EULAR study group on MHC-I-opathy: identifying disease-overarching mechanisms across disciplines and borders

887 Kuipe JJW, e al. Ann Rheum Dis 2023;82:887–896. doi:10.1136/ann heumdis-2022-222852 Re iew EULAR s udy g oup on ‘MHC- I- opa hy’: iden i ying disease- o e a ching mechanisms ac oss disciplines andbo de s Jonas JW Kuipe ,1 Jö g C P inz,2 E s a ios S a ikos,3 Pio Kuśnie czyk,4 Akiko A akawa,2 Sebas ian Sp inge ,5 Dillon Min o ,6,7 I an Padjen,8,9 Russka Shumnalie a,10 Seçil Vu al,11 Ina Kö e ,12,13 Ma leen G an de Sande,14,15 Ayşe Boy a ,16 Joke H de Boe ,1 Geo ge Be sias,17,18 Niek de V ies,14,15 Cha lo e LM K ieckae ,19,20 Inês Leal,21,22 Na aša Vido ič Valen inčič,23,24 Ilknu Tugal- Tu kun,25 Hanane el Khaldi Ahanach,26,27 Félicie Cos an ino,28,29 Simon Gla igny,30,31 Danijela M azo ac Zimak,32 Fabian Lö sche ,33 Floo G Ke s ens,19,20 Ma ija Bakula,8 Elsa Vie a Sousa ,34,35 Pe e Böhm,36 Kees Bosman,37 Tony J Kenna,38 Simon J Powis,39 Maxime B eban ,28,29 Ahme Gul,40 John Bowes ,41,42 Rik JU Lo ies ,43,44 Johannes Nowa zky,45,46 Ge i Jan Wolbink,19,47 Dennis G McGonagle,48,49 F ank ien Tu ks a ,19,20 EULAR s udyg oup MHC- I- opa hies To ci e: Kuipe JJW, P inzJC, S a ikosE, e al. Ann Rheum Dis 2023;82:887–896. Handling edi o Jose S Smolen ►Addi ional supplemen al ma e ial is published online only. To iew, please isi he jou nal online (h p:// dx. doi. o g/ 10. 1136/ a d- 2022- 222852). Fo numbe ed a ilia ions see end o a icle. Co espondence o D F ank ien Tu ks a, Rheuma ology, Ams e dam Rheuma ology and Immunology Cen e , Ams e dam 1105, The Ne he lands; . u ks a@ eade. nl Recei ed 21 July 2022 Accep ed 25 Janua y 2023 Published Online Fi s 27Ma ch2023 © Au ho (s) (o hei employe (s)) 2023. Re- use pe mi ed unde CC BY- NC. No comme cial e- use. See igh s and pe missions. Published by BMJ. ABSTRACT The ’MHC- I (majo his ocompa ibili y complex class I)- opa hy’ concep desc ibes a amily o in lamma o y condi ions wi h o e lapping clinical mani es a ions and a s ong gene ic link o he MHC- I an igen p esen a ion pa hway. Classical MHC- I- opa hies such as spondyloa h i is, Behçe ’s disease, pso iasis and bi dsho u ei is a e widely ecognised o hei s ong associa ion wi h ce ain MHC- I alleles and gene a ian s o he an igen p ocessing aminopep idases ERAP1 and ERAP2 ha implica es al e ed MHC- I pep ide p esen a ion o CD8+T cells in he pa hogenesis. P og ess in unde s anding he cause and ea men o hese diso de s is hampe ed by pa ien pheno ypic he e ogenei y and lack o sys ema ic in es iga ion o he MHC- I pa hway. He e, we discuss new insigh s in o he biology o MHC- I- opa hies ha s ongly ad oca e o disease- o e a ching and in eg a ed molecula and clinical in es iga ion o deciphe unde lying disease mechanisms. Because his equi es ans o ma i e mul idisciplina y collabo a ion, we in oduce he EULAR s udy g oup on MHC- I- opa hies o uni e clinical expe ise in heuma ology, de ma ology and oph halmology, wi h undamen al and ansla ional esea che s om mul iple disciplines such as immunology, genomics and p o eomics, alongside pa ien pa ne s. We p io i ise s anda disa ion o disease pheno ypes and scien i ic nomencla u e and p opose in e disciplina y gene ic and ansla ional s udies o exploi eme ging he apeu ic s a egies o unde s and MHC- I- media ed disease mechanisms. These collabo a i e e o s a e equi ed o add ess ou s anding ques ions in he e iopa hogenesis o MHC- I- opa hies owa ds imp o ing pa ien ea men and p ognos ica ion. THE INCEPTION OF THE MHC-I-OPATHY FAMILY In lamma ion agains sel is o ches a ed by a con inuum o incomple ely unde s ood inna e and adap i e immune mechanisms. The e m ‘au oin- lamma o y’ e e s o in lamma ion agains sel , caused by abno mal inna e immuni y, whe eas ‘au oimmuni y’ is caused by abe an adap i e immuni y.1 2 Since his dicho omous de ini ion o e looked condi ions such as pso iasis (PsO) and Behçe ’s disease (BD), he concep o ‘mixed- pa e n’ o ‘in e media e’ diseases was p oposed.3 Genome- wide associa ion s udies (GWAS) o MHC- I- associa ed diseases, such as BD (associ- a ed wi h HLA- B*51),4 5 PsO (associa ed wi h HLA- C*06:02),6–8 HLA- B*27- associa ed spondy- loa h i is (SpA)9–11 HLA- B*27- associa ed an e io u ei is (AU)12 and HLA- A*29- associa ed bi dsho u ei is (BU),13 14 e ealed ha hese ‘in e media e diseases’ sha e a dis inguishable gene ic backg ound de ined by MHC- I genes, he an igen p ocessing genes ERAP1 and ERAP2, and he IL- 17 pa hway gene IL23R. Such gene ic o e lap implica es MHC- I pep ide p esen a ion as he key mechanis ic commonali y. Fu he mo e, i subs an ia es he idea ha BD, PsO, SpA and BU belong o a dis inc disease clus e known as ‘MHC- I- opa hies’.15 The e is ongoing deba e and incomple e e idence ega ding unde lying mechanisms o MHC- I- opa hies.16–18 MHC- I p o eins (also called HLA- A, HLA- B and HLA- C) bind sho pep ides om deg aded o pa hogenic p o eins, which ha e been p o eolysed inside he cell by he p o ea- some.19–21 Mos MHC- I pep ides a e de i ed om p o eins om he hos . ERAP1 and ERAP2 a e endoplasmic e iculum aminopep idases associa ed wi h an igen p ocessing ha im a ce ain ac- ion o hese pep ides i hey a e no sho enough be o e loading hem on o MHC- I molecules.22 This p ocess enables MHC- I o p esen ens o hou- sands o pep ides on he cell su ace, collec i ely e e ed o as he ‘immunopep idome’.23 CD8+T cells ead ou he immunopep idome by binding o he pep ide- MHC- I complexes wi h hei T cell ecep o s (TCR) ( igu e 1). MHC- I molecules can P o ec ed by copy igh . on June 27, 2023 a Facul y o Medicine, Uni e si y o Lisbon.h p://a d.bmj.com/Ann Rheum Dis: i s published as 10.1136/a d-2022-222852 on 27 Ma ch 2023. Downloaded om P o ec ed by copy igh . on June 27, 2023 a Facul y o Medicine, Uni e si y o Lisbon.h p://a d.bmj.com/Ann Rheum Dis: i s published as 10.1136/a d-2022-222852 on 27 Ma ch 2023. Downloaded om P o ec ed by copy igh . on June 27, 2023 a Facul y o Medicine, Uni e si y o Lisbon.h p://a d.bmj.com/Ann Rheum Dis: i s published as 10.1136/a d-2022-222852 on 27 Ma ch 2023. Downloaded om P o ec ed by copy igh . on June 27, 2023 a Facul y o Medicine, Uni e si y o Lisbon.h p://a d.bmj.com/Ann Rheum Dis: i s published as 10.1136/a d-2022-222852 on 27 Ma ch 2023. Downloaded om P o ec ed by copy igh . on June 27, 2023 a Facul y o Medicine, Uni e si y o Lisbon.h p://a d.bmj.com/Ann Rheum Dis: i s published as 10.1136/a d-2022-222852 on 27 Ma ch 2023. Downloaded om P o ec ed by copy igh . on June 27, 2023 a Facul y o Medicine, Uni e si y o Lisbon.h p://a d.bmj.com/Ann Rheum Dis: i s published as 10.1136/a d-2022-222852 on 27 Ma ch 2023. Downloaded om P o ec ed by copy igh . on June 27, 2023 a Facul y o Medicine, Uni e si y o Lisbon.h p://a d.bmj.com/Ann Rheum Dis: i s published as 10.1136/a d-2022-222852 on 27 Ma ch 2023. Downloaded om P o ec ed by copy igh . on June 27, 2023 a Facul y o Medicine, Uni e si y o Lisbon.h p://a d.bmj.com/Ann Rheum Dis: i s published as 10.1136/a d-2022-222852 on 27 Ma ch 2023. Downloaded om 888 Kuipe JJW, e al. Ann Rheum Dis 2023;82:887–896. doi:10.1136/ann heumdis-2022-222852 Re iew also bind o kille - cell immunoglobulin- like ecep o s (KIRs) and o he ecep o s on na u al kille (NK) cells.24 The e is much conjec u e abou he cause(s) o MHC- I- opa hies15 25 26 and se e al popula hypo heses ha e been p oposed which a e no necessa ily mu ually exclusi e. The p ima y hypo hesis o he cause o MHC- I- opa hies is ha disease- associa ed MHC- I alleles p esen speci ic immunogenic pep ides ha igge ‘au oimmune’ eac ions (ie, he a h i o- genic pep ide heo y).27 The gene ic associa ion wi h ERAP genes also suppo s his hypo hesis since he ac i i y o hese enzymes can modi y he immunopep idome.28 P oo o concep has been shown in PsO and ecen ly in SpA and HLA- B*27+AU.29 The iden i ica ion o CD8+T cells in PsO eac agains melanocy es in he con ex o HLA- C06:02 as skin- speci ic a ge cells o he pso ia ic au oimmune esponse,30–32 while CD8+T cells om syno ial and eye luid o SpA and AU pa ien s ecognise bo h sel and mic obial pep ides p esen ed by HLA- B*27.29 The e is s ill no conclusi e e idence ha his mechanism unde lies o he MHC- I- opa hies since mechanis ic s udies a e echnically challenging o conduc , owing o mul io gan in ol e- men ha ing complex issues, which equi es labou - in ensi e echnology o sc een o many epi opes. Consequen ly, se e al al e na i e heo ies o MHC- I- opa hies ha e been p oposed, one o which sugges s ha MHC- I p o ein mis olding di ec ly leads o in lamma ion. Acco ding o his heo y, p edisposing MHC- I molecules may exhibi p ope ies which could cause excessi e mis olding and accumula ion in he ER, p omo ing he ‘un olded p o ein esponse’.33–38 S udies o (HLA- B*27) ans- genic animals and cellula models suppo his hypo hesis, bu he e is a pauci y o ansla ional e idence.39–43 A hi d popula hypo hesis sugges s ha he p edisposing MHC- I alleles a