Second B ea h: concep ual amewo k o localized immune
cascade p og amming in desmoplas ic umo s (in silico– alida ed,
li e a u e-ancho ed)
A icle ype: Concep Pape (P eclinical)
Disclaime : Resea ch concep o discussion and p eclinical es ing only. This documen con ains no
clinical ins uc ions and is no medical ad ice.
Consen s a emen /e hical app o al: This wo k does no equi e e hical app o al as he e a e no p oced-
u es in human o animal subjec s.
Funding suppo : This esea ch did no ecei e any speci ic g an om undingagencies in he public,
comme cial, o no - o -p o i sec o s.
E hics s a emen : This manusc ip does no equi e any e hical e iew, he e is no use o da a o
manipula ions in animals o subjec s.
A achmen s: The manusc ip includes 6 igu es and 2 ables.
Mu ad No uzo - co esponding au ho (ORCID: 0009-0007-2386-6332)
E-mail: [email p o ec ed]
Independen Biomedical Resea che - Aze baijan, Baku
P ima y Au ho & Concep O igina o o Second B ea h;P ojec Di ec o esponsible o o e all concep
de elopmen , s a egic planning, and coo dina ion o all esea ch s ages, including in silico and in i o
phases. De eloped he concep , cascade design and manages he implemen a ion o he me hodology in
silico.
Ke al Ra al, PhD (ORCID: 0000-0003-4114-0352)
Ramanbhai Pa el College o Pha macy, CHARUSAT Uni e si y - India, Anand
In Silico Modeling Enginee ; esponsible o he de elopmen , execu ion, and op imiza ion o in silico
modeling wo k lows, da a analysis, and in eg a ion o compu a ional esul s in o he cascade amewo k.
Ulka Shi aliye a, PhD candida e (ORCID: 0009-0005-1314-2774)
Aze baijan S a e Oil and Indus y Uni e si y - Aze baijan, Baku
P ojec Membe (Chemis y); Pa icipa ed in he p ojec as an addi ional con ibu o om he ield o
chemis y, suppo ing he in e disciplina y o ma o he eam and aking pa in gene al discussions
du ing he ea ly de elopmen phase.
ABSTRACT
We p esen “Second B ea h”, a es able concep ual amewo k o sequen ial, localized immune
p og amming in desmoplas ic solid umo s. Each cascade module is modeled in silico and explici ly
c oss- e e enced o published in i o and in i o e idence (see Re e ences), suppo ing biological
plausibili y wi hou claiming new we -lab expe imen s in i o. The amewo k speci ies con olled,
s epwise in a umo al ac i a ion o inna e and adap i e immuni y unde limi ed sys emic exposu e.
Ne wo k and en ichmen analyses highligh coo dina ed hubs (e.g., TNF, TLR4, STAT1, CTLA4, CD274)
and implica e β-ca enin/Wn -linked p og ams a a icking and checkpoin - eadiness s eps, aligning wi h
he p oposed sequence logic. Collec i ely, “Second B ea h” p o ides a mechanis ic a ionale and es able
p edic ions, including p ede ined go/no-go c i e ia, o con e ing immune-cold, desmoplas ic umo s in o
mo e esponsi e s a es and o e s a s uc u ed basis o p eclinical alida ion.
CONCEPTUAL FRAMEWORK OVERVIEW
Backg ound: Despi e he success o immune checkpoin inhibi o s (ICIs) in ce ain cance s, many la e-
s age solid umo s emain “immune-cold,” cha ac e ized by low T-cell in il a ion, dense ex acellula
ma ix (ECM), s omal and ascula ba ie s, and poo esponses o sys emic immuno he apy
[4,5,6,7,32,33]. O e coming hese esis ance mechanisms equi es localized and con olled
ep og amming o he umo mic oen i onmen (TME) o pe mi e ec i e an i- umo immuni y [4,5].
Objec i e: S a egy p oposes a bioma ke -guided, s aged, and locally con ined immune cascade designed
o enable ein il a ion and ac i a ion o endogenous o au ologous T cells in p e iously un esponsi e
solid umo s.
