E.R.; Yu, Q.; Yamamo o, K.; Enca naci ó n-Rosado, J.; Rose, R.E.; LaRue, M.M.; Sohn, A.S.W.; Biancu , D.E.; e al. Selec i e Alanine T anspo e U iliza ion C ea es a Ta ge able Me abolic Niche in Panc ea ic Cance . Cance Disco . 2020,10, 1018–1037. [C ossRe ] [PubMed] 86. Zhu, Z.; Ach eja, A.; Meu s, N.; Animasahun, O.; Owen, S.; Mi al, A.; Pa ikh, P.; Lo, T.; F anco-Ba aza, J.; Shi,J.; e al. Tumou - ep og ammeds omalBCAT1 uelsb anched-chainke oaciddependencyins omal- ich PDAC umou s. Na . Me ab. 2020, 1–18. [C ossRe ] [PubMed] 87. Oli a es, O.; Maye s, J.R.; Goui and, V.; To ence, M.E.; Gicquel, T.; Bo ge, L.; Lac, S.; Roques, J.; La au , M.-N.; Be hez è ne, P.; e al. Collagen-de i ed p oline p omo es panc ea ic duc al adenoca cinoma cell su i al unde nu ien limi ed condi ions. Na . Commun. 2017,8, 16031. [C ossRe ] 88. Al an, B.; Kai a, K.; Wa anabe, A.; Kubo, N.; Bao, P.; Dolgo maa, G.; Bilguun, E.-O.; A aki, K.; Kanai, Y.; Yokobo i, T.; e al. Rela ionship be ween LAT1 exp ession and esis ance o chemo he apy in panc ea ic duc al adenoca cinoma. Cance Chemo he . Pha macol. 2018,81, 141–153. [C ossRe ] 89. Dahe , B.; Pa ks, S.K.; Du i aul , J.; Co me ais, Y.; Baida jad, H.; Tambu e, E.; Pouyssegu , J.; Vuce ic, M. Gene ic Abla ion o he Cys ine T anspo e xCT in PDAC Cells Inhibi s mTORC1, G ow h, Su i al, and Tumo Fo ma ion ia Nu ien and Oxida i e S esses. Cance Res. 2019,79, 3877–3890. [C ossRe ] 90. Halb ook, C.J.; Lyssio is, C.A. Employing Me abolism o Imp o e he Diagnosis and T ea men o Panc ea ic Cance . Cance Cell 2017,31, 5–19. [C ossRe ] 91. Son, J.; Lyssio is, C.A.; Ying, H.; Wang, X.; Hua, S.; Ligo io, M.; Pe e a, R.M.; Fe one, C.R.; Mulla ky, E.; Shyh-Chang, N.; e al. Glu amine suppo s panc ea ic cance g ow h h ough a KRAS- egula ed me abolic pa hway. Na . Cell Biol. 2013,496, 101–105. [C ossRe ] J. Clin. Med. 2020,9, 4128 21 o 21 92. Daemen, A.; Pe e son, D.W.; Sahu, N.; Mcco d, R.; Du, X.; Liu, B.; Kowane z, K.; Hong, R.; Mo a , J.G.; Gao, M.; e al. Me aboli e p o iling s a i ies panc ea ic duc al adenoca cinomas in o sub ypes wi h dis inc sensi i i ies o me abolic inhibi o s. P oc. Na l. Acad. Sci. USA 2015 ,112, E4410–E4417. [C ossRe ] [PubMed] 93. Ka asinska, J.M.; Topham, J.T.; Kalloge , S.E.; Jang, G.H.; Den oche, R.E.; Culib k, L.; Williamson, L.M.; Wong, H.-L.; Lee, M.K.-C.; O’Kane, G.M.; e al. Al e ed Gene Exp ession along he Glycolysis–Choles e ol Syn hesis Axis Is Associa ed wi h Ou come in Panc ea ic Cance . Clin. Cance Res. 2020 ,26, 135–146. [C ossRe ] [PubMed] 94. Du, Y.; Zhao, B.; Liu, Z.; Ren, X.; Zhao, W.; Li, Z.; You, L.; Zhao, Y. Molecula Sub yping o Panc ea ic Cance : T ansla ing Genomics and T ansc ip omics in o he Clinic. J. Cance 2017,8, 513–522. [C ossRe ] [PubMed] Publishe ’s No e: MDPI s ays neu al wi h ega d o ju isdic ional claims in published maps and ins i u ional a ilia ions. © 2020 by he au ho s. Licensee MDPI, Basel, Swi ze land. This a icle is an open access a icle dis ibu ed unde he e ms and condi ions o he C ea i e Commons A ibu ion (CC BY) license (h p://c ea i ecommons.o g/licenses/by/4.0/).