e ecognised by KIRs o leucocy e immunoglobulin- like ecep o s (LILRs) on he cell su ace o NK cells.44 45 In ou opinion, he i s hypo hesis applies o he majo i y o MHC- I- opa hies (wi h he mos obus e idence o PsO and SpA), bu de ini i e p oo o CD8+T cell- media ed pa hologies is lacking o se e al o he condi ions. Hypo heses 2 and 3 may also apply o ce ain condi ions. Fo example, in BD, ERAP1 may media e HLA- B51 ecogni ion ia NK cells,17 46 47 and pa ho- gens ha can cause eac i e a h i is induce un olded p o ein esponses in HLA- B*27- posi i e indi iduals.48 While hese o he pa hways and mechanisms a e implica ed, including he e y in e es ing in e ac ions o al e ed mic obiomes in pa ien s,49 we ocus ou discussion on he MHC- I pa hway as he key de e mi- nan o his amily o complex condi ions. THE MANY FACES AND CHALLENGES OF MHC-I-OPATHIES Se e al condi ions a e conside ed o be ‘classical’ MHC- I- opa hies (PsO, pso ia ic a h i is (PsA), SpA, B*27- AU, BD and BU) and sha e s ikingly simila clinical symp oms ( able 1). BU is a a e and se e e ype o u ei is, leading o e inal damage and ision loss ha exclusi ely a ec s HLA- A*29- posi i e indi- iduals.50 51 Al hough i is unclea which o he clinical ea u es a e sha ed be ween BU and o he MHC- I- opa hies, 1 s udy o 118 cases e ealed ha many pa ien s also su e om a h algia and PsO.52 We also discuss PsA because i sha es many cha ac e - is ics wi h PsO, including s ong associa ion wi h MHC- I alleles and IL23R.7 8 53–55 While some pa ien s wi h in lamma o y bowel disease may ha e simila symp oms,56 we will only discuss clas- sical MHC- I- opa hies he e. MHC- I- opa hies o e lap in hei pa e n o o gan in ol e- men ( able 1). U ei is, o ins ance, is a disease ea u e epo ed in e e y classical MHC- I- opa hy, al hough wi h di e en p e - alence and ana omical loca ion (an e io /pos e io ).57 58 Sac o- ilii is is p esen in pa ien s wi h SpA, PsA as well as BD.59–61 Cu aneous in ol emen is also a sha ed ea u e o MHC- I- opa hies ( able 1). Howe e , no e e y pa ien exhibi s he symp oma ic hall- ma ks o e e y clinical en i y. Fo example, a e ial, enous and neu ological complica ions a e common in BD, bu in equen in o he MHC- I- opa hies.18 62 Fo se e al MHC- I- opa hies, pa ien s wi h he associa ed isk MHC- I alleles a e mo e likely o mani es ea ly- onse disease and a wo se p ognosis.63–66 Fu he mo e, subs an ial clinical and geog aphical a ia ion in disease pheno ypes exis s, o example, he p e alence o gas o- in es inal in ol emen in BD in Asian e sus Eu opean popula- ions.67 68 WHAT YOU (DO NOT) SEE IS WHAT YOU (DO NOT) GET! The clinical managemen o MHC- I- opa hy pa ien s is compli- ca ed by he e ogenei y in age o onse , symp oms and disease cou se. Unlike cases wi h commonly ecognised symp oms (e.g., Figu e 1 An o e iew o he ole o he MHC- I pa hway in MHC- I- opa hies. The p o easome p oduces pep ide agmen s ha a e anspo ed in o he endoplasmic e iculum by he anspo e associa ed wi h an igen p ocessing (TAP) and immed by ERAP1 and ERAP2 (ERAP) o a leng h o 8- 11 amino acids be o e binding o MHC- I molecules. A e a icking o he cell su ace he MHC- I- pep ide complex is “ ead ou ” by su eying immune cells, igge ing an igen- speci ic CD8+ T cell esponses o na u al kille (NK) cell ac i a ion. MHC- I- opa hies a e gene ically associa ed wi h unc ionally dis inc a ian s o MHC- I and ERAP which al e he pep ide epe oi e p esen ed by MHC- I. Au o eac i e T cells in he pe iphe y ha escape ole ance mechanisms and p omo e in lamma ion agains sel - pep ide epi opes. Bio ende so wa e was used o c ea e elemen s om his igu e unde an academic license. P o ec ed by copy igh . on June 27, 2023 a Facul y o Medicine, Uni e si y o Lisbon.h p://a d.bmj.com/Ann Rheum Dis: i s published as 10.1136/a d-2022-222852 on 27 Ma ch 2023. Downloaded om 889 Kuipe JJW, e al. Ann Rheum Dis 2023;82:887–896. doi:10.1136/ann heumdis-2022-222852 Re iew u ei is in SpA pa ien s), asymp oma ic o a ypical in ol emen o he skin, bowel o o he como bidi ies in pa ien s may be o e - looked ( able 1). Fo example, eexamina ion o SpA pa ien s e ealed ha up o one- hi d may ha e como bid PsO.69 In he DUET s udy, o e 40% o pa ien s wi h B*27- AU we e diag- nosed wi h SpA o PsA on e- e alua ion by a heuma ologis ,70 which was con i med by o he s udies.71 La ge popula ion- le el da a also co ela e disease mani es a ions o MHC- I- opa hies such as u ei is, PsO, PsA and BD.58 72 73 Obse a ions om well- powe ed coho s udies subs an ia e ha o al disease, which is a hallma k o BD, is also linked o SpA.74–76 Despi e conside able pheno ypic he e ogenei y, hese s udies suppo ha MHC- I- opa hies a e in e connec ed condi ions ha canno be unde - s ood in isola ion and equi e a mul idisciplina y app oach. The human pheno ype on ology (HPO) p o ides a amewo k o s anda dised nomencla u e o disease symp oms, which can acili a e imp o ed classi ica ion o disease pheno ypes.77 Al hough o iginally designed o sys ema ically cap u e he clin- ical mani es a ions o a e, monogenic condi ions, HPO has mo e ecen ly been used o success ully in e se e al a e pheno- ypes o he UK Biobank.78 In i s cu en o m, he HPO may no be op imal o he anno a ion o he clinical spec um o pa ien s wi h MHC- I- opa hies. As a esul , he EULAR s udy g oup aims o e alua e he HPO and adap i o i he symp oms o MHC- I- opa hies. The spec um o MHC- I- opa hies will bene i om s anda disa ion o disease mani es a ions, allowing exis ing coho s o be me ged in o a well- powe ed s udy. The p ecise delinea ion o clinical pheno ypes will allow us o ela e hem o molecula endo ypes. We expec ha his p ocess will acili a e he disco e y (and alida ion) o be e diagnos ic, p ognos ic and he apeu ic bioma ke s. A COMMON GENETIC ARCHITECTURE MHC-I, he ip o he icebe g S ong gene ic associa ion wi h ce ain MHC- I alleles is he hallma k o he MHC- I- opa hy clus e : MHC- I associa ion s udies da e back o 1973 wi h he disco e y o he associa ion o HLA- B*27 and SpA as well as HLA- B*51 and BD ( o me ly ‘HL- A5’),79–81 ollowed by epo s on HLA- C*06:02 (p e iously known as ‘HLA- Cw6’) and PsO in 1977,82 83 and he associa ion be ween HLA- A*29 and BU in 198284 ( able 2). In compa ison o genes associa ed wi h complex in lamma o y condi ions, he e ec size o MHC- I alleles accoun s o a disp opo iona e amoun o gene ic isk. Fo almos 50 yea s, esea che s ha e s uggled o unde s and he ole hese class I alleles play in hei disease biology. In e es ingly, ecen ine- mapping s udies showed ha s a is- ical adjus men o HLA- B*27 in SpA e ealed independen associa ions o o he MHC- I alleles, including HLA- A*02:01, HLA- B*07, HLA- B*57 and HLA- B*4085 86 ( able 2). This is signi ican because i also implica es he MHC- I pa hway o cases lacking he p ima y MHC- I isk allele and s ongly inc imina es pep ide p esen a ion a he han al e na i e mechanisms. Associa ion wi h se e al o hese alleles was also ound a e co ec ing o he p ima y isk MHC- I allele in PsO (HLA- A*02:01, HLA- B*27 and HLA- B*07), BD (eg, HLA- B*27 and HLA- B*57), PsA (eg, HLA- B*07 and HLA- A*02) and AU.7 12 87–90 To da e, small GWAS in he a e BU had limi ed powe o de ec HLA- A*29- independen loci in de ail, bu also epo ed inde- penden isk MHC- I alleles.13 14 These indings aise he possi- bili y ha an ensemble o disease- o e a ching MHC- I alleles con ibu e o MHC- I- opa hy suscep ibili y. Table 1 Summa y o issue in ol emen pe MHC- I- opa hy, o ganised pe clinical special y ( e e ences unde lying he summa ised da a and sco es can be ound in online supplemen al able 1 Disease PsO* PsA† SpA B*27 AU BD BU Medical special y P ima y isk MHC- I- allele(s) C*06 C*06/B*27 B*27 B*27 B*51 A*29 P ognosis wo se when p ima y MHC- I allele p esen 3 0 3 n.a 3 0 Oph halmology U ei is‡ 1 1 3 3 3 3 De ma ology O al ulce a ions 0 1 1 0 3 0 De ma ology Geni al ulce a ions 0 0 0 0 3 0 De ma ology Pso iasi o m de ma i is§ 3 3 2 2 1 1 De ma ology Pus ula lesions¶ 2 2 1 0 3 0 De ma ology E y hema nodosum- like lesions 0 0 0 0 3 0 Rheuma ology Spondyli is 1 3 3 3 1 0 Rheuma ology A h i is 2 3 3 2 3 0 Rheuma ology En hesi is 2 3 3 3 1 0 Rheuma ol/immunol Vasculi is** 1 1 1 0 3 0 Gas oen e ology In lamma o y bowel disease 1 1 2 1 2 0 In e nal medicine Como bid hype ension 1 2 2 0 0 2 Neu ol/In Med/ca diol Como bid ca dio asc disease 2 2 2 0 1 0 Legend: 3 pa o he disease ec um 2 egula ly epo ed 1 in equen ly epo ed 0 3 pa o he disease spec um. 2 egula ly epo . 