Me hods: “Second B ea h” in ol es a sequen ial in e en ion a ge ing physical and immunologic
ba ie s. Local enzyma ic ma ix dis up ion using a collagenase–hyalu onidase mix u e combined wi h
lysyl oxidase inhibi ion educes ECM densi y and s omal ba ie s [31,32]. T ansien ec ui men and
ac i a ion o inna e immune cells is induced using weakly immunogenic bac e ia o localized oll-like
ecep o agonis s o gene a e local dange signals [18,19,20,34]. Con olled, mic odosed in a umo al
cy okine pulses (IL-12, IFN-γ, TNF-α) ampli y local an igen p esen a ion and e ec o T-cell p iming
while minimizing sys emic exposu e [1,2,9,10,11,21,22,23]. Op ional au ologous T-cell augmen a ion can
be adminis e ed in a umo ally o sys emically du ing he window o heigh ened immune ac i a ion
[16,17,24,25,26,27]. A eco e y o con ainmen phase using local an ibio ics o immunomodula o s limi s
excessi e in lamma ion and es o es issue homeos asis a e bac e io he apy [18,19,20,34].
Resul s: Ta ge alida ion iden i ied key p o eins and genes in ol ed in immune ac i a ion, ECM
emodeling, and cy okine signaling. P o ein–p o ein in e ac ion analysis e ealed densely in e connec ed
hub nodes, including TNF, TLR4, CTLA4, STAT1, and CD274. Func ional en ichmen highligh ed
signi ican in ol emen o he Wn signaling pa hway, wi h hub nodes such as APC, LRP5, CTNNB1,
and AXIN1 po en ially egula ing β-ca enin ac i a ion and cell p oli e a ion. Gene co-occu ence ne wo k
analysis demons a ed s ong in e dependencies among IFNG, TLR4, CD86, TNF, NFKB1, CTLA4, and
CD8A, sugges ing coo dina ed egula ion o immune ac i a ion and checkpoin mechanisms wi hin he
p oposed cascade.
Conclusions: Second B ea h ep esen s a no el p eclinical app oach o con e immunologically “cold”
umo s in o esponsi e a ge s o an i- umo immuni y. I s sequen ial, localized design aims o enhance
e icacy while minimizing sys emic oxici y. P eclinical ne wo k and en ichmen analyses p o ide
mechanis ic suppo o i s p oposed mul i-s ep immune cascade, guiding u u e in i o and in i o
alida ion.
Keywo ds: Tumo , cance , oncology, Tumo mic oen i onmen , cance mic oen i onmen , immunology,
immune-cold umo s, in a umo al immuno he apy, ex acellula ma ix emodeling, immuno he apy
INTRODUCTION
Immunologically “cold” solid umo s a e de ined by low in a umo al CD8⁺ T-cell densi y, a weak IFN-γ
signa u e, and he p esence o abno mal ascula u e, ele a ed in e s i ial luid p essu e (IFP), and a dense
ex acellula ma ix (ECM) [6,7,32,33]. These ea u es hinde an igen p esen a ion, es ic e ec o -cell
in il a ion, and con ibu e o he poo esponsi eness o such umo s o immune checkpoin inhibi o s
(ICIs) [13,14,15]. To add ess his challenge, we p opose ha sequen ial local immune p og amming can
e ec i ely con e a cold umo pheno ype in o a ho , immunologically ac i e s a e. This s a egy
in ol es PRR-d i en inna e p iming, ein o cemen o he IL-12/IFN-γ/TNF-α axis, and con olled ECM
modula ion, he eby c ea ing a o able condi ions o ICI esponsi eness while minimizing sys emic
exposu e [1,2,4,5,9,10,11]. This s udy ep esen s a logical con inua ion o he a icle ‘Cascade Medicine:
A chi ec u e o The apy o a Sus ainable Ou come’ (M. No uzo , 2025), whe e he ini ial cascade
concep was in oduced.
