The dysfunctional immune system in common variable immunodeficiency increases the susceptibility to gastric cancer

cells A icle The Dys unc ional Immune Sys em in Common Va iable Immunode iciency Inc eases he Suscep ibili y o Gas ic Cance I ene Gullo 1,2,3,4,†, Ca a ina Cos a 1,2, Susana L. Sil a 5,6,7,*,†, C is ina Fe ei a 5,6,7, Ad iana Mo a 5,6, Sa a P. Sil a 5,6,7, Rúben Dua e Fe ei a 5,6, Ped o Rosmaninho 5,6, Emília Fa ia 8, JoséTo es da Cos a 9, Ri a Câma a 10, Gilza Gonçal es 1,3,4, João San os-An unes 11 , Ca la Oli ei a 2,3,4, JoséC. Machado 2,3,4, Fá ima Ca nei o 1,2,3,4,*,‡ and Ana E. Sousa 5,6,‡ 1Depa men o Pa hology, Cen o Hospi ala Uni e si á io de São João (CHUSJ), 4200-319 Po o, Po ugal; [email p o ec ed] (I.G.); [email p o ec ed] (C.C.); [email p o ec ed] (G.G.) 2 Depa men o Pa hology, Facul y o Medicine o he Uni e si y o Po o (FMUP), 4200-319 Po o, Po ugal; [email p o ec ed] (C.O.); [email p o ec ed] (J.C.M.) 3 Ins i u e o Molecula Pa hology and Immunology, Uni e si y o Po o (Ipa imup), 4200-135 Po o, Po ugal 4Ins i u o de In es igação e Ino ação em Saúde (i3S), Uni e si y o Po o, 4200-135 Po o, Po ugal 5Ins i u o de Medicina Molecula João Lobo An unes, Faculdade de Medicina, Uni e sidade de Lisboa, 1649-028 Lisbon, Po ugal; [email p o ec ed] (C.F.); [email p o ec ed] (A.M.); [email p o ec ed] (S.P.S.); dua e e [email p o ec ed] (R.D.F.); ped o. [email p o ec ed] (P.R.); [email p o ec ed] (A.E.S.) 6Cen o de Imunode iciências P imá ias do Cen o Académico de Medicina de Lisboa, 1649-028 Lisbon, Po ugal 7Cen o Hospi ala Uni e si á io Lisboa No e, 1600-190 Lisbon, Po ugal 8Se iço de Imunoale gologia, Cen o Hospi ala Hospi ala e Uni e si á io de Coimb a, 3004-561 Coimb a, Po ugal; [email p o ec ed] 9Se iço de Imunoale gologia, Cen o Hospi ala Hospi ala Uni e si á io de São João (CHUSJ), 4200-319 Po o, Po ugal; [email p o ec ed] 10 Se iço de Imunoale gologia, Hospi al D Nélio Mendonça, 9000-177 Funchal, Po ugal; [email p o ec ed] 11 Depa men o Gas oen e ology, Cen o Hospi ala Uni e si á io de São João (CHUSJ), 4200-319 Po o, Po ugal; [email p o ec ed] *Co espondence: [email p o ec ed] (S.L.S.); [email p o ec ed] (F.C.) †These au ho s con ibu ed equally o his wo k. ‡These au ho s con ibu ed equally o his wo k. Recei ed: 26 May 2020; Accep ed: 17 June 2020; Published: 19 June 2020   Abs ac : Gas ic ca cinoma (GC) ep esen s he mos common cause o dea h in pa ien s wi h common a iable immunode iciency (CVID). Howe e , a limi ed numbe o cases ha e been cha ac e ised so a . In his s udy, we analysed he clinical ea u es, bac e ial/ i al in ec ions, de ailed mo phology and immune mic oen i onmen o nine CVID pa ien s wi h GC. The s udy o he immune mic oen i onmen included au oma ed digi al coun s o CD20+, CD4+, CD8+, FOXP3+, GATA3+and CD138+immune cells, as well as he e alua ion o PD-L1 exp ession. Twen y-one GCs om non-CVID pa ien s we e used as a con ol g oup. GC in CVID pa ien s was diagnosed mos ly a ea ly-s age (n=6/9; 66.7%) and a younge age (median-age: 43y), when compa ed o non-CVID pa ien s (p<0.001). GC pa hogenesis was closely ela ed o Helicobac e pylo i in ec ion (n=8/9; 88.9%), bu no o Eps ein-Ba i us (0.0%) o cy omegalo i us in ec ion (0.0%). Non-neoplas ic mucosa (non-NM) in CVID-pa ien s displayed p ominen lymphocy ic gas i is (100%) and a dys unc ional immune mic oen i onmen , cha ac e ised by highe a es o CD4+/CD8+/Foxp3+/GATA3+/PD-L1+ immune cells and he expec ed pauci y o CD20+B-lymphocy es and CD138+plasma cells, when Cells 2020,9, 1498; doi:10.3390/cells9061498 www.mdpi.com/jou nal/cells Cells 2020,9, 1498 2 o 21 compa ed o non-CVID pa ien s (p<0.05). Changes in he immune mic oen i onmen be ween non-NM and GC we e no equi alen in CVID and non-CVID pa ien s, e lec ing he ele ance o immune dys unc ion o gas ic ca cinogenesis and GC p og ession in he CVID popula ion. Keywo ds: gas ic cance ; common a iable immunode iciency; immune mic oen i onmen ; Helicobac e pylo i; lymphocy ic gas i is; immune dys unc ionali y; inbo n e o s o immuni y 1. In oduc ion Inbo n e o s o immuni y comp ise an expanding g oup o mo e han 400 diseases which ha e p o ided a ema kable unexplo ed con ex o esea ch [ 1 ]. Common a iable immunode iciency (CVID) is he mos common clinically ele an p ima y immunode iciency [ 2 ]. I is de ined by de ec s in B-cell di e en ia ion in o memo y B cells and Ig-sec e ing plasma cells [ 2 – 4 ], hough pheno ypic and unc ional abno mali ies ha e been inc easingly ecognised also in T cells [ 5 , 6 ] and inna e immuni y [7,8]. Inc easing e idence poin s o a con inuous spec um o immunological dis u bances, leading o an ongoing deba e on he CVID diagnos ic c i e ia [ 9 ]. Ex eme deple ion o he naï e CD4 compa men (naï e CD4 T cells below 10%) has been conside ed he mos sensi i e indica o o sugges some cellula immunode iciency in pa ien s wi h CVID, which is associa ed wi h highe p e alence o se e e non-in ec ious complica ions [ 10 ]. In addi ion o ecu en in ec ions p ima ily in he espi a o y and gas oin es inal ac s, he mos ypical p esen a ion in CVID, clinical mani es a ions ela ed o immune dys egula ion in hese pa ien s include au oimmune diso de s, a a iable spec um o lymphop oli e a i e diseases and malignancies [11,12]. Pa ien s wi h CVID bea a 10 o 47- old inc eased isk o de eloping cance in compa ison wi h he gene al popula ion [ 2 , 13 – 15 ]. Based on a ecen me a-analysis, he o e all p e alence o malignancy in CVID pa ien s is 8.6% [ 16 ], he mos common encompassing lymphoma (40.5%) and gas ic adenoca cinoma (7.5%), ollowed a a much lowe equency by b eas , skin, hymic cance , leukaemia and melanoma, among o he s [16]. Gas ic cance anks second among CVID- ela ed cance s, a e he expec ed lymphop oli e a i e diseases, and ep esen s he i s cause o cance - ela ed dea h in his popula ion [ 17 ]. In con as , acco ding o he mos ecen GLOBOCAN es ima es, gas ic cance is he i h mos equen ly diagnosed malignancy and he hi d leading cause o cance - ela ed dea h wo ldwide in he gene al popula ion [18]. In spi e o he high incidence o gas ic cance in CVID pa ien s, only 59 cases ha e been epo ed o da e in a me a-analysis o 48 s udies wi h a o al o 8123 CVID pa ien s [ 16 ], and de ailed clinicopa hological cha ac e isa ion is limi ed o solely wo case se ies [17,19]. Pa ien s wi h gas ic cance and CVID a e usually 15 yea s younge han hose wi hou CVID [ 20 ]. CVID-associa ed gas ic cance s a e equen ly in es inal- ype mode a ely o poo ly di e en ia ed adenoca cinomas wi h high numbe s o in a- umo al lymphocy es [ 19 ]. Malignancy gene ally de elops on a backg ound o se e e a ophic me aplas ic pangas i is [19,20]. The high incidence o gas ic adenoca cinoma in CVID pa ien s has been linked o gene ic p edisposi ion, pe sis en mucosal in lamma ion and dec eased clea ance o oncogenic i al and bac e ial in ec ions, including Helicobac e pylo i in ec ion [ 20 , 21 ]. Au oimmune p ocesses, as in he case o pe nicious anaemia, a e also known o inc ease he isk o gas ic cance in CVID, as epo ed in he gene al popula ion [ 22 ]. I is plausible ha impai men o umou cell immune su eillance may acili a e su i al and p oli e a ion o p e-malignan cells. Despi e he impo an in luence ha he immune mic oen i onmen may exe in CVID-associa ed gas ic cance pa hogenesis and p og ession, an in-dep h analysis o he immune landscape in his con ex is s ill missing. Mo eo e , pa ien s wi h Cells 2020,9, 1498 3 o 21 inbo n e o s o immuni y may p o ide a unique, no ye explo ed, con ex o deciphe he in e play o immunological and en i onmen al ac o s in gas ic adenoca cinoma. In his s udy, we pe o med a de ailed clinical, his opa hological and immunopheno ypic analysis o CVID-associa ed gas ic cance and backg ound non-neoplas ic mucosa, aiming a unco e ing biological ea u es ha may be associa ed wi h inc eased suscep ibili y o CVID pa ien s o gas ic cance . In pa icula , o elucida e mechanisms ha could media e gas ic cance pa hogenesis and p og ession in CVID pa ien s, we explo ed he pu a i e in luence o bac e ial and i al in ec ions as well as he quali y, densi y and spa ial/ empo al dis ibu ion o he immune mic oen i onmen . This ep esen s he i s a emp o p o ide a de ailed desc ip ion o he disease. 