1 in equen ly epo ed. 0 ei he unknown / no epo s / no p esen . *Pso iasis: besides plaque pso iasis. This encompasses o he o ms o pso ia ic disease like pso iasis gu a e and (se e al ypes o ) pus ula pso iasis. †PsA: bo h axial and pe iphe al disease. ‡U ei is an e io is he main sub ype epo ed in PsO, PsA, SpA, whe eas in Behçe ’s mul iple ana omical sub ypes o u ei is a e epo ed. BU mani es s as pos e io u ei is. §Pso iasi o m lesions: e e s o he se e al ypes o pso iasis; classical plaque pso iasis, gu a e, nail lesions and e y hema ous as well as pus ula lesions. ¶Pus ula lesions: co e s acnei o m, papulopus ula and non- ollicula pus ules. **Vasculi is in PsO as well as in PsA and SpA asculi is is in he la ge essels (ao i is); in B27- AU and BU no epo ed ou side he eye; in Behçe ’s asculi is is in all ypes o essels, a e ies and eins. BU, bi dsho u ei is; PsA, pso ia ic a h i is; PsO, pso iasis; SpA, spondyloa h i is. P o ec ed by copy igh . on June 27, 2023 a Facul y o Medicine, Uni e si y o Lisbon.h p://a d.bmj.com/Ann Rheum Dis: i s published as 10.1136/a d-2022-222852 on 27 Ma ch 2023. Downloaded om 890 Kuipe JJW, e al. Ann Rheum Dis 2023;82:887–896. doi:10.1136/ann heumdis-2022-222852 Re iew The e o e, unc ional s udies ha conside only one MHC- I allele may no cap u e he complexi y o he MHC- I pa hway in pa ien s. This emphasises he need o use p ima y pa ien issues o in es iga e disease mechanisms. I emains o be de e mined whe he he ull MHC haplo ype (including ‘seconda y’ isk MHC- I alleles) imp o es pa ien s a i ica ion. La ge popula ion- based s udies (ie, UK Biobank) suppo ha MHC- I alleles a e associa ed wi h a a ie y o heal h bioma ke s.91 A i s s ep in o his di ec ion could be he conduc ion o a mul iances al MHC- I- opa hy GWAS analysis by combining se e al a ailable la ge- scale genome- wide da ase s and in e oga ing he MHC o di e en pheno ypic s a es. The de il is in he ERAP1 and ERAP2 de ails Pe haps one o he majo accomplishmen s o he p og ess in he unde s anding o MHC- I- opa hies was he disco e y o he associa ion wi h he ERAP1 and/o ERAP2 genes.4 6 7 9 12–14 92–95 These genes encode wo ER- esiden enzymes specialised in im- ming pep ides o acili a e o p e en hei binding in he g oo e o MHC- I.96 97 By gene a ing and des oying pep ide epi opes, ERAPs can a ec CD8+T cell and NK cell esponses.98–101 Gene ic a ian s in ERAP1 and ERAP2 a ec he enzyma ic ac i i y and exp ession le els o hese enzymes.93 102 Conse- quen ly, a change in ERAP ac i i y may expose CD8+T cells o al e ed pep ide epe oi es (sel o non- sel ) ia MHC- I isk alleles, which can be ha m ul.28 Gene ic associa ion be ween ERAP1 and MHC- I- opa hies is ypically obse ed in indi iduals ca ying he p ima y isk MHC- I.4 6 11 13 85 93 Coding a ian s in ERAP1 o ganise in o se e al common haplo ypes o en e e ed o as ERAP1 ‘allo ypes’103 104 ha exhibi a wide ange o enzyma ic ac i i ies owa ds pep ide subs a es and di e en ially shape he immunopep idome o MHC- I.28 105 Risk polymo phisms in ERAP1 (and ERAP2) a e also s ongly associa ed wi h mRNA and p o ein exp ession le els o hese aminopep idases.50 102 106 Haplo ype- based anal- yses ha e singled ou speci ic ERAP1 allo ypes as isk ac o s o MHC- I- opa hies. While se e al e ms ha e been p oposed o ERAP1 allo ypes, s anda dised nomencla u e has ye o be widely adop ed. One unc ionally dis inc ERAP1 allo ype (o en e e ed o as Haplo ype 10 (hap10)) is a isk ac o o BD and BU,93 107 bu p o ec i e o SpA, AU and PsO.28 108 In e es ingly, in PsO, he p o ec i e hap10 was less e ec i e in gene a ing he au oan igenic epi ope han he isk haplo ypes o ERAP1, leading o lowe HLA- C exp ession and immunogenici y o melanocy es.31 ERAP1 may also in luence NK cell esponses ia inhibi o y ecep o s NKG2A/CD94 (also exp essed by CD8+T cells109) o non- classical MHC- I molecule HLA- E.46 The inhibi o y ac i i y o HLA- E equi es he p esen a ion o a signal sequence om MHC- I molecules, which a e also p esen in HLA- A29, HLA- B27 and HLA- B51.110–112 The e o e, ERAPs may also a ec NK cells and CD8+T cells ia MHC- I- ela ed molecules, as was p e iously shown in cance models.46 113 Al hough KIR ecep- o s can ecognise immunopep idome changes caused by ERAP1, KIR genes do no in luence HLA- B*27 and ERAP1- media ed ankylosing spondyli is isk.114 115 This sugges s ha he disease mechanisms media ed by ERAP1 and MHC- I a e less dependen on KIRs. In con as o ERAP1, ERAP2 gene ic a ian s a e no associ- a ed wi h all MHC- I- opa hies (eg, BD). Also, ERAP2 is associ- a ed wi h SpA ega dless o HLA- B*27 s a us. Because he e is also epis asis be ween ERAP1 and HLA- B*40 in SpA (indepen- den o HLA- B*27),85 i is possible ha ERAP2 modi ies disease in SpA ia al e na i e isk MHC- I alleles. Func ional s udies suppo ha ERAP2 signi ican ly a ec s he immunopep idome o many MHC- I alleles, including HLA- B*40115 116 No e ha ERAP2 allo ypes co- occu non- andomly wi h ERAP1 allo ypes.93 105 Fu he mo e, al hough HLA- A*29 is common in many egions, HLA- A*29- posi i e indi iduals who ca y bo h ERAP1 and ERAP2 isk alleles a e only obse ed in coun ies whe e BU is p e alen .93 The e o e, an indi idual’s ERAP1 and ERAP2 allo ypes along wi h hei MHC- I p o ile (and T cell epe oi e) a e mos likely o de e mine hei suscep- ibili y o MHC- I- opa hies.117 S udies linking ERAP geno ypes wi h clinical end poin s may ha e po en ial,118 119 bu we would like o emphasise ha hese s udies should be ca e ully con olled and well powe ed. Bo h ERAP1 and ERAP2 a e common denomina o s o MHC- I- opa hies, which place an igenic pep ide p esen a ion a he hea o hei pa hogenesis. IL23R and T cells The e a e many o he genes associa ed wi h condi ions wi hin he MHC- I- opa hy spec um ha ha e been disco e ed h ough GWAS. While hey a e impo an o disease biology, we only b ie ly discuss IL23R, a ecep o o IL- 23 exp essed by T cells (and Table 2 Repo ed HLA class I associa ions in ou MHC- I- opa hies MHC- I- opa hy P e alence P ima y HLA class I associa ion % cases nega i e o p ima y HLA class I allele Independen * HLA class I associa ions Bi dsho u ei is 1- 5/500 000 HLA- A*29:02 0HLA- A*3013 14 HLA- A*3314 Spondyloa h i is† 0.5% HLA- B*27 ~ <30 HLA- B*4085 86 HLA- A*0285 HLA- B*0785 HLA- B*5785 HLA- C*1586 Behçe ’s Disease 0.19- 120/100 0009HLA- B*51 ~30–70 HLA- A*0287 HLA- B*2787 HLA- B*5787 HLA- A*0387 HLA- B*1587 HLA- B*4987 HLA- A*2687 89 HLA- C*0789 Pso iasis 2–4% HLA- C*06:02 ~30–70 HLA- A*02122 HLA- B*27122 HLA- B*07122 HLA- C*07176 *Iden i ied by s a is ical adjus ing o p ima y associa ed HLA class I allele. †Majo i y o da a a e om gene ic s udies in ankylosing spondyli is. Includes bo h isk and p o ec i e alleles. P o ec ed by copy igh . on June 27, 2023 a Facul y o Medicine, Uni e si y o Lisbon.h p://a d.bmj.com/Ann Rheum Dis: i s published as 10.1136/a d-2022-222852 on 27 Ma ch 2023. Downloaded om 891 Kuipe JJW, e al. Ann Rheum Dis 2023;82:887–896. doi:10.1136/ann heumdis-2022-222852 Re iew inna e lymphocy es), because i is common o MHC- I- opa hies and is associa ed wi h disease se e i y and pheno ypes.8 118 120–122 Fascina ingly, despi e IL23R exp ession by CD4+T cells, epigen- e ic analyses implica e CD8+T cells as majo pe pe a o s o MHC- I- opa hies.88 In e leukin- 17- p oducing CD8+T cells ( e med ‘Tc17’) in il a ing skin and syno ial lesions in PsO, BD, SpA and PsA pa ien s exp ess IL23R.123–125 Tc17 cells a e also mo e abundan in pa ien s wi h BU.126 127 IL23R’s ole in he pa hophysiology o MHC- I- opa hies is incomple ely unde - s ood, bu likely o be issue- dependen .128 This may explain why pa ien s wi h PsO129 and PsA130 exhibi clinical esponse o he apy ha dis up s T cell IL- 23 signalling, while ini ial ials we e less success ul in SpA.131 132 A be e unde s anding o clinical and molecula ea u es will help o e come challenges posed by pa ien he e ogenei y as well as iden i y he apeu ic bioma ke s which will guide he selec ion o candida es eligible o ea men wi h IL- 23 inhibi o s.21 128 131–134 UNMET NEEDS IN MHC-I-OPATHY PATHOPHYSIOLOGY UNDERSTANDING E idence o au o eac i e CD8+ T cell in ol emen A numbe o immunopep idome s udies in cell models ha e shown ha polymo phisms in ERAP cause change in he pep ides p esen ed by HLA- B27, HLA- B51, HLA- A29 and o he MHC- I alleles.28 115 116 Ci cums an ial e idence sugges s ha hese enzymes in oduce o emo e pep ides ha bind o isk MHC- I alleles and signal CD8+T cells o a ack heal hy issues. The ac ha CD8+T cells a e clono ypically expanded in pa ien s wi h SpA, PsO and PsA suppo s his concep .135–138 In BD, ca ie s o he disease- associa ed ERAP1 allo ype107 show en ichmen o ci cula ing an igen- expe ienced e ec o CD8+T cells and ERAP1 modula ion in luenced CD8+T cell esponses.107 The lack o iden i ica ion o causa i e au oan igens o indeed alloan igens has esul ed in discussion abou whe he CD8+T cells d i e pa hology in MHC- I- opa hies.17 Rega dless, au oan igen- de i ed pep ide ecogni ion by CD8+T cells in pa ien s has p e iously been epo ed, including an HLA- B51- p esen ed pep ide de i ed om a s ess- inducible au oan igen in BD,139 HLA- C06:02 p esen ed pep ide om inna e hos de ence p o ein LL- 37 in PsO,140 and HLA- B27- es ic ed epi ope om a pep ide ho mone ecep o and ca ilage- de i ed pep ides in SpA.141 142 To da e, he mos compelling concep ual p oo ha CD8+T cells media e au oimmune in lamma ion is based on