The concep ual amewo k p oposed o his app oach consis s o i e in e ela ed modules: (A) local
inna e p iming, (B) local Th1 ac i a ion, (C) con olled ECM modula ion o educe IFP and inc ease
issue po osi y, (D) a ge ed deli e y o CD8⁺ T cells and NK cells, and (E) sys emic ICI sensi iza ion.
T ansi ions be ween hese modules a e guided by local bioma ke signals, and eadiness o ICI he apy is
quan i ied by a composi e Wa m h Readiness Index (WRI). This model is pa icula ly ele an o la e-
s age desmoplas ic o exclusiona y umo s, which a e cha ac e ized by subs an ial physical deli e y
ba ie s and an ini ially cold pheno ype [31,32,33].
Despi e ad ances, signi ican knowledge gaps emain. In a umo al in e en ions such as PRR o STING
agonis s, local IL-12 o mula ions, and ECM- a ge ing me hods ha e yielded limi ed e icacy o
unaccep able oxici y unde sys emic exposu e [4,5,9,11]. I is no ye known whe he an o de ed and
bioma ke -guided sequence o PRR ac i a ion, Th1 ein o cemen , and ECM modula ion is essen ial o
pheno ype con e sion and ICI sensi i i y. Mo eo e , ope a ional e en ion c i e ia and clea go/no-go
ules o such in e en ions a e lacking.
The cen al esea ch ques ion, he e o e, is whe he bioma ke -guided sequen ial local ac i a ion
(PRR→Th1→ECM) in immunologically cold, desmoplas ic umo s can ele a e he WRI abo e a
p ede ined h eshold and imp o e CD8⁺/NK cell in il a ion and ICI sensi i i y, compa ed wi h ei he
simul aneous deli e y o he same componen s o pa ial applica ion o indi idual modules. The p ima y
endpoin s include an inc ease in IFN-γ signa u e, a educ ion in IFP, and enhanced CD8⁺ densi y pe mm²
[6,7].
Ope a ionally, an immune-cold umo is de ined by low CD8⁺ in il a ion densi y, a weak IFN-γ signa u e,
low o ocal PD-L1 exp ession, and low umo mu a ional bu den, combined wi h a dense ECM and
ele a ed IFP [6,7,32,33]. S a i ica ion elies on a combina ion o hese ea u es, wi h “cold” umo s
classi ied as hose mee ing a leas wo o he h ee key c i e ia, wi h p io i y gi en o CD8⁺ densi y and
IFN-γ sco e. Baseline bioma ke moni o ing includes CD8⁺/mm², NK signa u es, IFN-γ signa u e, MHC-
I/II exp ession, ascula no maliza ion ma ke s, IFP and pe usion le els, ECM densi y, TCR clonali y,
and epi ope sp eading.
Taken oge he , his amewo k emphasizes he need o a s uc u ed and bioma ke -d i en s a egy o
ans o m immune-cold umo s in o esponsi e pheno ypes. By sys ema ically es ing whe he sequen ial
local immune p og amming imp o es umo immunogenici y and sensi izes umo s o ICIs, his esea ch
seeks o close c i ical gaps in cance immuno he apy and p o ide a pa h owa d sa e and mo e e ec i e
in e en ions o desmoplas ic and exclusiona y umo ypes [4,13,14,15,31,32,33].
2. MATERIALS & METHODS
2.1 Ma e ials
2.1.1 Key Design P inciples and No el y
The s udy was buil on h ee cen al pilla s o no el y. Fi s , local sequen ial immune p og amming was
designed o engage inna e ac i a ion, Th1 axis pola iza ion, and con olled ECM emodeling in a s epwise
and causal manne , wi h he goal o con e ing immune-cold umo s in o immune-ho pheno ypes
[1,2,4,5,9,10,11]. Second, bioma ke -ga ed go/no-go ansi ions we e inco po a ed o ensu e ha each
s age ad anced only upon achie ing de ined local h esholds, including inc eases in IFN-γ signa u e,
educ ions in IFP wi h imp o ed pe usion, and enhanced CD8⁺ T-cell densi y [6,7,32]. Thi d, a sa e y-by-
design amewo k was implemen ed, which emphasized exposu e localiza ion and he inclusion o a
manda o y an ibac e ial sa e y window ollowing bac e ial p iming o minimize sys emic oxici y while
main aining a du able immune imp in [18,19,20,34].