2. Ma e ials and Me hods 2.1. Pa ien Se ies Nine gas ic adenoca cinomas om o malin ixed pa a in-embedded (FFPE) gas ic cance samples o CVID pa ien s, including biopsies (n=2) and su gical specimens (n=7), we e ob ained om a coho o CVID pa ien s unde ollow-up a Cen o de Imunode ici ê ncias P im á ias (CIDP) o Cen o Acad é mico de Medicina de Lisboa (Lisbon, Po ugal) (n=6) and om o he hospi als belonging o he na ional ne wo k o CIDP (n=3). CVID was diagnosed based on ma ked dec ease o IgG (a leas wo s anda d de ia ions below he mean o age) and a ma ked dec ease in a leas one o he iso ypes IgM o IgA, impai ed esponse o accines and exclusion o seconda y causes o hypogammaglobulinemia [ 2 ]. Twen y-one gas ic cance FFPE samples om non-CVID pa ien s wi h ea ly-s age gas ic cance we e also ob ained om he Depa men s o Pa hology and o Gas oen e ology a Cen o Hospi ala Uni e si á io S ã o Jo ã o (CHUSJ) and used as con ol g oup. Clinical ea u es and medical eco ds, collec ed om he iles o CIDP and CHUSJ, we e analysed, including gende , age, ele an pe sonal and amily his o y and ea men o gas ic cance . Da a om s ool cul u es and PCR analysis ela i e o gas oin es inal in ec ions, ou inely pe o med in CVID pa ien s, we e also e ie ed. His opa hological cha ac e isa ion o gas ic adenoca cinomas and he backg ound non-neoplas ic mucosa, dis an om and adjacen o gas ic cance (when a ailable), was pe o med on haema oxylin and eosin (H&E)-s ained slides. Pa icula ly, in non-neoplas ic mucosa he p esence o lymphocy ic gas i is, a ophic gas i is wi h o wi hou neu ophilic ac i i y, lymphoid agg ega es and in es inal me aplasia (IM) was e alua ed. Mo eo e , he p esence o p ecu so lesions was speci ied. The umou his o ype was de ined acco ding o he classi ica ions o Lau é n [ 23 ] and he Wo ld Heal h O ganiza ion (WHO) (2019) [24]. Tumou s aging was e alua ed acco ding o he 2018 AJCC s aging sys em [25]. The s udy was conduc ed upon he app o al o he E hical Boa d o he Cen o Hospi ala Uni e si á io Lisboa No e (CHULN) and o he Faculdade de Medicina da Uni e sidade de Lisboa (FMUL). 2.2. Immunohis ochemis y (IHC) and Eps ein-Ba Vi us (EBV) In Si u Hyb idiza ion (ISH) Se ial 3- µ m sec ions we e p epa ed om one ep esen a i e FFPE block. IHC s aining was pe o med in all CVID and non-CVID gas ic cance cases wi h an ibodies agains CMV (clone CCH2 and DDG9, 1:1000; DAKO), CD20, (clone L26, p edilu ed; Ven ana Medical Sys ems), CD8 (clone SP57, p edilu ed; Ven ana Medical Sys ems), CD4 (clone SP35, p edilu ed; Ven ana Medical Sys ems), GATA3 (clone D13C9, 1:200; Cell Signaling Technology), Foxp3 (clone AB54501, 1:100; abcam), CD138/Syndecan-1 (clone B-A38, p edilu ed; Cell Ma que) and PD-L1 (clone 22C3, 1:80; DAKO). Samples we e p ocessed in he au oma ic Ven ana Benchma k Ul a pla o m using an Op i iew Uni e sal DAB De ec ion Ki and an Op i iew Ampli ica ion Ki o PD-L1 s aining. EBV in ec ion was s udied using ch omogenic ISH o EBV-encoded RNA (EBER-ISH, INFORM EBER p obe, Ven ana Cells 2020,9, 1498 4 o 21 Medical Sys ems) using he same equipmen , wi h enzyma ic diges ion (ISH p o ease) and an iViewBlue de ec ion ki . The de ailed p o ocols a e p esen ed in Table S1. 2.3. Digi al Image Analysis Full hickness consecu i e sec ions, s ained o H&E, CMV, CD20, CD4, CD8, Foxp3, GATA3, CD138/Syndecan-1 and PD-L1 we e scanned wi h a 40 × objec i e, using a Nanozoome S60 slide scanne (Hamama su). QuPa h Open sou ce so wa e (0.2.0-m9) was used o digi al pa hology image analysis [26]. Fo each case, a ep esen a i e ield a 200 × magni ica ion was selec ed in he dis an and adjacen non-neoplas ic mucosa (when a ailable), p ecu so lesions (when p esen ) and gas ic cance . A posi i e cell coun unc ion, which includes au oma ic segmen a ion o cell de ec ion and au oma ic classi ica ion o posi i i y was conduc ed in he lamina p op ia/s oma and in aepi helial compa men s, p ope ly sepa a ed a e manually ou lining he polygon egion o in e es . 2.4. E alua ion o PD-L1 Exp ession Two pa hologis s (IG, CC) independen ly e alua ed PD-L1 exp ession, and disco dan e alua ions we e discussed o each an ag eemen . PD-L1 immunoexp ession was assessed in dis an and adjacen non-neoplas ic mucosa, p ecu so lesions and gas ic cance . An o e all slide eyeball assessmen was conduc ed in he non-neoplas ic mucosa. The p esence o absence o PD-L1+lymphocy es was no ed. An es ima e o he pe cen age o PD-L1+immune cells in he lamina p op ia (<10% and ≥ 10%) was gi en. Fo he e alua ion o p ecu so lesions and adenoca cinomas, he combined posi i e sco e (CPS) was applied as p e iously desc ibed [27]. 2.5. S a is ical Analysis IBM SPSS ( elease 23.0.0) and G aphPad P ism 8 we e used o s a is ical analysis. All es s we e wo-sided, and di e ences we e conside ed signi ican when p<0.05. Compa isons o ca ego ical a iables we e pe o med using a Chi squa e es o Fishe ’s exac es , as app op ia e. Fo mul iple ca ego ical a iables, a pos -hoc es was used by applying Bon e oni co ec ion. Compa isons o quan i a i e a iables we e pe o med using non-pa ame ic es s, as he dis ibu ion o no mali y was a iable wi hin di e en immune bioma ke s. Acco dingly, Mann–Whi ney and Wilcoxon es s we e used, o independen and dependen a iables, espec i ely. 3. Resul s The se ies o pa ien s analysed in his s udy encompasses nine pa ien s diagnosed wi h CVID and wi h gas ic cance . Six o he nine CVID pa ien s belong o a coho o CVID pa ien s p e iously desc ibed [ 28 ] and we e diagnosed du ing a ho ough ollow-up a CIDP (Lisbon, Po ugal), which includes a egula esophagogas oduodenoscopy su eillance p o ocol [ 22 ]. Du ing he pe iod 2007–2020, he o al numbe o pa ien s wi h CVID ollowed a CIDP was 98, and he six pa ien s who de eloped gas ic cance accoun ed o 6.1% o he coho subjec s. O e all, gas ic cance was he mos common malignancy diagnosed in his CVID coho . The emaining h ee pa ien s we e ec ui ed om o he hospi als belonging o he na ional ne wo k o CIDP. 3.1. Clinicopa hological Fea u es o CVID Pa ien s wi h Gas ic Cance 3.1.1. Clinical Da a Fi e o he nine gas ic cance cases occu ed in women (55.6%). CVID and gas ic cance we e diagnosed a he median