s udies o PsO, and e y ecen ly in HLA- B*27- posi i e SpA and AU pa ien s.29 30 Skin lesional CD8+T cells in PsO can ecognise an HLA- C06:02- es ic ed au oan igen epi ope om ADAMTSL5 highly exp essed in skin melanocy es.30 31 ADAMTSL5- speci ic CD8+T cells sec e e PsO- p omo ing cy okines (eg, IL- 17) speci ically a e ecognising melanocy e- pep ide p ocessed by ERAP1 and p esen ed by he disease- associa ed MHC- I HLA- C06:02.30–32 He e, he immunogenici y o melanocy es o sel - eac i e CD8+T cell esponses was inc eased by disease- associa ed ERAP1 haplo ypes h ough g ea e supply o he pep ide au oan igen.30 I has, he e o e, been sugges ed ha pha macological modula ion o ERAP ac i i y owa ds p ecu so pep ides speci ically p esen ed by MHC- I alleles could e e se in lamma ion in MHC- I- associa ed diseases.143 144 Resea che s ecen ly ound ha issue- in il a ing CD8+T cells sha ed TCRs in eye liquid as well as syno ial luid o HLA- B*27- posi i e pa ien s wi h AS and AU.29 These CD8+T cells speci ically ecognise mic obial (eg, YEIH p o ein om eac i e a h i is- igge ing pa hogens) and sel - an igens (eg, pep ides om GPER1 o PRPF3 p o eins) speci ically wi hin he con ex o HLA- B27. Acco ding o hese indings, en i onmen al pa ho- gens may igge au oimmuni y ia CD8+T cell ac i a ion in MHC- I- opa hies, hus suppo ing he p ima y hypo hesis o he MHC- I- opa hy pa hogenesis. Fu u e esea ch migh explo e whe he HLA- B27 p esen a ions o hese pep ides a e a ec ed by isk allo ypes o ERAP1 and whe he pha macological a ge ing o ERAPs in e e es wi h hese esponses. I emains unclea why o he housands o sel - pep ides in he immunopep idome only a mino i y become immunogenic, while he majo i y emain ole able. Howe e , T cell au oan i- gens o en ha e pos - ansla ional modi ica ions o show al e ed binding con o ma ion.145–147 Wha igge s CD8+T cell sel - eac i i y in MHC- I- opa hies emains unknown. The classical iew is ha nega i e selec ion in he hymus elimina es au o eac i e T cells. Some sel - eac i e CD8+T cells manage o escape his il e ing p ocess and a e ein oduced in o he ci cula ion (some imes a high equencies) bu kep in check by ole ance mechanisms.148–150 In e es ingly, ecen wo k sugges s ha hymic egula- o y T- cells, a he han nega i e selec ion o au o eac i e T cells, en o ce p o ec ion agains au oimmuni y.148 151 He e, he cy okine IL- 23 elimina es hymic egula o y T cells in an IL23R- dependen manne ,152 while selec i ely en iching IL23R- exp essing CD8+T cells.153 Mo eo e , he e is no sha p a ini y h eshold o he ecogni ion o MHC- pep ide complex by TCRs, and CD8+T cells wi h o he wise low a ini y TCRs can be ac i a ed by a la ge inc ease in p esen ed au oan igen.154 155 This also i s wi h he ecen ly p oposed ‘au oimmune su eillance o hype sec e ing mu an s’ heo y ha links high au oan igen le els o T- cell au oimmuni y.156 C oss- p esen a ion o ex acellula an igens in dend i ic cells can also lead o he en y o ex a- cellula an igens in o he MHC class I pa hway, he eby g ea ly expanding he po en ial pool o immunogenic pep ides. Concep- ually, his in eg a es he possibili y o mic obial agen s causing disease, as demons a ed o SpA, and AU.29 Vi us- igge ed Box 1 The aims o he EULAR s udy g oup on "MHC- I- opa hy" 1. Mul idisciplina y collabo a ion be ween heuma ologis s, de ma ologis s and oph halmologis s o consensus and s anda dised anno a ion o disease symp oms. 2. De ailed pheno ypic e alua ion by pa ien - epo ed symp oms/ou comes. 3. In eg a ion o GWAS da a o MHC- I- opa hy- ela ed diseases, ac oss a la ge numbe o exis ing coho s, o acili a e ine mapping o he gene ic basis. 4. Ha monisa ion o he nomencla u e (eg, ERAP allo ypes) and p o ide expe syn hesis o cu en bes p ac ice o he s udy o key aspec s o he biology in MHC- I- opa hies. 5. Es ablishmen o a pan- Eu opean conso ium wi h s anda dised clinicopa hological disease pheno ypes om aim 1 and 2, (complemen ed by molecula da a on ERAP and MHC- I haplo ypes and possibly o he biological da a such as me agenomics o assess mic obiome in ol emen and TCR- epe oi e da a) o imp o ed disease classi ica ion, diagnos ic c i e ia and p ognos ic bioma ke s o p edic ion o disease p og ession and e icacy o ( ype o ) he apy. 6. E alua ion o MHC- I- opa hies in di e en e hnic backg ounds, gi en he massi e he e ogenei y wi hin class- 1 an igens. 7. Pa ien pa icipa ion: in ol emen o pa ien esea ch pa ne s. P o ec ed by copy igh . on June 27, 2023 a Facul y o Medicine, Uni e si y o Lisbon.h p://a d.bmj.com/Ann Rheum Dis: i s published as 10.1136/a d-2022-222852 on 27 Ma ch 2023. Downloaded om 892 Kuipe JJW, e al. Ann Rheum Dis 2023;82:887–896. doi:10.1136/ann heumdis-2022-222852 Re iew clonal CD8+T cell esponses a e p ocessed h ough MHC class I, and some o hese esponses a e con olled by ERAP1.157 Recen echnological ad ancemen s which ha e inc eased he sensi i i y and scale o analysing immunopep idomes o p ima y pa ien issues (ideally sampled a he a ec ed o gans) as- well as high- h oughpu p o iling o (au o)an igen- speci ic T- cell epe oi es (ie, single- cell TCR sequencing) may help iden i y CD8+T cell- media ed disease mechanisms in MHC- I- opa hies in g ea e de ail.158–161 Towa ds MHC-I pa hway he apy This s udy g oup’s ul ima e goal is o imp o e disease ou come o MHC- I- opa hies. Al hough de ini e disease mechanisms need o be es ablished, a ailable clinical and molecula e idence allow us o ou line se e al po en ial s a egies. Gi en ha MHC- I is conside ed a oo cause o MHC- I- opa hies, he a- peu ic a ge ing o an igen p ocessing and p esen a ion seems sel - e iden . This may be achie ed by in e en ions aimed a dis up ing cy okine signalling (see sec ion IL23R and T cells) o s a egies ha acili a e es o a ion o he mic obiome.48 Pa ien s wi h MHC- I- opa hies may ha e an al e ed mic obio a,162–164 bu heal hy indi iduals may also show mic obio a composi ions ha clus e acco ding o hei HLA alleles (eg, HLA- B*27, HLA- A*29).165 Eme ging T cell- an igen disco e y app oaches wi hin he mic obiome may p o ide an exci ing ield o upcoming s udies.166 In case o au oan igen- media ed pa hology, i may be possible o speci ically nega e T cell in e ac ion by an ibodies o small compounds ha speci ically block access o MHC- I- pep ide complexes. T- cell engagemen may also be blocked by p e en ing o changing he abundance o a ge pep ide p esen a ion by manipula ion ups eam o MHC- I, including he cellula p o eome (eg, chemo he apy), o pha macological inhibi ion o modula ion o he p o easome, TAP o he an igen loading complex,167–171 al hough wi h limi a ions in speci ici y a he cos o po en ial ad e se e ec s. Inhibi ing o , depending on he disease, enhancing he ac ion o ERAP1 and ERAP2 may be a p omising app oach, since hese enzymes a e highly specialised o an igen p esen a ion, and much is known abou hei s uc u e and unc ion o allow he de elopmen o inhibi o s o enhance s.143 144 172 The ac ha hei impac on an igen p esen a ion may be limi ed o a pa o he immunopep- idome,173 may cons i u e a middle g ound be ween single an igen s a egies (an ibodies o MHC/pep ide complex) and gene al supp ession o he MHC- I pa hway. Mos o hese he apeu ic ‘op ions’ a e s ill in hei in ancy and equi e ansla ional s udies in sui able p eclinical models. Al hough he HLA- B*27- ansgenic oden models,174 ha e p o ided aluable insigh s in o he disease mechanisms o MHC- I- opa hies, he e emains an unme need o addi ional ansgenic MHC- I models. To de e mine i i is possible o a ge he MHC- I pa hway he apeu ically in pa ien s, hese models should be ‘ ully’ humanised and cap u e a b oade spec um o clin- ical and molecula cha ac e is ics. Mission o he EULAR s udy g oup on MHC-I-opa hies As a esul o he complexi y o he clinical pheno ypes and he lack o knowledge abou he unde lying mechanisms o MHC- I- opa hies, in e na ional c oss- disciplina y collabo a ions and comple- men a y scien i ic expe ise a e u gen ly needed. The EULAR s udy g oup on MHC- I- opa hies p o ides an in e na ional ne wo k ha b ings medical specialis s, ansla ional and undamen al scien is s unde one umb ella wi h he aim o coope a i ely o e coming long- s anding unme needs in he disease managemen and unde s anding o he biology o MHC- I- opa hies. The s udy g oup (cu en ly >50 pa icipan s: de ma ologis s, oph halmologis s, heuma ologis , scien is s and pa ien ep e- sen a i es om >15 coun ies) was ounded in 2020 amids he COVID- 19 pandemic. The global pandemic es ic ed ini ial discus- sion o online mee ings. An inaugu al mee ing ook place in May 2022, in Ams e dam, ollowed by a mee ing du ing EULAR in June 2022 in Copenhagen. S udy g oup esea ch and collabo a ions will ocus on he pa hophysiology o MHC- I pa hway in hese condi- ions. B ie ly, he s udy g oup aims a e summa ised in box 1 and he objec i e is o ha monise, acili a e and imp o e esea ch me hod- ology and e minology, s udy disease mechanisms mo e collec