Dis inc i e elemen s included he aming o immune p e-condi ioning as an in es iga ional s age wi h
de ined ma ke s and ansi ion ules, physiologically gen le ECM co-modula ion using a combina ion o
collagenase, hyalu onidase, and a lysyl oxidase inhibi o in iso onic saline (0.9% NaCl) [31,32], and
p o ec i e sca olding o suppo i e modules such as NK/IL-15 o ex aco po eal measu es applied
s ic ly by indica ion [17,27].
2.1.2 Classes o Tools
Local inna e p iming (PRR): a enua ed bac e ia, bac e ial pa e ns, TLR/STING agonis s, and
PAMP ca ie s [18,19,20,34].
Local Th1 axis: IL-12 → IFN-γ → TNF-α adminis e ed a low, localized exposu e wi h ma ix-
bound ca ie s o e en ion [1,2,9,10,11,21,22,23].
ECM modula ion: con olled deli e y o collagenase, hyalu onidase, and a lysyl oxidase
inhibi o in 0.9% NaCl o educe IFP and enhance po osi y wi hou sys emic exposu e [31,32].
E ec o -cell ec o s: au ologous o HLA-compa ible CD8⁺ T cells and/o NK cells wi hou
manda o y gene ic modi ica ion [16,17,24,25,26,27].
Sys emic sensi iza ion: immune checkpoin inhibi o s (an i-PD-1/PD-L1, an i-CTLA-4)
adminis e ed a e success ul pheno ype con e sion [13,14,15,28,29,30].
Sys emic/suppo i e he apy: NK-cell p oduc s, γc-cy okine suppo (IL-15 class),
c ys alloids/plasma, ex aco po eal me hods, mic obio a-di ec ed he apies, and hepa op o ec i e
measu es applied s ic ly unde p ede ined sa e y indica ions [17,27].
An imic obial p o ec ion: an ibio ic egimens ailo ed o he bac e ial agen used o PRR
p iming [18,19,20,34].
2.1.3 ECM Modula ion Mix u e
A minimal-dose o mula ion o collagenase, hyalu onidase, and lysyl oxidase inhibi o is p oposed o
in a umo al deli e y in cycles, un il bioma ke h esholds would indica e ECM emodeling ( educ ion in
IFP and imp o emen in pe usion). T ea men would be wi hd awn upon any indica ion o o e -
deg ada ion, ascula comp omise, edema, o leakage.
2.1.4 Compu a ional and Bioin o ma ic Inpu s
To e alua e pha macological and genomic in luences on cascade p og ession, an open-sou ce
compu a ional oolchain was applied:
Chemical & PK/ADME ools: RDKi , Open Babel, SwissADME, pkCSM, adme SAR.
Genomics (pha macogenomics): Pha mGKB, CPIC, gnomAD, 1000 Genomes, Ensembl VEP,
and SnpE .
Sys ems/Ne wo k analysis: STRING ( 11.5), Cy oscape ( 3.9.1), cy oHubba plugin, DAVID
( 2021), and g:P o ile .
Cance co-occu ence and ou come da a: TCGA and cBioPo al.
2.1.5 P eclinical Models
O ho opic, desmoplas ic, and immune-excluding umo models a e p oposed o p eclinical alida ion o
mimic cold umo mic oen i onmen s wi h deli e y ba ie s [31,32,33]. In e en ion a ms a e designed
o include single-s ep abla ions (e.g., PRR only), sequen ial combina ions (PRR→Th1→ECM), and
suppo i e addi ions (e ec o cells, ICI). Nega i e con ols we e es ablished by swi ching o PRR
pa hways, neu alizing IFN-γ, o inhibi ing ECM modula o s.