age o 21 yea s ( ange: 15–46) and 43 yea s ( ange: 27–62), espec i ely. The de elopmen o gas ic cance occu ed wi hin a median pe iod o 19.0 yea s ( ange: 6–38) om he CVID diagnosis. De ailed epidemiological and clinicopa hological ea u es o CVID pa ien s wi h gas ic cance a e p esen ed in Table S2. Cells 2020,9, 1498 5 o 21 Only one pa ien p esen ed amily his o y o p ima y immunode iciency diso de s, namely one b o he (pa ien 5) wi h selec i e IgA de iciency. This is he usual pa e n in CVID since he majo i y o he cases ha e a polygenic cause. Se en CVID pa ien s epo ed malignancy o he han gas ic cance in 15 ela i es, including b eas , colo ec al, p os a e, ce ical cance , enal cell ca cinoma and cen al ne ous sys em (CNS) malignancy (Table S2). A amily his o y o gas ic cance was p esen in wo pa ien s (pa ien 2 and 9), one (pa ien 2) being diagnosed wi h di use gas ic cance a he age o 43 yea s. This pa ien was submi ed o CDH1 and CTNNA1 gene ic es ing, which excluded ge mline causa i e a ian s. Two CVID pa ien s we e diagnosed wi h a second malignancy a e gas ic cance , namely ec al adenoca cinoma (pa ien 2) and hepa ocellula ca cinoma (pa ien 5). Six o he nine pa ien s (66.6%) su e ed om au oimmune diso de s, which included pe nicious anaemia (44.4%), au oimmune panc ea i is, heuma oid a h i is-like symp oms, immune h ombocy openia, alopecia and pso iasis. Au oimmune diso de s we e p esen in ou ela i es o h ee pa ien s ( ype 1 diabe es melli us, i iligo, Sjog en’s synd ome and au oimmune hy oid diseases). Fou o he nine pa ien s (44.4%) had g anuloma ous diseases, con i med using his opa hological analysis, namely g anuloma ous lymphocy ic in e s i ial lung disease (n=4), associa ed wi h g anulomas a ec ing skin and CNS in one pa ien . All pa ien s we e ecei ing IgG eplacemen he apy when gas ic cance was diagnosed. Sc eening o gas oin es inal in ec ion h oughou ollow-up e ealed he p esence o Campylobac e jejuni (n=5/9; 55.6%), Salmonella spp (n=3/9; 33.3%) and Gia dia lamblia (n=7/9; 77.8%) in ec ion, as well as gu i al in ec ions, including CMV (n=3/9; 33.3%), EBV (n=1/9; 11.1%) and No o i us (n=1/9; 11.1%). H. pylo i in ec ion was sea ched o in all cases h oughou ollow-up and eigh es ed posi i e (8/9; 88.9%) ia a u ea b ea h es and/o his opa hology in gas ic biopsy samples. In se en pa ien s, gas ec omy was he p ima y ea men ( o al (n=6) o sub o al dis al (n=1). Du ing he ollow-up pe iod, no umou ecu ence was de ec ed. Fou ou o he nine pa ien s died (44.4%) du ing he s udy ime. Two pa ien s su i ed less han one yea a e p esen ing wi h un esec able gas ic ca cinomas (also wi h hepa ocellula ca cinoma in one pa ien ) and wo pa ien s died om CVID- ela ed complica ions. 3.1.2. His opa hological Findings in Non-Neoplas ic Gas ic Mucosa Table S2 displays he de ailed his opa hological ea u es in non-neoplas ic mucosa o each CVID pa ien . Abundan mononuclea , p edominan ly lymphocy ic, in lamma o y in il a e in he lamina p op ia associa ed wi h glandula a ophy (ch onic a ophic gas i is) was iden i ied in all specimens o which he non-neoplas ic gas ic mucosa was a ailable (n=7/9). These pa ien s ea u ed a signi ican dec ease in ci cula ing o al lymphocy e coun s (median: 1240 cells/ µ L; CI 95%: 480–2220), as compa ed o heal hy indi iduals (median: 1900 cells/ µ L; CI 95%: 1700–2300; p=0.0301), sugges ing lymphocy ic a ic al e a ions wi h inc eased mucosal homing. In ag eemen wi h he diagnosis o CVID, a pauci y o plasma cells in he lamina p op ia was a ea u e in all cases, as e ealed using immunohis ochemis y. Non-neoplas ic mucosa o wo CVID pa ien s disclosed ew plasma cells (coun s pe 20 × powe ield: 22 (pa ien 4) and 40 (pa ien 8)). These wo CVID pa ien s ea u ed quan i iable equencies o ci cula ing plasma cells ( equency o CD38b igh IgMnega i e wi hin CD19+cells: 1.1% (pa ien 4) and 2.2% (pa ien 8)). Lymphocy ic gas i is, de ined by he p esence o a leas 25 in aepi helial lymphocy es pe 100 epi helial cells [ 29 ] was a cons an ea u e o he cases analysed (n=7/7; 100.0%). The majo i y o in aepi helial lymphocy es we e CD8+(median alue pe 20 × powe ield: 51; CI 95%: 11.1–303.0) and/o GATA3+(median alue pe 20 × powe ield: 30.0; CI 95%: 2.1–43.9) (Sec ion 3.3). Two cases showed exube an lymphocy ic gas i is, wi h an absolu e numbe o CD8+in aepi helial lymphocy es o 374 and 303 pe 20 × powe ield, espec i ely, in pa ien s 1 and 4. O no e, hese pa ien s ea u ed wi hin he lowes coun s o ci cula ing CD8 T cells (median coun : 405 cells/ µ L; CI 95%: 125–994; pa ien 1: 333 cells/µL; pa ien 4: 125 cells/µL). Cells 2020,9, 1498 6 o 21 Mul i ocal neu ophilic ac i i y was iden i ied in wo pa ien s (n=2/7; 28.6%). Ch onic a ophic gas i is wi h IM was iden i ied in all pa ien s (n=7/7; 100%), and in h ee samples, IM was ex ensi e, occupying almos he o ali y o he gas ic mucosa (n=3/7; 42.9%). Lymphoid ollicles we e iden i ied in ou pa ien s (n=4/7; 57.1%), bu ge minal cen es we e only iden i ied in one case (Figu e 1). Cells 2020, 9, x 6 o 22 Figu e 1. Non-neoplas ic gas ic mucosa displaying (a) lymphoid ollicle wi h ge minal cen e (Haema oxylin and Eosin, HE, 100×) (c) wi h CD20+ lymphocy es (IHC, 100×) and absence o CD138+ plasma cells (e) (IHC, 100×)—Pa ien 9. (b) Non-neoplas ic gas ic mucosa displaying a lymphoid ollicle wi hou ge minal cen e (HE, 100×) wi h (d) CD20+ lymphocy es and ( ) sca e ed CD138+ plasma cells a he pe iphe y (a ows)—Pa ien 4. 3.1.3. His opa hological Findings in Gas ic Cance Fi e o he nine (n = 5/9; 55.6%) gas ic cance cases we e localised in he gas ic an um and he emaining ou cases (n = 4/9; 44.4%) in he co pus. Pa hological s aging was a ailable in all gas ec omy specimens and mos pa ien s submi ed o gas ec omy (n = 7/9) we e diagnosed a ea ly s ages: pT1a (n = 4); pT1b (n = 2); pT2 (n = 1). No lymph node me as ases we e iden i ied in any gas ec omy specimen. The mo phology o gas ic cance de eloping in he con ex o CVID was he e ogenous (Figu e 2). Acco ding o he 2019 WHO Classi ica ion o Diges i e Sys em Tumou s [24], he majo i y o he cases we e ubula and/o papilla y adenoca cinomas (n = 6/9; 66.7%), bo h low-g ade (n = 3/6; 50.0%) and high-g ade (n = 3/6; 50%). The emaining cases included wo mucinous adenoca cinomas (n = 2/9; 22.2%) and one poo ly cohesi e ca cinoma (n = 1/9; 11.1%) o he non-signe ing cell ype (PCC- NOS), al hough sca e ed signe ing cells we e iden i ied. P ecu so adenoma ous lesions we e iden i ied in wo cases and consis ed o in es inal ype adenoma wi h low g ade dysplasia (Figu e 2). Figu e 1. Non-neoplas ic gas ic mucosa displaying ( a ) lymphoid ollicle wi h ge minal cen e (Haema oxylin and Eosin, HE, 100 × ) ( c ) wi h CD20+lymphocy es (IHC, 100 × ) and absence o CD138+ plasma cells ( e ) (IHC, 100 × )—Pa ien 9. ( b ) Non-neoplas ic gas ic mucosa displaying a lymphoid ollicle wi hou ge minal cen e (HE, 100 × ) wi h ( d ) CD20+lymphocy es and ( ) sca e ed CD138+ plasma cells a he pe iphe y (a ows)—Pa ien 4. 