i ely; os e basic and ansla ional knowledge exchange in an in e disci- plina y ashion h ough mee ings ia symposia du ing EULAR mee - ings(h ps://www.eula .ch/myUploadDa a/ iles/s udy_g oup_aims_ mhc_i_opa hy_ o _web.pd ) and dissemina e p og ess ia social media (eg, an open Linked- in page o in e es ed colleagues, h ps:// www.linkedin.com/g oups/12722534/). To accomplish hese objec- i es, he S udy G oup o med se e al mul idisciplina y ask o ces composed o clinicians, biologis s and pa ien ep esen a i es o p io i ise unme esea ch needs ha would equi e c oss- Eu opean collabo a ion. Fo example, one o he ask o ces aims o conduc me a- analysis o GWAS da a o he MHC- I- opa hies o ine map he MHC and iden i y no el isk loci in ela ion o clinical ea u es. Ano he ask o ce cu en ly wo ks on e alua ion o a pa ien - epo ed symp om in as uc u e, which has al eady been success- ully employed in COVID- 19 s udies.175 Al hough cu en ly all wo k wi hin he s udy g oup is con ibu ed in kind by i s membe s, he apidly g owing s udy g oup aims o apply o ex e nal unding o esea ch. This will also be equi ed o achie e mo e ambi ious goals, such as he collec ion o bioma e ials o os e inno a i e esea ch by deep immunop o iling (eg, T- cell epe oi es, MHC- I immu- nopep idomes) and ansla ional s udies (eg, ERAP modula ion in pa ien issues). The EULAR s udy g oup will complemen hei scien i ic objec i es wi h he o ganisa ion o in e ac i e wo kshops and symposia connec ed o EULAR o exchange basic, ansla ional and clinical knowledge in an in e disciplina y ashion and u he acili a e he g ow h o he s udy g oup by inclusion o physicians and scien is s ac i e in his ield. In conclusion, he EULAR s udy g oup on MHC- I- opa hies b idges a a ie y o medical scien i ic disciplines wi h he ambi- ious join objec i e o conduc an in eg a ed in es iga ion o MHC- I- opa hies o disco e he cause and cu e o a a ie y o complex in lamma o y condi ions. Au ho a ilia ions 1Depa men o Oph halmology, Cen e o T ansla ional Immunology, Uni e si y Medical Cen e U ech , U ech , The Ne he lands 2Uni e si y Hospi al, depa men o De ma ology and Alle gy, Ludwig Maximilians Uni e si y Munich, Munchen, Ge many 3Labo a o y o Biochemis y, Depa men o Chemis y, Na ional and Kapodis ian Uni e si y o A hens, A hens, G eece 4Labo a o y o Immunogene ics and Tissue Immunology, Ins i u e o Immunology and Expe imen al The apy Ludwik Hi sz eld Polish Academy o Sciences, W oclaw, Poland 5School o Science, Cons uc o Uni e si y B emen gGmbH, B emen, Ge many 6Depa men o De ma ology, Ma e Dei Hospi al, Msida, Mal a 7Depa men o Pa hology, Uni e si y o Mal a Facul y o Medicine and Su ge y, Msida, Mal a 8Di ision o Clinical Immunology and Rheuma ology, Uni e si y Hospi al Cen e Zag eb Depa men o In e nal Medicine, Zag eb, C oa ia 9Uni e si y o Zag eb School o Medicine, Zag eb, C oa ia 10Clinic o Rheuma ology, Depa men o Rheuma ology, Medical Uni e si y o So ia, So ia, Bulga ia 11School o Medicine, Depa men o De ma ology, Koç Uni e si y, Is anbul, Tu key 12Clinic o Rheuma ology and Immunology, Bad B amds ed Hospi al, Bad B ams ed , Ge many 13Di ision o Rheuma ology and Sys emic In lamma o y Diseases, Uni e si y Medical Cen e Hambu g- Eppendo , Hambu g, Ge many P o ec ed by copy igh . on June 27, 2023 a Facul y o Medicine, Uni e si y o Lisbon.h p://a d.bmj.com/Ann Rheum Dis: i s published as 10.1136/a d-2022-222852 on 27 Ma ch 2023. Downloaded om 893 Kuipe JJW, e al. Ann Rheum Dis 2023;82:887–896. doi:10.1136/ann heumdis-2022-222852 Re iew 14Uni e si y o Ams e dam, Depa men o Rheuma ology & Clinical Immunology and Depa men o Expe imen al Immunology, Ams e dam Ins i u e o In ec ion & Immuni y, Ams e dam UMC Loca ion AMC, Ams e dam, The Ne he lands 15Ams e dam Rheuma ology and Immunology Cen e (ARC) | Ams e dam UMC, Uni e si y o Ams e dam, Ams e dam, The Ne he lands 16Depa men o De ma ology, Anka a Uni e si y Facul y o Medicine, Anka a, Tu key 17Depa men o Rheuma ology and Clinical Immunology, Uni e si y o C e e School o Medicine, I aklio, G eece 18Labo a o y o Au oimmuni y- In lamma ion, Ins i u e o Molecula Biology and Bio echnology, He aklion, G eece 19Ams e dam Rheuma ology and immunology Cen e (ARC)| Reade, Ams e dam, The Ne he lands 20Depa men o Rheuma ology, Reade Hoo dloca ie D Jan an B eemens aa , Ams e dam, The Ne he lands 21Depa men o Oph halmology, Hospi al de San a Ma ia, Cen o Hospi ala Uni e si á io Lisboa No e EPE, Lisboa, Po ugal 22Cen o de Es udeos das Ciencias da Visão, Uni e sidade de Lisboa Faculdade de Medicina, Lisboa, Po ugal 23Uni e si y Eye Clinic, Uni e si y Medical Cen e Ljubljana, Ljubljana, Slo enia 24Facul y o medicine, Uni e si y o Ljubljana, Ljubljana, Slo enia 25Depa men o Oph halmology, Is anbul Uni e si y Is anbul Facul y o Medicine, Is anbul, Tu key 26Depa emen o Oph halmology, Ams e dam UMC Loca ion AMC, Ams e dam, The Ne he lands 27Depa men o Oph halmology, Onze Lie e V ouwe Gas huis, Ams e dam, The Ne he lands 28Se ice de Rheuma ology, Hospi al Amb oise- Pa e, Boulogne- Billancou , F ance 29In ec ion & In lamma ion, UMR 1173, Inse m, UVSQ, Uni e si y Pa is- Saclay, Mon igny- le- B e onneux, F ance 30In ec ion & In lamma ion, UMR 1173, Inse m, UVSQ/Uni e si é Pa is Saclay, Mon igny- le- B e onneux, F ance 31Labo a oi e d’Excellence In lamex, Pa is, F ance 32Depa men o Oph halmology, Uni e si y Hospi al Cen e Zag eb, Zag eb, C oa ia 33Depa men o Rheuma ology and Immunology, Inselspi al Uni e si y Hospi al Be n, Uni e si y o Be n, Be n, Swi ze land 34Rheuma ology Resea ch Uni Molecula João Lobo An unes, Uni e si y o Lisbon Medical Facul y, Lisboa, Po ugal 35Rheuma ology Depa men San a Ma ia Cen o Hospi al, Academic Medical Cen e o Lisbon, Lisboa, Po ugal 36Pa ien pa ne , Ge man League agains Rheuma ism, Bonn, Ge many 37Pa ien pa ne , Na ionale Ve eniging ReumaZo g, Nijmegen, The Ne he lands 38T ansla ional Resea ch Ins i u e, Queensland Uni e si y o Technology, B isbane, Queensland, Aus alia 39School o Medicine, Uni e si y o S And ews School o Medicine, S And ews, UK 40Di ision o Rheuma ology, Is anbul Uni e si y Is anbul Facul y o Medicine, Is anbul, Tu key 41Cen e o Gene ics and Genomics Ve sus A h i is, Cen e o Musculoskele al Resea ch, Facul y o Biology, Medicine and Heal h, Manches e Academic Heal h Science Cen e , The Uni e si y o Manches e , Manches e , UK 42NIHR Manches e Musculoskele al Biomedical Resea ch Uni , Manches e Uni e si y NHS Founda ion T us , Manches e , UK 43Depa men o Rheuma ology, KU Leu en Uni e si y Hospi als Leu en, Leu en, Belgium 44Skele al Biology and Enginee ing Resea ch Cen e , Depa men o De elopmen and Regene a ion, KU Leu en, Leu en, Belgium 45Depa men o Medicine, Di ision o Rheuma ology, NYU Langone Behçe ’s Disease P og am, NYU Langone Ocula Rheuma ology P og am, New Yo k Uni e si y G ossman School o Medicine, New Yo k Uni e si y, New Yo k, New Yo k, USA 46Depa men o Pa hology, New Yo k Uni e si y G ossman School o Medicine, New Yo k, New Yo k, USA 47Depa men Immunopa hology, Sanquin Resea ch, Ams e dam, The Ne he lands 48Leeds Ins i u e o Rheuma ic and Musculoskele al Medicine, Uni e si y o Leeds, Leeds, UK 49NIHR Leeds Biomedical Resea ch Cen e, Leeds Teaching Hospi als NHS T us , Leeds, UK Twi e Jonas JW Kuipe @jonas_kuipe and Tony J Kenna @ onykenna3 Con ibu o s The inal e sion was app o ed by all au ho s. Funding The au ho s ha e no decla ed a speci ic g an o his esea ch om any unding agency in he public, comme cial o no - o - p o i sec o s. Compe ing in e es s JCP: G an s o con ac s om any en i y Ge man Resea ch Founda ion g an s PR 241/5- 2Consul ing ees Boeh inge IngelheimPaymen o hono a ia o lec u es, p esen a ions, speake s bu eaus, manusc ip w i ing o educa ional e en s paid ac i i ies as a speake o Almi all, Boeh inge Ingelheim, Janssen- Cilag, No a is and P ize PK Leade ship o iducia y ole in o he boa d, socie y, commi ee o ad ocacy g oup, paid o unpaid All unpaid: Human Immunology (Else ie ) Edi o ial Boa d, F on ie s in Immunology Gues associa e, Edi o and e iew edi o . Edi o ial boa d o In e na ional Jou nal o Immunogene ics SS: all suppo o he p esen manusc ip (eg, unding, p o ision o s udy ma e ials, medical w i ing, a icle p ocessing cha ges, e c. Deu sche Fo schungsgemeinscha (DFG) o Jacobs Uni e si y B emen DM G an s o con ac s om any en i y Go e nmen o Mal aPaymen o hono a ia o lec u es, p esen a ions, speake s bu eaus, manusc ip w i ing o educa ional e en s Sano i, U iageSuppo o a ending mee ings and/o a el A ene, Biode ma, U iageLeade ship o iducia y ole in o he boa d, socie y, commi ee o ad ocacy g oup, paid o unpaid Mal ese Associa ion o De ma ologis s and Vene eologis sI an IP Paymen o hono a ia o lec u es, p esen a ions, speake s bu eaus, manusc ip w i ing o educa ional e en s No a is, Eli Lilly, P ize , Abb ie, Hono a ia o lec u es, paymen s di ec ly o me RS Suppo o a ending mee ings and/o a el Abb ie and P ize Ina IK Consul ing