2.1.6 Sa e y and Bioe hics
All in e en ions we e designed exclusi ely o p eclinical esea ch unde IACUC and biosa e y o e sigh .
Sa e y p o isions a e in eg a ed o include an ibac e ial phases ollowing PRR p iming, cy okine
localiza ion wi h ma ix-bound ca ie s o a oid sys emic leakage, con olled ECM modula ion o p e en
o e -deg ada ion, moni o ing o hype in lamma o y esponses du ing e ec o deli e y, and s ic dono -
ma e ial compa ibili y checks o adop i e ans e s [9,10,11,18,19,20,24,25,26,27,34].
2.2 Me hods
2.2.1 Sequen ial In e en ion F amewo k
The expe imen al p o ocol s uc u ed a en-s age sequence designed o g adually con e immune-cold
umo s in o immune-ho pheno ypes. Each s age is de ined by a speci ic goal, ansi ion bioma ke s, and
sa e y checks.
1. Immunos imula ion (P epa a o y): Mic obio a, me aboli es, and mic onu ien s we e op imized
o es o e baseline immune compe ence p io o umo -di ec ed in e en ions. T ansi ion o he
nex s age equi ed no maliza ion o immune pa ame e s and exclusion o ac i e in ec ions.
2. In a umo al Bac e io he apy (PRR P iming): A enua ed bac e ial p epa a ions o PRR
agonis s we e in ended o localized in a umo al deli e y in u u e p eclinical se ings o ac i a e
dend i ic cells (DCs) and ini ia e a local dange cascade [18,19,20,34]. Ea ly inc eases in IFN-γ
signa u e wi hou sys emic in lamma ion we e equi ed o ad ance.
3. An ibac e ial Phase (Sa e y Window): An an ibio ic egimen ailo ed o he bac e ial agen was
adminis e ed sys emically o mi iga e sepsis isk while p ese ing local immune imp in ing
[18,19,20,34].
4. Local An icoagulan B idge (By Indica ion): Applied only in cases o hype coagulabili y, his
s age in ol ed an icoagulan deli e y o educe mic o h ombosis and asospasm. P og ession
equi ed imp o ed coagula ion/pe usion ma ke s wi hou bleeding isk.
5. Local Cy okine Axis (IL-12 → IFN-γ → TNF-α): Cy okines we e in ended o localized
in a umo al deli e y in u u e p eclinical se ings in low doses ia ma ix-bound ca ie s o
p omo e Th1 pola iza ion and ascula pe meabili y. T ansi ion equi ed inc eases in IFN-γ
signa u e and MHC-I/II exp ession wi hou sys emic leakage [1,2,9,10,11,21,22,23].
6. ECM Modula ion: A minimal-dose in a umo al mix u e o collagenase, hyalu onidase, and a
lysyl oxidase inhibi o in 0.9% NaCl was deli e ed cyclically o educe IFP and inc ease po osi y.
Ad ancemen equi ed educ ions in ECM densi y and IFP, wi h imp o ed pe usion bu wi hou
bleeding o edema.
7. E ec o -Cell Adminis a ion: Au ologous o HLA-compa ible CD8⁺ T cells and/o NK cells
we e adminis e ed in a umo ally o sys emically a e ba ie elie [16,17,24,25,26,27].
T ansi ion equi ed e idence o in il a ion, inc eased TCR clonali y, and absence o
hype in lamma ion.
8. Pe sonalized Onco accina ion (Op ional): Applied selec i ely o consolida e epi ope co e age.
T ansi ion equi ed induc ion o an igen-speci ic T-cell esponses wi hou in e e ence in he base
cascade.
9. Suppo i e Immunomodula ion and ICI (Op ional): Sys emic checkpoin inhibi o s (an i-PD-
1/PD-L1, an i-CTLA-4) and suppo i e modules (e.g., NK p oduc s, IL-15) we e in oduced only
when bioma ke e idence o ICI sensi i i y was achie ed [13,14,15,28,29,30].