3.1.3. His opa hological Findings in Gas ic Cance Fi e o he nine (n=5/9; 55.6%) gas ic cance cases we e localised in he gas ic an um and he emaining ou cases (n=4/9; 44.4%) in he co pus. Pa hological s aging was a ailable in all gas ec omy specimens and mos pa ien s submi ed o gas ec omy (n=7/9) we e diagnosed a ea ly s ages: pT1a (n=4); pT1b (n=2); pT2 (n=1). No lymph node me as ases we e iden i ied in any gas ec omy specimen. The mo phology o gas ic cance de eloping in he con ex o CVID was he e ogenous (Figu e 2). Acco ding o he 2019 WHO Classi ica ion o Diges i e Sys em Tumou s [ 24 ], he majo i y o he cases we e ubula and/o papilla y adenoca cinomas (n=6/9; 66.7%), bo h low-g ade (n=3/6; 50.0%) and high-g ade (n=3/6; 50%). The emaining cases included wo mucinous adenoca cinomas (n=2/9; Cells 2020,9, 1498 7 o 21 22.2%) and one poo ly cohesi e ca cinoma (n=1/9; 11.1%) o he non-signe ing cell ype (PCC-NOS), al hough sca e ed signe ing cells we e iden i ied. P ecu so adenoma ous lesions we e iden i ied in wo cases and consis ed o in es inal ype adenoma wi h low g ade dysplasia (Figu e 2). Cells 2020, 9, x 7 o 22 Figu e 2. Mo phology o common a iable immunode iciency (CVID)-associa ed gas ic cance . (a) Tubula gas ic cance , low-g ade (HE, 100× magni ica ion); (b) ubula gas ic cance , high-g ade (HE, 100× magni ica ion); (c) mucinous adenoca cinoma (HE, 50× magni ica ion); (d) poo ly-cohesi e ca cinoma, PCC-NOS (HE, 200× magni ica ion). No e he p esence o sca e ed signe ing cells (a ows); (e, ) adenoma ous lesion, low-g ade dysplasia, wi h ubula (e) and illous ( ) a chi ec u e (HE, 50× magni ica ion). 3.1.4. His opa hological Fea u es Compa ed wi h Non-CVID Pa ien s To in es iga e in dep h he clinical ea u es o CVID pa ien s and he biological ele ance o he abo e desc ibed his opa hological indings, we compa ed ou esul s wi h hose ob ained om a con ol g oup o non-CVID pa ien s (n = 21) diagnosed wi h ea ly s age gas ic cance [pT1a (n = 15/21; 71.4%) and pT1b (n = 6/21; 28.6%)]. The esul s a e p esen ed in Table 1. Gas ic cance in CVID pa ien s was diagnosed a a younge age compa ed wi h non-CVID pa ien s (median age: 43y e sus 75y; Table 1; p < 0.001). His o y o H. pylo i in ec ion was highe wi hin he CVID coho (n = 8/9; 88.9% e sus n = 6/14; 44.9%; Table 1; p = 0.036). Rega ding he his opa hological cha ac e is ics o non-neoplas ic mucosa, he compa ison be ween CVID and non-CVID pa ien s showed ha lymphocy ic gas i is is a dis inc i e cha ac e is ic o CVID pa ien s (n = 7/7; 100.0% e sus n = 1/21; 4.8%; Table 1; p < 0.001). Lymphoid ollicles we e less p ominen in CVID pa ien s when compa ed wi h non-CVID pa ien s (n = 4/7; Figu e 2. Mo phology o common a iable immunode iciency (CVID)-associa ed gas ic cance . ( a ) Tubula gas ic cance , low-g ade (HE, 100 × magni ica ion); ( b ) ubula gas ic cance , high-g ade (HE, 100 × magni ica ion); ( c ) mucinous adenoca cinoma (HE, 50 × magni ica ion); ( d ) poo ly-cohesi e ca cinoma, PCC-NOS (HE, 200 × magni ica ion). No e he p esence o sca e ed signe ing cells (a ows); ( e, ) adenoma ous lesion, low-g ade dysplasia, wi h ubula ( e ) and illous ( ) a chi ec u e (HE, 50×magni ica ion). 3.1.4. His opa hological Fea u es Compa ed wi h Non-CVID Pa ien s To in es iga e in dep h he clinical ea u es o CVID pa ien s and he biological ele ance o he abo e desc ibed his opa hological indings, we compa ed ou esul s wi h hose ob ained om a con ol g oup o non-CVID pa ien s (n=21) diagnosed wi h ea ly s age gas ic cance [pT1a (n=15/21; 71.4%) and pT1b (n=6/21; 28.6%)]. The esul s a e p esen ed in Table 1. Gas ic cance in CVID pa ien s was diagnosed a a younge age compa ed wi h non-CVID pa ien s (median age: 43y e sus 75y; Table 1;p<0.001). His o y o H. pylo i in ec ion was highe wi hin he CVID coho (n=8/9; 88.9% e sus n =6/14; 44.9%; Table 1;p=0.036). Rega ding he his opa hological cha ac e is ics o non-neoplas ic mucosa, he compa ison be ween CVID and non-CVID pa ien s showed ha lymphocy ic gas i is is a dis inc i e cha ac e is ic o CVID pa ien s (n=7/7; 100.0% e sus n=1/21; 4.8%; Table 1;p<0.001). Lymphoid ollicles we e less Cells 2020,9, 1498 8 o 21 p ominen in CVID pa ien s when compa ed wi h non-CVID pa ien s (n=4/7; 57.1% e sus n =21/21; 100.0%; Table 1;p=0.011). Rega ding gas ic cance his o ype, he compa ison be ween he wo g oups showed no signi ican di e ences ega ding WHO and Lau é n classi ica ion (p=0.089 and p=0.502, espec i ely). Table 1. Clinicopa hological cha ac e is ics o CVID pa ien s wi h gas ic cance in compa ison wi h a con ol g oup o non-CVID pa ien s wi h ea ly s age gas ic cance . Fea u e CVID Pa ien s n=9 Non-CVID Pa ien s n=21 pValue Median alue ( ange) Median alue ( ange) Age 43 (27–62) 75 (45–84) 0.000 * N (%) N (%) Gende Female Male 5/9 (55.6%) 4/9 (44.4%) 11/21 (52.4%) 10/21 (47.6%) 1.000 Helicobac e pylo i in ec ion Posi i e Nega i e 8/9 (88.9%) 1/9 (11.1%) 6/14 (42.9%) 8/14 (57.1%) 0.036 * His opa hological Findings in Non-Neoplas ic Mucosa Lymphoci ic gas i is P esen Absen NA 7/7 (100.0%) 0/7 (0.0%) 2 1/21 (4.8%) 20/21 (95.2%) 0 0.000 * Lymphoid ollicles P esen Absen NA 4/7 (57.1%) 3/7 (42.9%) 2 21/21 (100.0%) 0/21 (0.0%) 0 0.011 * Neu ophilic ac i i y P esen Absen NA 2/7 (28.6%) 5/7 (71.4%) 2 0/21 (0.0%) 21/21 (100%) 0 0.056 In es inal me aplasia P esen Absen NA 7/7 (100.0%) 0/7 (0.0%) 2 21/21 (100.0%) 0/21 (0.0%) 0 NA His opa hological ea u es o gas ic cance pT s age (AJCC 8 h Ed) [25] pT1a pT1b pT2 NA 4/7 (57.1%) 2/7 (28.6%) 1/7 (14.3) 2 15/21 (71.4%) 6/21 (28.6%) 0/21 0 0.21 Lau én classi ica ion [23] In es inal Di use Mixed Inde e mina e 6/9 (66.7%) 1/9 (11.1%) 0/9 (0.0%) 2/9 (22.2%) 13/21 (61.9%) 1/21 (4.8%) 4/21 (19.0%) 3/21 (14.3%) 0.502 WHO classi ica ion (2019) [24] Tubula /papilla y Poo ly cohesi e Mucinous Mixed GCLS 6/9 (66.7%) 1/9 (11.1%) 2/9 (22.2%) 0/9 (0.0%) 0/9 (0.0%) 13/21 (61.9%) 1/21 (4.8%) 0/21 (0.0%) 4/21 (19.0%) 3/21 (14.3%) 0.089 P ecu so lesions P esen (adenoma) Absen NA 2/7 (28.6%) 5/7 (71.4%) 2 5/21 (23.8%) 16/21 (76.2%) 0 1.000 Cells 2020,9, 1498 9 o 21 Table 1. Con . Fea u e CVID Pa ien s n=9 Non-CVID Pa ien s n=21 pValue EBV and CMV in ec ion EBV in ec ion (EBER-ISH) Posi i e Nega i e 0/9 (0.0%) 9/9 (100.0%) 1/21 (4.8%) 20/21 (95.2%) 1.000 CMV in ec ion (IHC) Posi i e Nega i e 0/9 (0.0%) 9/9 (100.0%) 0/21 (0.0%) 21/21 (100.0%) NA NA, no a ailable; GCLS, gas ic cance wi h lymphoid s oma; EBV, Eps ein-Ba i us; CMV, cy omegalo i us; EBER-ISH, EBV encoded RNA in si u hyb idiza ion; IHC, immunohis ochemis y. S a is ically signi ican esul s a e highligh ed by an as e isk (*). 