ees Amgen, Boeh inge ,GSK, SobiPaymen o hono a ia o lec u es, p esen a ions, speake s bu eaus, manusc ip w i ing o educa ional e en s Abb ie, Amgen. Boeh inge , GSK, Janssen, Lilly, MSD, No a is, P ize , Sobi M dS G an s o con ac s om any en i y No a is, UCB, EliLillyPaymen o hono a ia o lec u es, p esen a ions, speake s bu eaus, manusc ip w i ing o educa ional e en s UCBSuppo o a ending mee ings and/o a el UCBPa icipa ion on a Da a Sa e y Moni o ing Boa d o Ad iso y Boa d No a is, UCB, Abb ie GB G an s o con ac s om any en i y GSK, P ize Paymen o hono a ia o lec u es, p esen a ions, speake s bu eaus, manusc ip w i ing o educa ional e en s GSK, As aZeneca, P ize , Abb ie, Aeno asis, No a is, Lilly IL Consul ing ees No a is, Alime a lT- T Consul ing ees AbbVie, Tu key, No a isPaymen o hono a ia o lec u es, p esen a ions, speake s bu eaus, manusc ip w i ing o educa ional e en s AbbVie, Tu key FC Paymen o hono a ia o lec u es, p esen a ions, speake s bu eaus, manusc ip w i ing o educa ional e en s LillySuppo o a ending mee ings and/o a el UCB, No a isPa icipa ion on a da a sa e y moni o ing boa d o Ad iso y Boa d UCB, No a is DMZ G an s o con ac s om any en i y Eu opean Socie y o Oph halmologyPaymen o hono a ia o lec u es, p esen a ions, speake s bu eaus, manusc ip w i ing o educa ional e en s Zen i a, Alkaloid d.o.o., Inspha ma d.o.o. MB Paymen o hono a ia o lec u es, p esen a ions, speake s bu eaus, manusc ip w i ing o educa ional e en s P ize , Via is, Lilly, MSD TJK All suppo o he p esen manusc ip (eg, unding, p o ision o s udy ma e ials, medical w i ing, a icle p ocessing cha ges, e c). Na ional Heal h & Medical Resea ch Council GNT2011115Leade ship o iducia y ole in o he boa d, socie y, commi ee o ad ocacy g oup, paid o unpaid P esiden , Aus alian Socie y o Medical Resea ch MB G an s o con ac s om any en i y PFIZERConsul ing ees PFIZERPaymen o hono a ia o lec u es, p esen a ions, speake s bu eaus, manusc ip w i ing o educa ional e en s FRENESIUS KABI, LILLYSuppo o a ending mee ings and/o a el BIOGEN, PFIZER, JANSSEN RJUL Consul ing ees UCB, No a is, Abb ie, Eli- LillyPaymen o hono a ia o lec u es, p esen a ions, speake s bu eaus, manusc ip w i ing o educa ional e en s UCB, No a is, Abb ie, Eli- Lilly, Amgen JN All esea ch unding suppo o he p esen manusc ip : NEI- NIH R01EY033495— esea ch unds and R01EY031383— esea ch undsHono a ia o lec u es: Ha a d Uni e si y, No hwes e n Uni e si y, Massachuse s Gene al Hospi al.Suppo o a ending mee ings and/o a el: NYU Depa men o Medicine, NIH- NEIPa icipa ion Medical Ad iso y Boa d: ABDA (Ame ican Behçe ’s Disease Associa ion) DGMcG Paymen o hono a ia o lec u es, p esen a ions, speake s bu eaus, manusc ip w i ing o educa ional e en s Janssen, Abb ie, No a is, UCB, BMS, Lilly. Pa ien consen o publica ion No applicable. P o enance and pee e iew No commissioned; ex e nally pee e iewed. Supplemen al ma e ial This con en has been supplied by he au ho (s). I has no been e ed by BMJ Publishing G oup Limi ed (BMJ) and may no ha e been pee - e iewed. Any opinions o ecommenda ions discussed a e solely hose o he au ho (s) and a e no endo sed by BMJ. BMJ disclaims all liabili y and esponsibili y a ising om any eliance placed on he con en . Whe e he con en includes any ansla ed ma e ial, BMJ does no wa an he accu acy and eliabili y o he ansla ions (including bu no limi ed o local egula ions, clinical guidelines, e minology, d ug names and d ug dosages), and is no esponsible o any e o and/o omissions a ising om ansla ion and adap a ion o o he wise. Open access This is an open access a icle dis ibu ed in acco dance wi h he C ea i e Commons A ibu ion Non Comme cial (CC BY- NC 4.0) license, which pe mi s o he s o dis ibu e, emix, adap , build upon his wo k non- comme cially, and license hei de i a i e wo ks on di e en e ms, p o ided he o iginal wo k is p ope ly ci ed, app op ia e c edi is gi en, any changes made indica ed, and he use is non- comme cial. See:h p://c ea i ecommons.o g/licenses/by-nc/4.0/. ORCID iDs Jonas JWKuipe h p://o cid.o g/0000-0002-5370-6395 DillonMin o h p://o cid.o g/0000-0003-3705-0119 Cha lo e LMK ieckae h p://o cid.o g/0000-0001-6591-7838 ElsaVie a Sousa h p://o cid.o g/0000-0002-7170-8802 MaximeB eban h p://o cid.o g/0000-0002-6932-9395 JohnBowes h p://o cid.o g/0000-0003-4659-031X Rik JULo ies h p://o cid.o g/0000-0002-5986-3092 F ank ienTu ks a h p://o cid.o g/0000-0002-8055-9036 P o ec ed by copy igh . on June 27, 2023 a Facul y o Medicine, Uni e si y o Lisbon.h p://a d.bmj.com/Ann Rheum Dis: i s published as 10.1136/a d-2022-222852 on 27 Ma ch 2023. Downloaded om 894 Kuipe JJW, e al. Ann Rheum Dis 2023;82:887–896. doi:10.1136/ann heumdis-2022-222852 Re iew REFERENCES 1 Medzhi o R. The spec um o in lamma o y esponses. Science 2021;374:1070–5. 2 Szekanecz Z, McInnes IB, Sche G, e al. Au oin lamma ion and au oimmuni y ac oss heuma ic and musculoskele al diseases. Na Re Rheuma ol 2021;17:585–95. 3 McGonagle D, McDe mo MF. A p oposed classi ica ion o he immunological diseases. PLoS Med 2006;3:e297. 4 Ki ino Y, Be sias G, Ishiga subo Y, e al. Genome- wide associa ion analysis iden i ies new suscep ibili y loci o Behçe ’s disease and epis asis be ween HLA- B*51 and ERAP1. Na Gene 2013;45:202–7. 5 Takeuchi M, Mizuki N, Megu o A, e al. Dense geno yping o immune- ela ed loci implica es hos esponses o mic obial exposu e in Behçe ’s disease suscep ibili y. Na Gene 2017;49:438–43. 6 S ange A, Capon F, Spence CCA, e al. A genome- wide associa ion s udy iden i ies new pso iasis suscep ibili y loci and an in e ac ion be ween HLA- C and ERAP1. Na Gene 2010;42:985–90. 7 Bowes J, Ashc o J, Dand N, e al. C oss- pheno ype associa ion mapping o he MHC iden i ies gene ic a ian s ha di e en ia e pso ia ic a h i is om pso iasis. Ann Rheum Dis 2017;76:1774–9. 8 S ua PE, Nai RP, Tsoi LC, e al. Genome- wide associa ion analysis o pso ia ic a h i is and cu aneous pso iasis e eals di e ences in hei gene ic a chi ec u e. Am J Hum Gene 2015;97:816–36. 9 Bu on PR, Clay on DG, Ca don LR, e al. Associa ion scan o 14,500 nonsynonymous snps in ou diseases iden i ies au oimmuni y a ian s. Na Gene 2007;39:1329–37. 10 Re eille JD, Sims A- M, Danoy P, e al. Genome- wide associa ion s udy o ankylosing spondyli is iden i ies non- MHC suscep ibili y loci. Na Gene 2010;42:123–7. 11 E ans DM, Spence CCA, Poin on JJ, e al. In e ac ion be ween ERAP1 and HLA- B27 in ankylosing spondyli is implica es pep ide handling in he mechanism o HLA- B27 in disease suscep ibili y. Na Gene 2011;43:761–7. 12 Huang X- F, Li Z, De Guzman E, e al. Genomewide associa ion s udy o acu e an e io u ei is iden i ies new suscep ibili y loci. In es Oph halmol Vis Sci 2020;61:3. 13 Kuipe JJW, Van Se en J, Ripke S, e al. A genome- wide associa ion s udy iden i ies a unc ional ERAP2 haplo ype associa ed wi h bi dsho cho io e inopa hy. Hum Mol Gene 2014;23:6081–7. 14 Gel man S, Monne D, Ligocki AJ, e al. ERAP1, ERAP2, and wo copies o HLA- aw19 alleles inc ease he isk o bi dsho cho io e inopa hy in HLA- A29 ca ie s. In es Oph halmol Vis Sci 2021;62:3. 15 McGonagle D, Aydin SZ, Gül A, e al. MHC- I- opa hy’-uni ied concep o spondyloa h i is and behçe disease. Na Re Rheuma ol 2015;11:731–40. 16 McGonagle D, Aydın S, Gül A, e al. Reply o: behçe ’s disease: an MHC- I- opa hy? unde ined. 2017. A ailable: h ps://www.seman icschola .o g/pape /Reply- o%3A- Beh%C3%A7e %27s-disease%3A-an-MHC-I-opa hy-McGonagle-Ayd%C4%B1n/ 5840d4ead0babea3 db727 b06b271c7c9 7993 [Accessed 14 Oc 2022]. 17 Giza M, Ko o i D, Chen L, e al. Is behçe ’s disease a “class 1- opa hy”? The ole o HLA- B*51 in he pa hogenesis o behçe ’s disease. Clin Exp Immunol 2018;191:11–8. 18 Ha emi G, Ka a emiz G, Yazici H. Behçe ’s disease: an MHC- I- opa hy? Clin Exp Rheuma ol 2017;35 Suppl 104:5. 19 Rock KL, Rei s E, Nee jes J. P esen you sel ! by MHC class I and MHC class II molecules. T ends Immunol 2016;37:724–37. 20 Yewdell JW, Hollý J. DRiPs ge molecula . Cu Opin Immunol 2020;64:130–6. 21 Rock KL, Yo k IA, Goldbe g AL. Pos - p o easomal an igen p ocessing o majo his ocompa ibili y complex class I p esen a ion. Na Immunol 2004;5:670–7. 22 Blees A, Januliene D, Ho mann T, e al. S uc u e o he human MHC- I pep ide- loading complex. Na u e 2017;551:525–8. 23 Admon A, Bassani- S e nbe g M. The human immunopep idome p ojec , a sugges ion o ye ano he pos genome nex big hing. Mol Cell P o eomics 2011;10:111. 24 Mbi ibindi B, Mukhe jee S, Welling on D, e al. Spa ial clus e ing o ecep o s and signaling molecules egula es NK cell esponse o pep ide epe oi e changes. F on Immunol 2019;10:605. 25 Gill T, Rosenbaum JT. Pu a i e pa hobion s in HLA- B27- associa ed spondyloa h opa hy. F on Immunol 2020;11:586494. 26 Ga ido- Mesa J, B own MA. T cell epe oi e p o iling and he mechanism by which HLA- B27 causes ankylosing spondyli is. Cu Rheuma ol Rep 2022;24:398–410. 