10. Adju an Local Me hods (Res ic ed): Local modali ies we e ese ed o esidual lesions a e
con i ming he e ec s o s ages 1–7.
2.2.2 Causal Checks and Go/No-Go C i e ia
A p ede ined checkpoin s, bioma ke h esholds de e mined whe he o p oceed, pause, o e mina e:
A e PRR p iming: DC ac i a ion and ea ly IFN esponses [18,19,20,34].
A e Th1 axis: Sus ained IFN-γ signa u e wi hou sys emic cy okine leakage [1,2,9,10,11].
A e ECM modula ion: Signi ican educ ion in IFP and inc eased pe usion [31,32].
A e e ec o deli e y: Ele a ed CD8⁺ T-cell densi y and cy o oxic ac i i y [16,24,25].
Be o e ICI ini ia ion: Con i med ansi ion o a “ho ” pheno ype (↑ IFN-γ, ↑ in il a ion, ↑ PD-L1
exp ession) [13,14,15].
Sa e y o e ides we e igge ed by sys emic cy okine leakage, ECM o e -deg ada ion, mic o-bleeding,
edema, o lack o CD8⁺ in il a ion despi e IFP educ ion.
2.2.3 P edic ion Ma ix and Falsi ica ion C i e ia
To ensu e causal a ibu ion, p edic ions we e mapped o speci ic ou comes:
PRR p iming → DC ac i a ion (assessed by IHC o CD11c and RNA signa u es) [18,19,20,34].
IL-12/IFN-γ/TNF-α signaling → “wa ming” (assessed by IFN-γ sco e, MHC-I/II up egula ion)
[1,2,9,10,11].
ECM modula ion → imp o ed in il a ion (assessed by SHG mic oscopy, IHC, IFP/pe usion
measu emen s) [31,32].
E ec o deli e y → umo con ol (assessed by g ow h kine ics, TCR clonali y, epi ope sp eading)
[16,24,25].
Wa m pheno ype → ICI sensi i i y (assessed by esponse analogs such as ORR/PFS in p eclinical
models) [13,14,15,28,29,30].
Falsi ica ion was de ined as he absence o p edic ed bioma ke changes compa ed wi h ma ched con ols.
2.2.4 In-Silico Modeling and Ne wo k Pha macology
To complemen p eclinical es ing, a dynamic cascade simula o (Py hon/NumPy/Ma plo lib code bundle)
was applied unde he ou e (PRR→Th1→ECM→E ec o s→ICI). In e en ions we e modeled
exclusi ely wi h in a umo al local deli e y; sys emic pha macokine ics (Cmax, AUC) we e analyzed
only o an ibio ics.
Pha macogenomics in eg a ion: Va ian s in PRR pa hways (TLR4, TLR9, STING), cy okine
signaling (IFNGR1/2, STAT1), and immune checkpoin s (CD274, CTLA4) we e mapped om
Pha mGKB, CPIC, gnomAD, and Ensembl VEP o pa ame e mul iplie s in he simula o .
Ne wo k cons uc ion: Candida e gene se s we e mapped in o p o ein–p o ein in e ac ion
ne wo ks using STRING 11.5. Ne wo k opology was analyzed wi h Cy oscape 3.9.1 and
cy oHubba.
Func ional en ichmen : Top hub p o eins we e subjec ed o KEGG pa hway en ichmen ia
DAVID 2021, wi h signi icance a Benjamini-adjus ed p < 0.05.
Pa hway and co-occu ence analysis: Gene ic pa hway en ichmen was c oss- alida ed using
TCGA and cBioPo al da ase s. Mu ual exclusi i y and co-occu ence o op hub p o eins we e
assessed ac oss 10,897 umo s spanning 32 cance ypes.
2.2.5 Endpoin s
P ima y endpoin s: Inc ease in IFN-γ signa u e, educ ion in IFP, and ise in in a umo al CD8⁺
densi y [6,7].