3.2. EBV and CMV We e No De ec ed he in Non-Neoplas ic Mucosa o Gas ic Cance in CVID Pa ien s We sea ched o EBV in ec ion by EBER-ISH and CMV in ec ion by IHC bo h in CVID pa ien s and in he non-CVID con ol g oup (Table 1). None o CVID cases showed EBER-ISH posi i i y in gas ic cance cells. One gas ic cance om he con ol g oup (n=1/21; 4.8%) exhibi ed di use posi i i y. This case showed p ominen lymphoid in il a ion in he umou s oma and abundan in a umou al lymphocy es, ul illing he c i e ia o he diagnosis o gas ic cance wi h lymphoid s oma (GCLS) [ 30 ]. In non-neoplas ic mucosa, sca e ed EBER-ISH posi i e lymphocy es we e ound in one case om he CVID pa ien coho (n =1/7; 14.3%). This is a non-speci ic inding and sca e ed EBER-ISH posi i e lymphocy es we e also ound in non-neoplas ic mucosa om non-CVID pa ien s (n=15/21; 71%). No CMV immuno eac i i y was obse ed ei he in non-neoplas ic mucosa o gas ic cance in any o he CVID and non-CVID pa ien s. 3.3. Immune Mic oen i onmen Changes in CVID Pa ien s Re lec Dys unc ional Immune Mic oen i onmen To explo e he pu a i e immune mechanisms ha may inc ease he suscep ibili y o CVID pa ien s o gas ic cance , we analysed he densi y and quali y o he immune in il a ion in non-neoplas ic mucosa (“dis an om” and “adjacen o”) and umou al issue om CVID pa ien s and compa ed he esul s wi h hose ob ained om he gas ic specimens o non-CVID pa ien s. We aimed o p o ide a spa ial/ empo al model o explain he ole o he immune in il a e in gas ic pa hogenesis in he con ex o CVID. Using immunohis ochemis y and subsequen analysis h ough digi al au oma ic quan i ica ion, we explo ed he in il a ion a e o cy o oxic CD8+T lymphocy es, CD4+helpe T cells, Foxp3+ egula o y T cells, GATA3+T lymphocy es, as well as CD20+B lymphocy es and CD138+ plasma cells. CVID cases showed s a is ically signi ican modi ica ions in he immune mic oen i onmen when compa ed wi h he esul s ob ained in non-CVID pa ien s (Tables 2and 3). Table 2. Immune cell in il a ion (Foxp3, GATA3, CD4, CD8, CD20, CD138) in in aepi helial and s omal compa men s o non-neoplas ic mucosa (“dis an om” and “adjacen o” gas ic cance ) and in gas ic cance o CVID cases in compa ison wi h non-CVID cases. Immune Cell Bioma ke CVID Pa ien s n=9 Non-CVID Pa ien s n=21 pValue 1. Non-neoplas ic mucosa dis an om gas ic cance (in aepi helial) n=7/9 Median alue (95% CI) (No. posi i e cells pe 20×PF) n=21/21 Median alue (95% CI) (No posi i e cells pe 20×PF) Foxp3 3.0 (1.0–8.0) 0.0 (0.0–1.0) 0.000 * GATA3 30.0 (2.1–43.9) 0.0 (0.0–6.9) 0.014 * Cells 2020,9, 1498 16 o 21 In non-CVID pa ien s, he e was an inc ease o Foxp3+, GATA3+, CD4+and CD8+T cell coun s as well CD20+B cell coun s when he non-neoplas ic mucosa dis an om umou was compa ed o gas ic cance (p=0.001; p=0.002; p=0.006; p=0.003; p=0.004, espec i ely). Simila esul s we e ob ained when he non-neoplas ic mucosa adjacen o umou was compa ed o gas ic cance (p alues: p=0.002; p=0.009; p=0.010; p=0.001; p=0.050, espec i ely). In con as , in CVID pa ien s, Foxp3+, GATA3+, CD4+and CD8+T cell coun s and CD20+B cell coun s did no di e signi ican ly when he non-neoplas ic mucosa was compa ed o gas ic cance (p>0.05). We obse ed an inc ease o CD8+/GATA3+lymphocy es o gas ic cance s compa ed o non-neoplas ic mucosa in non-CVID pa ien s along wi h a lack o signi ican di e ence in immune cell coun s in CVID-pa ien s. 3.3.2. H. pylo i In ec ion May In luence he Immune Mic oen i onmen in CVID Pa ien s. We compa ed he immune cell coun s in he CVID and non-CVID pa ien s wi h his o y o H. pylo i in ec ion. The de ailed esul s a e p esen ed in Supplemen a y Table S3. When H. pylo i-in ec ed CVID and non-CVID pa ien s we e compa ed, we ob ained simila esul s in mos lymphocy e coun compa isons. Howe e , in gas ic cance , H. pylo i-in ec ed CVID pa ien s ha bou ed mo e abundan GATA3+and CD8+T cells in compa ison wi h H. pylo i-in ec ed non-CVID pa ien s (p=0.008; p=0.005), while no signi ican di e ences we e obse ed in he whole se ies (Sec ion 3.3). Foxp3+and CD4+T cell coun s we e also inc eased in gas ic cance in CVID pa ien s (p=0.001; p=0.003), in his case in keeping wi h he esul s ob ained in he whole se ies (Sec ion 3.3). No ably, in non-neoplas ic mucosa adjacen o gas ic cance , GATA3+T cell coun s in he lamina p op ia we e simila be ween CVID (n =8) and non-CVID (n=6) pa ien s wi h a his o y o H. pylo i in ec ion (p=0.065). 4. Discussion Gas ic cance is an eme ging pheno ype wi hin he spec um o malignancies ha a ec pa ien s wi h CVID, anking second among CVID-associa ed cance s and i s among causes o cance - ela ed dea h [ 16 , 17 ]. In ou s udy, gas ic cance was no only he i s cause o cance - ela ed dea h, bu also he i s mos common malignancy in CVID pa ien s, p obably e lec ing he he e ogenei y o gas ic cance geog aphical dis ibu ion [ 18 ]. Mo eo e , as p e iously epo ed [ 17 ], CVID pa ien s had ea ly-onse gas ic cance , simila ly o wha happens in gas ic cance pa ien s ha bou ing he edi a y gas ic cance -associa ed synd omes [ 31 ]. This age an icipa ion may be explained by he pe missi e en i onmen o cance de elopmen in an immune-de ec i e backg ound. Addi ionally, he ea ly age o gas ic cance onse in ou coho may also be explained by he igo ous endoscopic ollow-up pe o med in ou cen e. The majo i y o CVID pa ien s in his coho we e diagnosed wi h ea ly-s age gas ic cance (66.7%), and none o hese pa ien s died due o cance - ela ed causes du ing he ollow-up pe iod. In con as , wo pa ien s we e diagnosed wi h un esec able disease and died wi hin one yea due o cance p og ession. Indeed, he o e all su i al o gas ic cance pa ien s wi h un esec able/me as a ic disease is less han one yea , e en wi h chemo he apy and he bes suppo i e ca e [ 32 ]. All oge he hese indings ein o ce he impo ance o uppe endoscopic su eillance in his popula ion [22]. The dys unc ionali y o he immune sys em in CVID pa ien s encompasses de ec i e B-cell ac i i y and di e en ia ion in o plasma cells, as well as he p esence o an abe an and pe sis en immune ac i a ion, which may a ou au oimmune diso de s [ 33 ]. Wo hy o no e, he p esence o an au oimmune backg ound seems o be ele an o he de elopmen o malignancies in his popula ion, as he incidence o au oimmune diso de s in CVID pa ien s wi h cance has been epo ed o be much highe han ha o CVID pa ien s who do no de elop malignancies [ 16 ]. In ou se ies, he majo i y o gas ic cance CVID pa ien s had a wide spec um o au oimmune diso de s (66.7%), poin ing o he impo ance o u he explo ing he possible con ibu ion o au oimmuni y o gas ic pa hology and Cells 2020,9, 1498 17 o 21 ca cinogenesis in CVID. Wi hin his con ex , i is ele an o emphasise ha non-neoplas ic mucosa in all CVID pa ien s exhibi ed lymphocy ic gas i is which may be a pheno ypic exp ession o a highe p opensi y o he immune sys em o igge an au oimmune esponse agains gas ic mucosa. The au oma ed digi al coun s o immune cells in his s udy p o ided, o he i s ime, a de ailed cha ac e isa ion o he immune dys unc ionali y in he gas ic mucosa o CVID pa ien s. In keeping wi h a s a us o T cell ane gy and exhaus ion o he immune sys em, he compa ison be ween CVID and non-CVID samples showed highe coun s o Foxp3+lymphocy es and PD-L1+immune cells bo h in he in aepi helial and lamina p op ia compa men s o CVID-associa ed gas ic mucosa. We obse ed also highe GATA3+lymphocy e coun s in he gas ic mucosa o CVID pa ien s. This inding may e lec he abundancy in ou samples o Foxp3+cells, which also exp ess GATA3 [ 34 ], as well as cy o oxic T cell dys unc ionali y. Indeed, GATA3 exp ession by cy o oxic T lymphocy es has been desc ibed as a mechanism o T cell exhaus ion, pa icula ly in au oimmune condi ions [ 35 , 36 ], as well as a mechanism o umou escape om he immune su eillance [ 37 ]. Mo eo e , i should be emphasised ha he high coun s o GATA3+lymphocy es in gas ic mucosa may be he consequence o a IL-4 d i en T-helpe -2-cell di e en ia ion o CD4+T cells [ 38 ], a mechanism desc ibed in he con ex o H. pylo i in ec ion and ha