27 Mau o D, Thomas R, Guggino G, e al. Ankylosing spondyli is: an au oimmune o au oin lamma o y disease? Na Re Rheuma ol 2021;17:387–404. 28 López de Cas o JA. How ERAP1 and ERAP2 shape he pep idomes o disease- associa ed MHC- I p o eins. F on Immunol 2018;9:2463. 29 Yang X, Ga ne LI, Z yagin IV, e al. Au oimmuni y- associa ed T cell ecep o s ecognize HLA- B*27- bound pep ides. Na u e 2022;612:771–7. 30 A akawa A, Ree es E, Vollme S, e al. E ap1 con ols he au oimmune esponse agains melanocy es in pso iasis by gene a ing he melanocy e au oan igen and egula ing i s amoun o HLA- C*06:02 p esen a ion. J Immunol 2021;207:2235–44. 31 A akawa A, Siewe K, S öh J, e al. Melanocy e an igen igge s au oimmuni y in human pso iasis. J Exp Med 2015;212:2203–12. 32 Boni acio KM, Kunj a ia N, K uege JG, e al. Cu aneous exp ession o a disin eg in- like and me allop o ease domain con aining h ombospondin ype 1 mo i - like 5 (ADAMTSL5) in pso iasis goes beyond melanocy es. J Pigmen Diso d 2016;3:244. 33 Loll B, Rücke C, Uchanska- Ziegle B, e al. Con o ma ional plas ici y o HLA- B27 molecules co ela es in e sely wi h e iciency o nega i e T cell selec ion. F on Immunol 2020;11:179. 34 Blanco- Gelaz MA, Suá ez- Al a ez B, Díaz- Peña R, e al. Hla- B27 polymo phism a posi ion 116 c i ically in luences he associa ion wi h TAP/ apasin, in acellula a icking and con o ma ional homodime s o ma ion. Mol Immunol 2009;46:1304–11. 35 Abual ous ET, F i zsche S, Hein Z, e al. F pocke lexibili y in luences he apasin dependence o wo di e en ially disease- associa ed MHC class I p o eins. Eu J Immunol 2015;45:1248–57. 36 Hein Z, Uch enhagen H, Abual ous ET, e al. Pep ide- independen s abiliza ion o MHC class I molecules b eaches cellula quali y con ol. J Cell Sci 2014;127:2885–97. 37 G anados DP, Tanguay P- L, Ha dy M- P, e al. E s ess a ec s p ocessing o MHC class I- associa ed pep ides. BMC Immunol 2009;10:10. 38 de Almeida SF, Fleming JV, Aze edo JE, e al. S imula ion o an un olded p o ein esponse impai s MHC class I exp ession. J Immunol 2007;178:3612–9. 39 Tu ne MJ, Sowde s DP, DeLay ML, e al. HLA- B27 mis olding in ansgenic a s is associa ed wi h ac i a ion o he un olded p o ein esponse. J Immunol 2005;175:2438–48. 40 Robinson PC, Lau E, Kei h P, e al. ERAP2 unc ional knockou in humans does no al e su ace hea y chains o HLA- B27, in lamma o y cy okines o endoplasmic e iculum s ess ma ke s. Ann Rheum Dis 2015;74:2092–5. 41 Kenna TJ, Lau MC, Kei h P, e al. Disease- associa ed polymo phisms in ERAP1 do no al e endoplasmic e iculum s ess in pa ien s wi h ankylosing spondyli is. Genes Immun 2015;16:35–42. 42 Amba us CA, Ye emenko N, Bae en DL. Al e ed cy okine exp ession by mac ophages om HLA- B27- posi i e spondyloa h i is pa ien s wi hou e idence o endoplasmic e iculum s ess. Rheuma ol Ad P ac 2018;2: ky014. 43 Campbell EC, Fe ke F, Bha S, e al. Exp ession o MHC class I dime s and ERAP1 in an ankylosing spondyli is pa ien coho . Immunology 2011;133:379–85. 44 Díaz- Peña R, Vidal- Cas iñei a JR, Mule o J, e al. Ac i a ing kille immunoglobulin- like ecep o s genes a e associa ed wi h inc eased suscep ibili y o ankylosing spondyli is. Clin Exp Immunol 2015;180:201–6. 45 Cauli A, Shaw J, Giles J, e al. The a h i is- associa ed HLA- B*27:05 allele o ms mo e cell su ace B27 dime and ee hea y chain ligands o KIR3DL2 han HLA- B*27:09. Rheuma ology (Ox o d) 2013;52:1952–62. 46 Ci aldi L, Romania P, Falco M, e al. ERAP1 egula es na u al kille cell unc ion by con olling he engagemen o inhibi o y ecep o s. Cance Res 2015;75:824–34. 47 D’Amico S, D’Alicand o V, Compagnone M, e al. ERAP1 con ols he in e ac ion o he inhibi o y ecep o KIR3DL1 wi h HLA- B51:01 by a ec ing na u al kille cell unc ion. F on Immunol 2021;12:778103. 48 Ba on A, Hill J, Bibi S, e al. Gene ic suscep ibili y o en e ic e e in expe imen ally challenged human olun ee s. In ec Immun 2022;90:e0038921. 49 Manasson J, Blank RB, Sche JU. The mic obiome in heuma ology: whe e a e we and whe e should we go? Ann Rheum Dis 2020;79:727–33. 50 Kuipe JJW, Venema WJ. HLA- A29 and bi dsho u ei is: u he down he abbi hole. F on Immunol 2020;11:599558. 51 Pulido JS, Canal I, Salomão D, e al. His ological indings o bi dsho cho io e inopa hy in an eye wi h ciliocho oidal melanoma. Eye (Lond) 2012;26:862–5. 52 Pagnoux C, Mah A, Aouba A, e al. Ex aocula mani es a ions o bi dsho cho io e inopa hy in 118 F ench pa ien s. P esse Med 2010;39:e97–102. 53 Okada Y, Han B, Tsoi LC, e al. Fine mapping majo his ocompa ibili y complex associa ions in pso iasis and i s clinical sub ypes. Am J Hum Gene 2014;95:162–72. 54 Winches e R, Giles J, Jadon D, e al. Implica ions o he di e si y o class I HLA associa ions in pso ia ic a h i is. Clin Immunol 2016;172:29–33. 55 Fi zGe ald O, Ha oon M, Giles JT, e al. Concep s o pa hogenesis in pso ia ic a h i is: geno ype de e mines clinical pheno ype. A h i is Res The 2015;17:115. 56 Schwa zman M, E mann J, Kuhn KA, e al. Spondyloa h i is in in lamma o y bowel disease coho s: sys ema ic li e a u e e iew and c i ical app aisal o s udy designs. RMD Open 2022;8:e001777. 57 Can ini F, Nannini C, Cassa à E, e al. U ei is in spondyloa h i is: an o e iew. J Rheuma ol Suppl 2015;93:27–9. 58 Wang K, Gai sch H, Poon H, e al. Classi ica ion o common human diseases de i ed om sha ed gene ic and en i onmen al de e minan s. Na Gene 2017;49:1319–25. 59 Gu A, Sa ac AJ, Bu kan YK, e al. A h opa hy, quali y o li e, dep ession, and anxie y in behce ’s disease: ela ionship be ween a h i is and hese ac o s. Clin Rheuma ol 2006;25:524–31. 60 Poddubnyy D, Jadon DR, Van den Bosch F, e al. Axial in ol emen in pso ia ic a h i is: an upda e o heuma ologis s. Semin A h i is Rheum 2021;51:880–7. 61 Bena en D, Plasencia C, Poddubnyy D, e al. Un eiling axial in ol emen in pso ia ic a h i is: an ancilla y analysis o he ASAS- pe spa s udy. Semin A h i is Rheum 2021;51:766–74. P o ec ed by copy igh . on June 27, 2023 a Facul y o Medicine, Uni e si y o Lisbon.h p://a d.bmj.com/Ann Rheum Dis: i s published as 10.1136/a d-2022-222852 on 27 Ma ch 2023. Downloaded om 895 Kuipe JJW, e al. Ann Rheum Dis 2023;82:887–896. doi:10.1136/ann heumdis-2022-222852 Re iew 62 Kö e I, Lö sche F. Behçe ’s synd ome apa om he iple symp om complex: ascula , neu ologic, gas oin es inal, and musculoskele al mani es a ions. A mini e iew. F on Med (Lausanne) 2021;8:639758. 63 P inz JC. Human leukocy e an igen- class I alleles and he au o eac i e T cell esponse in pso iasis pa hogenesis. F on Immunol 2018;9:954. 64 Maldini C, La alley MP, Cheminan M, e al. Rela ionships o HLA- B51 o b5 geno ype wi h Behce ’s disease clinical cha ac e is ics: sys ema ic e iew and me a- analyses o obse a ional s udies. Rheuma ology (Ox o d) 2012;51:887–900. 65 A é alo M, G a acós Masmi jà J, Mo eno M, e al. In luence o HLA- B27 on he ankylosing spondyli is pheno ype: esul s om he REGISPONSER da abase. A h i is Res The 2018;20:221. 66 Zhang S, Wang Y, Peng L, e al. Compa ison o clinical ea u es in HLA- B27 posi i e and nega i e pa ien s wi h axial spondyloa h i is: esul s om a coho o 4,131 pa ien s. F on Med (Lausanne) 2020;7:609562. 67 Leccese P, Yazici Y, Oli ie i I. Behce ’s synd ome in nonendemic egions. Cu Opin Rheuma ol 2017;29:12–6. 68 Ske W, Hamil on MJ, A ayssi T. Gas oin es inal Behçe ’s disease: a e iew. Wo ld J Gas oen e ol 2015;21:3801–12. 69 Meie K, Schloegl A, Poddubnyy D, e al. Skin mani es a ions in spondyloa h i is. The Ad Musculoskele Dis 2020;12. 70 Ha oon M, O’Rou ke M, Ramasamy P, e al. A no el e idence- based de ec ion o undiagnosed spondyloa h i is in pa ien s p esen ing wi h acu e an e io u ei is: he due (Dublin u ei is e alua ion ool). Ann Rheum Dis 2015;74:1990–5. 71 Monne D, B eban M, Hud y C, e al. Oph halmic indings and equency o ex aocula mani es a ions in pa ien s wi h HLA- B27 u ei is: a s udy o 175 cases. Oph halmology 2004;111:802–9. 72 Chaiyabu C, Ungp ase P, Silpa- A cha N, e al. Pso iasis and isk o u ei is: a sys ema ic e iew and me a- analysis. Biomed Res In 2020;2020:9308341. 73 Hahn HJ, Kwak SG, Kim D- K, e al. Associa ion o behçe disease wi h pso iasis and pso ia ic a h i is. Sci Rep 2021;11:2531. 74 Lee YC, Jeong SJ, Eun Y- G, e al. Risk o au oimmune diseases in ecu en aph hous ulce pa ien s: a na ionwide popula ion s udy. O al Dis 2021;27:1443–50. 75 Abbood HM, Pa han E, Che uka a GP. The link be ween ankylosing spondyli is and o al heal h condi ions: wo nes ed case- con ol s udies using da a o he UK biobank. J Appl O al Sci 2018;27:e20180207. 76 Lin K- C, Tsai LL, Ko EC, e al. Como bidi y p o iles among pa ien s wi h ecu en aph hous s oma i is: a case- con ol s udy. J Fo mos Med Assoc 2019;118:664–70. 77 Robinson PN. Deep pheno yping o p ecision medicine. Hum Mu a 2012;33:777–80. 78 Tcheandjieu C, Agui e M, Gus a sson S, e al. A phenome- wide associa ion s udy o 26 Mendelian genes e eals pheno ypic exp essi i y o common and a e a ian s wi hin he gene al popula ion. PLoS Gene 2020;16:e1008802. 