Seconda y endpoin s: TCR clonali y, epi ope sp eading, pe usion/po osi y, umo g ow h con ol,
su i al analogs, and sa e y o localized exposu e (no sys emic signal leakage)
[4,5,13,14,15,31,32,33].
3. RESULTS
3.0.1 Suppo ing in i o and in i o e idence om he li e a u e
Published expe imen al s udies om o he independen g oups p o ide empi ical suppo o each s ep o
he p oposed cascade. Enzyma ic deg ada ion o ex acellula ma ix has been shown o educe in e s i ial
luid p essu e and enhance T-cell in il a ion in mu ine umo models [31,32]. Local adminis a ion o IL-
12, IFN-γ, o TNF-α inc eased an igen p esen a ion and p omo ed CD8⁺ T-cell p iming while minimizing
sys emic exposu e [1,2,9–11,21–23]. Toll-like ecep o agonis s and a enua ed bac e ial ec o s ha e
success ully igge ed local inna e immune ac i a ion and dange signaling [18–20,34]. T ansien
con ainmen s a egies, including local an ibio ics o immunomodula o s, ha e been epo ed o esol e
excessi e in lamma ion ollowing bac e io he apy [18–20]. Collec i ely, hese in i o and in i o
obse a ions [1–34] con i m he easibili y o he indi idual modules, p o iding a biological ounda ion
o he subsequen in silico ne wo k alida ion o he ull Second B ea h cascade.
3.0.2 In silico alida ion
3.1 Ta ge Valida ion
A lis o p o eins/genes co esponding o each se is summa ized in Table 1.
Table 1: Lis o p o eins/genes co esponding o di e en ou es
Lis o genes
TLR4, LY96, TLR9, MB21D1, TMEM173, MYD88, TICAM1, TBK1, IRF3, IRF7, NFKB1,
RELA, CD80, CD86, CCR7, IL12A, IL12B, IL12RB1, IL12RB2, STAT4, IFNG, IFNGR1,
IFNGR2, JAK1, JAK2, STAT1, TNF, TNFRSF1A, TNFRSF1B, MAP3K7, NFKBIA,
COL1A1, COL3A1, HAS2, HYAL1, HYAL2, LOX, MMP2, MMP9, ICAM1, VCAM1,
VEGFA, KDR, ANGPT1, ANGPT2, TEK, CD8A, CD8B, PRF1, GZMB, KLRK1, MICA,
MICB, HLA-A, HLA-B, B2M, TAP1, TAP2, CTSS, PDCD1, CD274, CTLA4, CD80, CD86,
LAG3, TIGIT, HAVCR2
3.2 P o ein-P o ein In e ac ions
The base ne wo k was gene a ed om he de ined a ge lis , o ming disc e e, module-speci ic clus e s.
STRING en ichmen p oduced an en iched ne wo k wi h dense in e connec i i y, indica ing ex ensi e
unc ional ela ionships. Hub analysis (deg ee >10) iden i ied he op 10 key nodes including TNF, TLR4,
CTLA4, STAT1, and CD274 - ep esen ing c i ical egula o s wi hin he ou e cascade.
6. Jain RK. No maliza ion o umo ascula u e: an eme ging concep in an iangiogenic he apy. Science.
2005 Jan 7;307(5706):58-62. doi: h ps://doi.o g/10.1126/science.1104819 . PMID: 15637262.
7. Ca melie P, Jain RK. P inciples and mechanisms o essel no maliza ion o cance and o he
angiogenic diseases. Na Re D ug Disco . 2011 Jun;10(6):417-27. doi: h ps://doi.o g/10.1038/n d3455 .
PMID: 21629292.
8. E ic Van Cu sem e al. Randomized Phase III T ial o Peg o hyalu onidase Al a Wi h Nab-Pacli axel
Plus Gemci abine o Pa ien s Wi h Hyalu onan-High Me as a ic Panc ea ic Adenoca cinoma. JCO 38,
3185-3194(2020). DOI: h ps://doi.o g/10.1200/JCO.20.00590
9. Nguyen KG, V abel MR, Man oo h SM, Hopkins JJ, Wagne ES, Gabaldon TA, Zaha o DA.
Localized In e leukin-12 o Cance Immuno he apy. F on Immunol. 2020 Oc 15;11:575597.
doi: h ps://doi.o g/10.3389/ immu.2020.575597 . PMID: 33178203; PMCID: PMC7593768.