could in luence gas ic cance pa hogenesis [ 39 , 40 ]. This in e p e a ion is in line wi h he high p e alence o H. pylo i in ec ion in ou coho o CVID pa ien s (88.9%), as well as wi h he di e en esul s ob ained compa ing exclusi ely H. pylo i-in ec ed CVID and non-CVID pa ien s. Speci ically, we obse ed ha , in non-neoplas ic mucosa adjacen o gas ic cance , GATA3+ T cell coun s we e simila in CVID and non-CVID pa ien s wi h H. pylo i in ec ion. The e o e, he high GATA3+lymphocy e coun s in CVID pa ien s could be d i en o some ex en by H. pylo i in ec ion. Howe e , he numbe o a ailable cases o his compa ison was limi ed o only 8 and 6 cases, espec i ely in CVID and non-CVID pa ien s, and u he da a will be needed o con i m ou esul s. In keeping wi h CVID- ela ed de ec s in B-cell di e en ia ion, CD138+plasma cells and CD20+B lymphocy es showed lowe coun s bo h in non-neoplas ic mucosa and gas ic cance in CVID pa ien s. In ag eemen , lymphoid ollicles we e also less p ominen in non-neoplas ic gas ic mucosa o CVID pa ien s and he majo i y lacked ge minal cen es. In e es ingly, he e was consis ency be ween he inding o ci cula ing plasma cells (CD38 b igh IgM nega i e ) and he iden i ica ion o plasma cells in non-neoplas ic mucosa o CVID pa ien s. One possible explana ion o he highe gas ic cance isk in CVID is he impai ed an ibody esponses. The isk o gas ic cance in immunode iciencies cha ac e ised by hypogammaglobulinemia needs o be u he elucida ed. A ecen s udy demons a ed ha gas ic cance isk is no signi ican ly inc eased in pa ien s wi h IgA de iciency [ 15 ]. Isola ed case epo s o gas ic cance ha e been epo ed in young pa ien s wi h X-linked hypogammaglobulinaemia [ 41 – 44 ], bu he inc ease o gas ic cance isk in his popula ion has no been demons a ed ye . Abno mali ies in he inna e immune sys em and T cells, a he han immunoglobulin de iciencies, could be ele an o gas ic cance pa hogenesis in CVID pa ien s. In his s udy, we add essed in dep h he cha ac e isa ion o he immune mic oen i onmen in CVID-associa ed gas ic cance , as compa ed wi h a con ol g oup o gas ic cance pa ien s wi hou p ima y immunode iciency. The mos in e es ing indings came up when we explo ed he changes be ween non-neoplas ic mucosa and gas ic cance in he wo g oups. CD8+and GATA3+T cell coun s we e highe in non-neoplas ic mucosa o CVID pa ien s. Howe e , CD8+and GATA3+T cells coun s did no di e in gas ic cance when we compa ed he wo g oups. These indings could be due ei he o a dec ease o CD8+/GATA3+T cell coun s in CVID pa ien s and/o an inc ease o CD8+/GATA3+T cell coun s in non-CVID pa ien s. To explo e hese possibili ies, we analysed sepa a ely he modi ica ions ha occu ed be ween he non-neoplas ic mucosa and gas ic cance wi hin each o he wo coho s. Ou esul s sugges ha he disco dan indings encoun e ed in non-neoplas ic mucosa and gas ic cance , when compa ing he wo g oups, a e de e mined by an inc ease o CD8+and GATA3+T cells in gas ic cance o non-CVID pa ien s, con as ing wi h he lack o signi ican al e a ions o immune cell coun s in CVID pa ien s. The e o e, changes in he immune esponse o ca cinogenesis in CVID and Cells 2020,9, 1498 18 o 21 non-CVID pa ien s a e no equi alen , e lec ing he dys unc ionali y o he CVID-associa ed immune sys em and he inabili y o moun an e ec i e esponse o gas ic cance -d i en s imuli. Finally, we explo ed immune cell coun s in gas ic cance by compa ing exclusi ely H. pylo i-in ec ed CVID and non-CVID pa ien s. This analysis lead o he obse a ion ha , in he con ex o H. pylo i in ec ion, CVID-associa ed gas ic cance ha bou ed mo e abundan GATA3+and CD8+T cells in compa ison wi h non-CVID gas ic cance , as obse ed in non-neoplas ic mucosa in he whole se ies. These indings may e lec he impo ance o H. pylo i in ec ion in CVID-associa ed gas ic ca cinogenesis, as he p esence o his pa icula bac e ium may igge an addi ional s imula ion o he immune sys em, o e coming he in insic immune impai men in CVID pa ien s. As men ioned abo e, he limi ed numbe o cases in his s udy weakens he ele ance o ou indings and u he da a a e needed o explo e hese esul s. Ne e heless, se e al s udies ha e highligh ed he pu a i e ole o H. pylo i in ec ion in CVID-associa ed gas ic ca cinogenesis [ 20 ]. Mo eo e , he equen and p olonged use o p o on-pump inhibi o s in he CVID pa ien s o ou s udy could ha e enhanced gas ic ca cinogenesis by p omo ing gas ic dysbiosis [45,46]. An in eg a i e scheme depic ing he pu a i e mechanisms leading o inc eased suscep ibili y o gas ic cance de e mined by he de ec i e immune en i onmen associa ed wi h CVID is shown in Figu e 8. A limi a ion o ou s udy is he small numbe o CVID-associa ed gas ic cance s, in keeping wi h he a i y o his diso de . In he li e a u e, only 59 cases ha e been desc ibed so a [ 16 ]. Ou esul s could ha e been in luenced by he young age o he CVID pa ien s, a hypo hesis ha we excluded by showing, in he non-CVID con ol g oup, ha he immune cells coun s we e no in luenced by he age o he pa ien s. Cells 2020, 9, x 19 o 22 Figu e 8. De ec i e immune en i onmen in CVID and pu a i e mechanisms leading o inc eased suscep ibili y o gas ic cance . 5. Conclusions The s udy he ein epo ed ep esen s, o he bes o ou knowledge, he i s in dep h analysis o he immune mic oen i onmen in gas ic cance s (and espec i e non-neoplas ic mucosa) de eloped in CVID pa ien s in compa ison wi h non-CVID pa ien s. Ou indings poin o he ole o he dys unc ional immune sys em in CVID in he de elopmen o gas ic cance . Supplemen a y Ma e ials: The ollowing a e a ailable online a www.mdpi.com/xxx/s1, Table A1: De ailed IHC p o ocol o he an ibodies used in his s udy, Table A2: De ailed clinico-pa hological cha ac e is ics o CVID pa ien coho , Table A3: De ailed esul s o he immune cell coun ing (Foxp3, GATA3, CD4, CD8, CD20) in non- neoplas ic mucosa (dis an om and adjacen o umou ) and adenoca cinomas o CVID pa ien s in compa ison wi h non-CVID pa ien s in ec ed by Helicobac e pylo i. Figu e A1: Immune cell coun ing in epi helial and lamina p op ia compa men s in adenoma ous lesions om CVID and non-CVID pa ien s. No s a is ically signi ican di e ences we e ound when CVID and non-CVID pa ien s we e compa ed. Au ho Con ibu ions: Concep ualiza ion, I.G., S.L.S., C.O., J.C.M., F.C. and A.E.S.; me hodology, I.G. and C.C.; alida ion, I.G., S.L.S., J.C.M., F.C. and A.E.S; o mal analysis, I.G., C.C.; in es iga ion, I.G., C.C., S.L.S., C.F., A.M., S.P.S., R.D.F., P.R., E.F., J.T.C., R.C., G.G., J.S.A.; esou ces, I.G., C.C., S.L.S., J.A., F.C. and A.E.S.; w i ing— o iginal d a p epa a ion, I.G. and C.C.; w i ing— e iew and edi ing, I.G., C.C., S.L.S., C.F., A.M., S.P.S., R.D.F., P.R., E.F., J.T.C., R.C., G.G., J.S.A., C.O., J.C.M., F.C. and A.E.S.; isualiza ion, I.G. and C.C.; supe ision, S.L.S., F.C. and A.E.S.; p ojec adminis a ion, S.L.S., F.C. and A.E.S.; unding acquisi ion, S.L.S., J.C.M., F.C and A.E.S. All au ho s ha e ead and ag eed o he published e sion o he manusc ip . Funding: This a icle is a esul o he p ojec s DOCne (NORTE-01-0145-FEDER-000003/000029), suppo ed by No e Po ugal Regional Ope a ional P og amme (NORTE 2020), unde he PORTUGAL 2020 Pa ne ship Ag eemen , h ough he Eu opean Regional De elopmen Fund (ERDF). This esea ch was unded by FCT- Founda ion o Science and Technology/Minis é io da Ciência, Tecnologia e Ino ação, g an numbe PTDC/MED-PAT/32462/2017 and PTDC/BIM-MEC/2834/2014. This wo k is unded by g an PAC-PRECISE- LISBOA-01-0145-FEDER-016394, co- unded by FEDER h ough POR Lisboa 2020—P og ama Ope acional Regional de Lisboa PORTUGAL 2020 and FCT, and UID/BIM/50005/2019 unded by FCT/ Minis é io da Ciência, Tecnologia e Ensino Supe io (MCTES) h ough Fundos do O çamen o de Es ado. Acknowledgmen s: We a e hank ul o he s udy pa icipan s o he dona ion o hei samples. Con lic s o In e es : The au ho s decla e no con lic s o in e es . Figu e 8. De ec i e immune en i onmen in CVID and pu a i e mechanisms leading o inc eased suscep ibili y o gas ic cance . 