79 Schloss ein L, Te asaki PI, Blues one R, e al. High associa ion o an HL- A an igen, W27, wi h ankylosing spondyli is. N Engl J Med 1973;288:704–6. 80 B ewe on DA, Ha FD, Nicholls A, e al. Ankylosing spondyli is and HL- A 27. Lance 1973;1:904–7. 81 Ohno S, Aoki K, Sugiu a S, e al. Le e : HL- A5 and Behçe ’s disease. Lance 1973;2:1383–4. 82 Tiilikainen A, Lassus A, Pi skanen R, e al. An a emp o e alua e ia ype an igens in pa ien s wi h pso iasis o myas henia g a is. Tissue An igens, 1977: 10. 83 Tsuji K, Nose Y, Hoshino K. Fu he s udy on HLA- C, -D and ia an igens in pso iasis ulga is. Tissue An igens, 1977: 10. 84 Nussenbla RB, Mi al KK, Ryan S, e al. Bi dsho e inocho oidopa hy associa ed wi h HLA- A29 an igen and immune esponsi eness o e inal S- an igen. Am J Oph halmol 1982;94:147–58. 85 Co es A, Puli SL, Leo PJ, e al. Majo his ocompa ibili y complex associa ions o ankylosing spondyli is a e complex and in ol e u he epis asis wi h ERAP1. Na Commun 2015;6:7146. 86 Wang G, Kim T- H, Li Z, e al. MHC associa ions o ankylosing spondyli is in eas asians a e complex and in ol e non- HLA- B27 HLA con ibu ions. A h i is Res The 2020;22:74. 87 Omb ello MJ, Ki ino Y, de Bakke PIW, e al. Behçe disease- associa ed MHC class I esidues implica e an igen binding and egula ion o cell- media ed cy o oxici y. P oc Na l Acad Sci U S A 2014;111:8867–72. 88 Bowes J, Budu- Agg ey A, Hu meie U, e al. Dense geno yping o immune- ela ed suscep ibili y loci e eals new insigh s in o he gene ics o pso ia ic a h i is. Na Commun 2015;6:6046. 89 Su G, Zhong Z, Zhou Q, e al. Iden i ica ion o no el isk loci o behçe ’s disease- ela ed u ei is in a chinese popula ion in a genome- wide associa ion s udy. A h i is Rheuma ol 2022;74:671–81. 90 Robinson PC, Claushuis TAM, Co es A, e al. Gene ic dissec ion o acu e an e io u ei is e eals simila i ies and di e ences in associa ions obse ed wi h ankylosing spondyli is. A h i is Rheuma ol 2015;67:140–51. 91 Sakaue S, Kanai M, Tanigawa Y, e al. A c oss- popula ion a las o gene ic associa ions o 220 human pheno ypes. Na Gene 2021;53:1415–24. 92 Sun L- D, Cheng H, Wang Z- X, e al. Associa ion analyses iden i y six new pso iasis suscep ibili y loci in he Chinese popula ion. Na Gene 2010;42:1005–9. 93 Kuipe JJW, Se en J an, De all M, e al. Func ionally dis inc ERAP1 and ERAP2 a e a hallma k o HLA- A29- (bi dsho ) u ei is. Hum Mol Gene 2018;27:4333–43. 94 Robinson PC, Cos ello M- E, Leo P, e al. ERAP2 is associa ed wi h ankylosing spondyli is in HLA- B27- posi i e and HLA- B27- nega i e pa ien s. Ann Rheum Dis 2015;74:1627–9. 95 Tang H, Jin X, Li Y, e al. A la ge- scale sc een o coding a ian s p edisposing o pso iasis. Na Gene 2014;46:45–50. 96 Ha oon N, Inman RD. Endoplasmic e iculum aminopep idases: biology and pa hogenic po en ial. Na Re Rheuma ol 2010;6:461–7. 97 Weime shaus M, E nouchidou I, Sa eanu L, e al. Pep idases imming MHC class I ligands. Cu Opin Immunol 2013;25:90–6. 98 Mpakali A, Maben Z, S e n LJ, e al. Molecula pa hways o an igenic pep ide gene a ion by ER aminopep idase 1. Mol Immunol 2019;113:50–7. 99 Tedeschi V, Paldino G, Paladini F. The impac o he “mis- pep idome” on HLA class I- media ed diseases: con ibu ion o ERAP1 and ERAP2 and e ec s on he immune esponse. In J Mol Sci 2020;21:9608. 100 Babaie F, Hosseinzadeh R, Eb azeh M, e al. The oles o ERAP1 and ERAP2 in au oimmuni y and cance immuni y: new insigh s and pe spec i e. Mol Immunol 2020;121:7–19. 101 de Cas o JAL, S a ikos E. In acellula an igen p ocessing by ERAP2: molecula mechanism and oles in heal h and disease. Hum Immunol 2019;80:310–7. 102 Hanson AL, Cuddihy T, Haynes K, e al. Gene ic a ian s in ERAP1 and ERAP2 associa ed wi h immune- media ed diseases in luence p o ein exp ession and he iso o m p o ile. A h i is Rheuma ol 2018;70:255–65. 103 Omb ello MJ, Kas ne DL, Remme s EF. Endoplasmic e iculum- associa ed amino- pep idase 1 and heuma ic disease: gene ics. Cu Opin Rheuma ol 2015;27:349–56. 104 Ree es E, Coleba ch- Bou n A, Ellio T, e al. Func ionally dis inc ERAP1 allo ype combina ions dis inguish indi iduals wi h ankylosing spondyli is. P oc Na l Acad Sci U S A 2014;111:17594–9. 105 Hu chinson JP, Tempone as I, Kuipe J, e al. Common allo ypes o ER aminopep idase 1 ha e subs a e- dependen and highly a iable enzyma ic p ope ies. J Biol Chem 2021;296:100443. 106 Cos an ino F, Talpin A, E nouchidou I, e al. ERAP1 gene exp ession is in luenced by nonsynonymous polymo phisms associa ed wi h p edisposi ion o spondyloa h i is. A h i is Rheuma ol 2015;67:1525–34. 107 Ca e s A, Kugle MC, Ozgule Y, e al. Behçe ’s disease isk- a ian HLA- B51/ERAP1- hap10 al e s human CD8 T cell immuni y. Ann Rheum Dis 2022;81:1603–11. 108 Robe s AR, Apple on LH, Co es A, e al. ERAP1 associa ion wi h ankylosing spondyli is is a ibu able o common geno ypes a he han a e haplo ype combina ions. P oc Na l Acad Sci U S A 2017;114:558–61. 109 Colonna M, Mo e a A, Vély F, e al. A high- esolu ion iew o NK- cell ecep o s: s uc u e and unc ion. Immunol Today 2000;21:428–31. 110 Lee N, Llano M, Ca e e o M, e al. HLA- E is a majo ligand o he na u al kille inhibi o y ecep o CD94/NKG2A. P oc Na l Acad Sci U S A 1998;95:5199–204. 111 Lampen MH, Hassan C, Sluij e M, e al. Al e na i e pep ide epe oi e o HLA- E e eals a binding mo i ha is s ikingly simila o HLA- A2. Mol Immunol 2013;53:126–31. 112 B aud V, Jones EY, McMichael A. The human majo his ocompa ibili y complex class Ib molecule HLA- E binds signal sequence- de i ed pep ides wi h p ima y ancho esidues a posi ions 2 and 9. Eu J Immunol 1997;27:1164–9. 113 Ci aldi L, Lo Monaco E, Fo loni M, e al. Na u al kille cells e icien ly ejec lymphoma silenced o he endoplasmic e iculum aminopep idase associa ed wi h an igen p ocessing. Cance Res 2011;71:1597–606. 114 Cauli A, Dessole G, Piga M, e al. Exp ession analysis o HLA- E and NKG2A and NKG2C ecep o s poin s a a ole o na u al kille unc ion in ankylosing spondyli is. RMD Open 2018;4:e000597. 115 Hanson AL, Vukce ic D, Leslie S, e al. Epis a ic in e ac ions be ween kille immunoglobulin- like ecep o s and human leukocy e an igen ligands a e associa ed wi h ankylosing spondyli is. PLoS Gene 2020;16:e1008906. 116 Venema WJ, Hiddingh S, de Boe JH, e al. ERAP2 inc eases he abundance o a pep ide submo i highly selec i e o he bi dsho u ei is- associa ed HLA- A29. F on Immunol 2021;12:634441. 117 López de Cas o JA, Al a ez- Na a o C, B i o A, e al. Molecula and pa hogenic e ec s o endoplasmic e iculum aminopep idases ERAP1 and ERAP2 in MHC- I- associa ed in lamma o y diso de s: owa ds a uni ying iew. Mol Immunol 2016;77:193–204. 118 Ruyssen- Wi and A, Luxembou ge C, Can ag el A, e al. Associa ion be ween IL23R and ERAP1 polymo phisms and sac oiliac o spinal MRI in lamma ion in spondyloa h i is: DESIR coho da a. A h i is Res The 2019;21:22. 119 Nossen JC, Johnsen S, Bakland G. The in luence o ERAP1 gene a ian s on clinical pheno ype in ankylosing spondyli is. Scand J Rheuma ol 2016;45:474–9. 120 O iz- Fe nández L, Ca mona F- D, Mon es- Cano M- A, e al. Gene ic analysis wi h he immunochip pla o m in Behçe disease. Iden i ica ion o esidues associa ed in he HLA class I egion and new suscep ibili y loci. PLoS One 2016;11:e0161305. 121 Nikamo P, Lysell J, S åhle M. Associa ion wi h gene ic a ian s in he IL- 23 and NF- κB pa hways disc imina es be ween mild and se e e pso iasis skin disease. J In es De ma ol 2015;135:1969–76. P o ec ed by copy igh . on June 27, 2023 a Facul y o Medicine, Uni e si y o Lisbon.h p://a d.bmj.com/Ann Rheum Dis: i s published as 10.1136/a d-2022-222852 on 27 Ma ch 2023. Downloaded om BS Be iol, A., P isco, D., & Emmi, G. (2020). Behçe : he synd ome. Rheuma ology (Ox o d, England), 59(Suppl 3), iii101–iii107. doi:10.1093/ heuma ology/kez626 Hahn, H. J., Kwak, S. G., Kim, D.-K., & Kim, J.-Y. (2021). Associa ion o Behçe disease wi h pso iasis and pso ia ic a h i is. Scien i ic Repo s, 11(1), 2531. doi:10.1038/s41598- 021-81972-4 Ka iyanna, P. T., Shah, P., Jaya angaiah, A., Chowdhu y, Y. S., & Laza o, D. (2021). Acu e co ona y synd ome in Behce ’s synd ome: A sys ema ic e iew. Eu opean Jou nal o Rheuma ology, 8(1), 31–35. doi:10.5152/eu j heum.2020.19213 Seyahi, E., Ugu lu, S., Cumali, R., Balci, H., Ozdemi , O., Melikoglu, M., … Yazici, H. (2008). A he oscle osis in behçe ’s synd ome. Semina s in A h i is and Rheuma ism, 38(1), 1–12. doi:10.1016/j.sema h i .2007.09.009 BU Pagnoux, C., Mah , A., Aouba, A., Bé ezné, A., Monne , D., Cohen, P., … Guille in, L. (2010). Ex aocula mani es a ions o bi dsho cho io e inopa hy in 118 F ench pa ien s. P esse Medicale (Pa is, F ance: 1983), 39(5), e97–e102. doi:10.1016/j.lpm.2009.12.005 BMJ Publishing G oup Limi ed (BMJ) disclaims all liabili y and esponsibili y a ising om any eliance Supplemen al ma e ial placed on his supplemen al ma e ial which has been supplied by he au ho (s) Ann Rheum Dis doi: 10.1136/a d-2022-222852–10.:10 2023;Ann Rheum Dis, e al. Kuipe JJW