5. Conclusions The s udy he ein epo ed ep esen s, o he bes o ou knowledge, he i s in dep h analysis o he immune mic oen i onmen in gas ic cance s (and espec i e non-neoplas ic mucosa) de eloped in CVID pa ien s in compa ison wi h non-CVID pa ien s. Ou indings poin o he ole o he dys unc ional immune sys em in CVID in he de elopmen o gas ic cance . Cells 2020,9, 1498 19 o 21 Supplemen a y Ma e ials: The ollowing a e a ailable online a h p://www.mdpi.com/2073-4409/9/6/1498/s1, Table S1: De ailed IHC p o ocol o he an ibodies used in his s udy, Table S2: De ailed clinico-pa hological cha ac e is ics o CVID pa ien coho , Table S3: De ailed esul s o he immune cell coun ing (Foxp3, GATA3, CD4, CD8, CD20) in non-neoplas ic mucosa (dis an om and adjacen o umou ) and adenoca cinomas o CVID pa ien s in compa ison wi h non-CVID pa ien s in ec ed by Helicobac e pylo i. Figu e S1: Immune cell coun ing in epi helial and lamina p op ia compa men s in adenoma ous lesions om CVID and non-CVID pa ien s. No s a is ically signi ican di e ences we e ound when CVID and non-CVID pa ien s we e compa ed. Au ho Con ibu ions: Concep ualiza ion, I.G., S.L.S., C.O., J.C.M., F.C. and A.E.S.; me hodology, I.G. and C.C.; alida ion, I.G., S.L.S., J.C.M., F.C. and A.E.S; o mal analysis, I.G., C.C.; in es iga ion, I.G., C.C., S.L.S., C.F., A.M., S.P.S., R.D.F., P.R., E.F., J.T.d.C., R.C., G.G., J.S.-A.; esou ces, I.G., C.C., S.L.S., J.A., F.C. and A.E.S.; w i ing—o iginal d a p epa a ion, I.G. and C.C.; w i ing— e iew and edi ing, I.G., C.C., S.L.S., C.F., A.M., S.P.S., R.D.F., P.R., E.F., J.T.d.C., R.C., G.G., J.S.-A., C.O., J.C.M., F.C. and A.E.S.; isualiza ion, I.G. and C.C.; supe ision, S.L.S., F.C. and A.E.S.; p ojec adminis a ion, S.L.S., F.C. and A.E.S.; unding acquisi ion, S.L.S., J.C.M., F.C and A.E.S. All au ho s ha e ead and ag eed o he published e sion o he manusc ip . Funding: This a icle is a esul o he p ojec s DOCne (NORTE-01-0145-FEDER-000003/000029), suppo ed by No e Po ugal Regional Ope a ional P og amme (NORTE 2020), unde he PORTUGAL 2020 Pa ne ship Ag eemen , h ough he Eu opean Regional De elopmen Fund (ERDF). This esea ch was unded by FCT-Founda ion o Science and Technology/Minis é io da Ci ê ncia, Tecnologia e Ino aç ã o, g an numbe PTDC/MED-PAT/32462/2017 and PTDC/BIM-MEC/2834/2014. This wo k is unded by g an PAC-PRECISE-LISBOA-01-0145-FEDER-016394, co- unded by FEDER h ough POR Lisboa 2020—P og ama Ope acional Regional de Lisboa PORTUGAL 2020 and FCT, and UID/BIM/50005/2019 unded by FCT/Minis é io da Ciência, Tecnologia e Ensino Supe io (MCTES) h ough Fundos do O çamen ode Es ado. Acknowledgmen s: We a e hank ul o he s udy pa icipan s o he dona ion o hei samples. Con lic s o In e es : The au ho s decla e no con lic o in e es . Re e ences 1. Bous iha, A.; Jeddane, L.; Pica d, C.; Al-He z, W.; Ailal, F.; Cha ila, T.; Cunningham-Rundles, C.; E zioni, A.; F anco, J.L.; Holland, S.M.; e al. Human Inbo n E o s o Immuni y: 2019 Upda e o he IUIS Pheno ypical Classi ica ion. J. Clin. Immunol. 2020,40, 66–81. [C ossRe ] [PubMed] 2. Bonilla, F.A.; Ba lan, I.; Chapel, H.; Cos a-Ca alho, B.T.; Cunningham-Rundles, C.; De La Mo ena, M.T.; Espinosa-Rosales, F.J.; Hamma s öm, L.; Nonoyama, S.; Quin i, I.; e al. In e na ional Consensus Documen (ICON): Common Va iable Immunode iciency Diso de s. J. Alle gy Clin. Immunol. P . 2015 ,4, 38–59. [C ossRe ] [PubMed] 3. Blanco, E.; P é ez-And é s, M.; M é ndez, S.A.; Se ano, C.; C iado, I.; Del Pino-Molina, L.; Sil a, S.; Mad uga, I.; Baka djie a, M.; Ma ins, C.; e al. De ec s in memo y B-cell and plasma cell subse s exp essing di e en immunoglobulin-subclasses in pa ien s wi h CVID and immunoglobulin subclass de iciencies. J. Alle gy Clin. Immunol. 2019,144, 809–824. [C ossRe ] [PubMed] 4. Ba bosa, R.R.; Sil a, S.L.; Sil a, S.P.; Melo, A.C.; Pe ei a-San os, M.C.; Ba a a, J.; Hamma s öm, L.; Cascalho, M.; Sousa, A.E. Reduced BAFF-R and Inc eased TACI Exp ession in Common Va iable Immunode iciency. J. Clin. Immunol. 2014,34, 573–583. [C ossRe ] 5. Azizi, G.; Rezaei, N.; Kiaee, F.; Ta akolinia, N.; Yazdani, R.; Mi sha iey, A.; Aghamohammadi, A. T Cell Abno mali ies in Common Va iable Immunode iciency. J. In es ig. Alle gol. Clin. Immunol. 2016 ,26, 233–243. [C ossRe ] 6. Vialla d, J.F.; Ruiz, C.; Guille , M.; Pelleg in, J.-L.; Mo eau, J.-F. Pe u ba ions o he CD8(+) T-cell epe oi e in CVID pa ien s wi h complica ions. Resul s Immunol. 2013,3, 122–128. [C ossRe ] 7. Vialla d, J.-F.; Camou, F.; And é , M.; Li e man, F.; Mo eau, J.-F.; Pelleg in, J.-L.; Blanco, P. Al e ed dend i ic cell dis ibu ion in pa ien s wi h common a iable immunode iciency. A h i is Res. The . 2005 ,7, R1052–R1055. [C ossRe ] 8. Ba bosa, R.R.; Sil a, S.P.; Sil a, S.L.; Tendei o, R.; Melo, A.C.; Ped o, E.; Ba bosa, M.A.D.C.P.; San os, M.C.; Vic o ino, R.M.; Sousa, A.E. Monocy e ac i a ion is a ea u e o common a iable immunode iciency i espec i e o plasma lipopolysaccha ide le els. Clin. Exp. Immunol. 2012,169, 263–272. [C ossRe ] 9. Von Spee-Maye , C.; Koemm, V.; Weh , C.; Goldacke , S.; Kindle, G.; Bulashe ska, A.; P oie i, M.; G imbache , B.; Ehl, S.; Wa na z, K. E alua ing labo a o y c i e ia o combined immunode iciency in adul pa ien s diagnosed wi h common a iable immunode iciency. Clin. Immunol. 2019 ,203, 59–62. [C ossRe ] Cells 2020,9, 1498 20 o 21 10. Ho, H.-E.; Cunningham-Rundles, C. Non-in ec ious Complica ions o Common Va iable Immunode iciency: Upda ed Clinical Spec um, Sequelae, and Insigh s o Pa hogenesis. F on . Immunol. 2020 ,11, 149. [C ossRe ] 11. Chapel, H.; Cunningham-Rundles, C. Upda e in unde s anding common a iable immunode iciency diso de s (CVIDs) and he managemen o pa ien s wi h hese condi ions. B . J. Haema ol. 2009 ,145, 709–727. [C ossRe ] [PubMed] 12. Lopes-Da-Sil a, S.; Rizzo, L.V. Au oimmuni y in Common Va iable Immunode iciency. J. Clin. Immunol. 2008,28, 46–55. [C ossRe ] [PubMed] 13. Mayo , P.C.; Eng, K.H.; Singel, K.L.; Ab ams, S.I.; Odunsi, K.; Moysich, K.B.; Fuleihan, R.; Ga abedian, E.; Luga , P.; Ochs, H.D.; e al. Cance in p ima y immunode iciency diseases: Cance incidence in he Uni ed S a es Immune De iciency Ne wo k Regis y. J. Alle gy Clin. Immunol. 2018,141, 1028–1035. [C ossRe ] 14. Mo az, E.; Taba si, P.; Mansou i, D.; Khos a i, A.; Ga ssen, J.; Velaya i, A.; Adcock, I.M. Cance s Rela ed o Immunode iciencies: Upda e and Pe spec i es. F on . Immunol. 2016,7, 197. [C ossRe ] [PubMed] 15. Mellemkjæ , L.; Hamma s öm, L.; Ande sen, V.; Yuen, J.; Heilmann, C.; Ba ing on, T.; Bjö kande , J.; Olsen, J.H. Cance isk among pa ien s wi h IgA de iciency o common a iable immunode iciency and hei ela i es: A combined Danish and Swedish s udy. Clin. Exp. Immunol. 2002 ,130, 495–500. [C ossRe ] [PubMed] 16. Kiaee, F.; Azizi, G.; Ra iemanesh, H.; Zainaldain, H.; Riz i, F.S.; Alizadeh, M.; Jamee, M.; Mohammadi, S.; Habibi, S.; Sha i i, L.; e al. Malignancy in common a iable immunode iciency: A sys ema ic e iew and me a-analysis. Expe Re . Clin. Immunol. 2019,15, 1105–1113. [C ossRe ] 17. Pul i en i, F.; Peco a o, A.; Cine o, F.; Mili o, C.; Valen e, M.; San angeli, E.; C escenzi, L.; Rizzo, F.; Tabolli, S.; Spada o, G.; e al. Gas ic Cance Is he Leading Cause o Dea h in I alian Adul Pa ien s Wi h Common Va iable Immunode iciency. F on . Immunol. 2018,9, 9. [C ossRe ] 18. B ay, F.; Fe lay, J.; Soe joma a am, I.; Siegel, R.L.; To e, L.A.; Jemal, A. Global cance s a is ics 2018: GLOBOCAN es ima es o incidence and mo ali y wo ldwide o 36 cance s in 185 coun ies. CA Cance J. Clin. 2018,68, 394–424. [C ossRe ] 19. De Pe is, G.; Dhungel, B.M.; Chen, L.; Chang, Y.-H.H. Gas ic Adenoca cinoma in Common Va iable Immunode iciency. In . J. Su g. Pa hol. 2014,22, 600–606. [C ossRe ] 20. Leone, P.; Vacca, A.; Dammacco, F.; Racanelli, V. Common Va iable Immunode iciency and Gas ic Malignancies. In . J. Mol. Sci. 2018,19, 451. [C ossRe ] 21. Manesh, A.T.; Azizi, G.; Heyda i, A.; Kiaee, F.; Shaghaghi, M.; Hossein-Khannaze , N.; Yazdani, R.; Abolhassani, H.; Aghamohammadi, A. Epidemiology and pa hophysiology o malignancy in common a iable immunode iciency? Alle gol. E Immunopa hol. 2017,45, 602–615. [C ossRe ] [PubMed] 22. Dhalla, F.; Da Sil a, S.P.; Lucas, M.; T a is, S.; Chapel, H. Re iew o gas ic cance isk ac o s in pa ien s wi h common a iable immunode iciency diso de s, esul ing in a p oposal o a su eillance p og amme. Clin. Exp. Immunol. 2011,165, 1–7. [C ossRe ] [PubMed] 23. Lau en, P. The wo his ological main ypes o gas ic ca cinoma: Di use and so-called in es inal- ype ca cinoma. an a emp a a his o-clinical classi ica ion. Ac a Pa hol. Mic obiol. Scand. 1965 ,64, 31–49. [C ossRe ] [PubMed] 24. WHO Classi ica ion o Tumou s. Diges i e Sys em Tumou s, 5 h ed.; WHO Classi ica ion o Tumou s Edi o ial Boa d: Lyon, F ance, 2019. 25. Dominguez-Muñoz, J.E. The Obes a in/G p o ein-coupled ecep o 39 sys em egula es pepsinogen I sec e ion in human s omach. In AJCC Cance S aging Manual; Mo essie : Be lin, Ge many, 2017. 26. Bankhead, P.; Lough ey, M.B.; Fe n á ndez, J.A.; Domb owski, Y.; McA , D.G.; Dunne, P.D.; McQuaid, S.; G ay, R.T.; Mu ay, L.J.; Coleman, H.G.; e al. QuPa h: Open sou ce so wa e o digi al pa hology image analysis. Sci. Rep. 2017,7, 16878. [C ossRe ] 27. Kulanga a, K.; Zhang, N.; Co igliano, E.; Gue e o, L.; Wald oup, S.; Jaiswal, D.; Ms, M.J.; Shah, S.; Hanks, D.; Wang, J.; e al. Clinical U ili y o he Combined Posi i e Sco e o P og ammed Dea h Ligand-1 Exp ession and he App o al o Pemb olizumab o T ea men o Gas ic Cance . A ch. Pa hol. Lab. Med. 2019 ,143, 330–337. [C ossRe ] 28. Ba bosa, R.R.; Sil a, S.P.; Sil a, S.L.; Melo, A.C.; Ped o, E.; Ba bosa, M.A.D.C.P.; San os, M.C.; Vic o ino, R.M.; Sousa, A.E. P ima y B-Cell De iciencies Re eal a Link be ween Human IL-17-P oducing CD4 T-Cell Homeos asis and B-Cell Di e en ia ion. PLoS ONE 2011,6, e22848. [C ossRe ] [PubMed] Cells 2020,9, 1498 21 o 21 29. S i as a a, A.; Lauwe s, G.Y. Pa hology o non-in ec i e gas i is. His opa hol. 2007 ,50, 15–29. [C ossRe ] [PubMed] 30. Gullo, I.; Oli ei a, P.; A helogou, M.; Gonçal es, G.; Pin o, M.L.; Ca alho, J.; Valen e, A.; Pinhei o, H.; And ade, S.; Almeida, G.M.; e al. New insigh s in o he in lamed umo immune mic oen i onmen o gas ic cance wi h lymphoid s oma: F om mo phology and digi al analysis o gene exp ession. Gas ic Cance 2018,22, 77–90. [C ossRe ] 31. Oli ei a, C.; Pinhei o, H.; Figuei edo, J.; Se uca, R.; Ca nei o, F. Familial gas ic cance : Gene ic suscep ibili y, pa hology, and implica ions o managemen . Lance Oncol. 2015,16, e60–e70. [C ossRe ] 32. Du ã es, C.; Almeida, G.M.; Se uca, R.; Oli ei a, C.; Ca nei o, F. Bioma ke s o gas ic cance : P ognos ic, p edic i e o a ge s o he apy? Vi chows A ch. 2014,464, 367–378. [C ossRe ] 33. Paquin-P oulx, D.; Sandbe g, J.K. Pe sis en Immune Ac i a ion in CVID and he Role o IVIg in I s Supp ession. F on . Immunol. 2014,5, 5. [C ossRe ] 34. Wan, Y. GATA3: A mas e o many ades in immune egula ion. T ends Immunol. 2014 ,35, 233–242. [C ossRe ] [PubMed] 35. Singe , M.; Wang, C.; Cong, L.; Ma jano ic, N.D.; Kowalczyk, M.S.; Zhang, H.; Nyman, J.; Sakuishi, K.; Ku ulus, S.; Genne , D.; e al. A Dis inc Gene Module o Dys unc ion Uncoupled om Ac i a ion in Tumo -In il a ing T Cells. Cell 2016,166, 1500.e9–1511.e9. [C ossRe ] [PubMed] 36. Medsge , T.A.; I anco, D.E.; Ka da a, L.; Mo el, P.A.; Lucas, M.R.; Fuschio i, P. GATA-3 up- egula ion in CD8+T cells as a bioma ke o immune dys unc ion in sys emic scle osis, esul ing in excessi e in e leukin-13 p oduc ion. A h i is Rheum. 2011,63, 1738–1747. [C ossRe ] [PubMed] 37. Sowell, R.T.; Kaech, S.M. P obing he Di e si y o T Cell Dys unc ion in Cance . Cell 2016 ,166, 1362–1364. [C ossRe ] [PubMed] 38. Lee, G.R. T ansc ip ional egula ion o T helpe ype 2 di e en ia ion. Immunol. 2014 ,141, 498–505. [C ossRe ] 39. Sugimo o, M.; Yamaoka, Y.; Fu u a, T. In luence o in e leukin polymo phisms on de elopmen o gas ic cance and pep ic ulce . Wo ldj. Gas oen e ol. 2010,16, 1188–1200. [C ossRe ] 40. Liu, X.; Cao, K.; Xu, C.; Hu, T.; Zhou, L.; Cao, D.; Xiao, J.; Luo, L.; Guo, Y.; Qi, Y. GATA-3 augmen a ion down- egula es Connexin43 in Helicobac e pylo i associa ed gas ic ca cinogenesis. Cance Biol. The . 2015 , 16, 987–996. [C ossRe ] 41. Hasan, S.; Hemmige, V.; Fo bes, L.R.; O ange, J.S.; Hajja , J. Gas ic Adenoca cinoma in he Se ing o X-Linked Agammaglobulinemia (XLA) and HIV. J. Alle gy Clin. Immunol. 2015,135, AB184. [C ossRe ] 42. Lackmann, G.M.; Niehues, T.; Wahn, V. 143 Gas ic Adenoca cinoma In a Boy wi h X-Linked Agammaglobulinemia. Pedia . Res. 2004,56, 488. [C ossRe ] 43. Boone, A.T.S.; Ma í nez, M.G.T.; He e a, G.L.; Po illa, J.O.D.L.; Padilla, S.E.E.; Rosales, F.J.E.; Reyes, S.O.L. Gas ic Adenoca cinoma in he Con ex o X-linked Agammaglobulinemia. J. Clin. Immunol. 2013 ,34, 134–137. [C ossRe ] [PubMed] 44. Hajja , J.; Hasan, S.; Fo bes, L.R.; Hemmige, V.; O ange, J.S. Gas ic Adenoca cinoma in a Pa ien wi h X-Linked Agammaglobulinemia and HIV: Case Repo and Re iew o he Li e a u e. F on . Pedia . 2016 ,4, 142. [C ossRe ] [PubMed] 45. B uno, G.; Zacca i, P.; Rocco, G.; Scalese, G.; Pane a, C.; Po owska, B.; Pon one, S.; Se e i, C. P o on pump inhibi o s and dysbiosis: Cu en knowledge and aspec s o be cla i ied. Wo ldj. Gas oen e ol. 2019 ,25, 2706–2719. [C ossRe ] [PubMed] 46. Fe ei a, R.M.; Ma ques, J.P.; Pin o-Ribei o, I.; Cos a, J.L.; Ca nei o, F.; Machado, J.C.; Figuei edo, C. Gas ic mic obial communi y p o iling e eals a dysbio ic cance -associa ed mic obio a. Gu 2017 ,67, 226–236. [C ossRe ] [PubMed] © 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/).