10. Hong, Y., Robbins, Y., Yang, X., Mydla z, W. K., Sowe s, A., Mi chell, J. B., Gulley, J. L., Schlom,
J., Gamei o, S. R., Sie e s, C., & Allen, C. T. (2022). Cu e o syngeneic ca cinomas wi h a ge ed IL-12
h ough obliga e ep og amming o lymphoid and myeloid immuni y. JCI insigh , 7(5),
e157448. h ps://doi.o g/10.1172/jci.insigh .157448
11. Ghosn, M., Tselikas, L., Champia , S., Deschamps, F., Bonne , B., Ca e, É., Tes an, M., Danlos, F.
X., Fa hane, S., Susini, S., Suzzoni, S., Amma i, S., Ma abelle, A., & De Bae e, T. (2023). In a umo al
Immuno he apy: Is I Ready o P ime Time?. Cu en oncology epo s, 25(8), 857–
867. h ps://doi.o g/10.1007/s11912-023-01422-4
12. Mele o, I., Cas anon, E., Al a ez, M., Champia , S., & Ma abelle, A. (2021). In a umou al
adminis a ion and umou issue a ge ing o cance immuno he apies. Na u e e iews. Clinical
oncology, 18(9), 558–576. h ps://doi.o g/10.1038/s41571-021-00507-y
13. Pa doll D. M. (2012). The blockade o immune checkpoin s in cance immuno he apy. Na u e e iews.
Cance , 12(4), 252–264. h ps://doi.o g/10.1038/n c3239
14. Sha ma, P., & Allison, J. P. (2015). The u u e o immune checkpoin he apy. Science (New Yo k,
N.Y.), 348(6230), 56–61. h ps://doi.o g/10.1126/science.aaa8172
15. Wei, S. C., Du y, C. R., & Allison, J. P. (2018). Fundamen al Mechanisms o Immune Checkpoin
Blockade The apy. Cance disco e y, 8(9), 1069–1086. h ps://doi.o g/10.1158/2159-8290.CD-18-0367
16. Rosenbe g, S. A., Res i o, N. P., Yang, J. C., Mo gan, R. A., & Dudley, M. E. (2008). Adop i e cell
ans e : a clinical pa h o e ec i e cance immuno he apy. Na u e e iews. Cance , 8(4), 299–
308. h ps://doi.o g/10.1038/n c2355
17. Romee, R., Rosa io, M., Be ien-Ellio , M. M., Wagne , J. A., Jewell, B. A., Schappe, T., Leong, J.
W., Abdel-La i , S., Schneide , S. E., Willey, S., Neal, C. C., Yu, L., Oh, S. T., Lee, Y. S., Mulde , A.,
Claas, F., Coope , M. A., & Fehnige , T. A. (2016). Cy okine-induced memo y-like na u al kille cells
exhibi enhanced esponses agains myeloid leukemia. Science ansla ional medicine, 8(357),
357 a123. h ps://doi.o g/10.1126/sci anslmed.aa 2341
18. Robe s NJ, Zhang L, Janku F, Collins A, Bai RY, S aed ke V, Rusk AW, Tung D, Mille M, Roix J,
Khanna KV, Mu hy R, Benjamin RS, Helgason T, Sz alb AD, Bi d JE, Roy-Chowdhu i S, Zhang HH,
Qiao Y, Ka im B, McDaniel J, Elpine A, Saho a A, Lachowicz J, Phillips B, Tu ne A, Klein MK, Pos
G, Diaz LA J , Riggins GJ, Papadopoulos N, Kinzle KW, Vogels ein B, Be egowda C, Huso DL,
Va e asian M, Saha S, Zhou S. In a umo al injec ion o Clos idium no yi-NT spo es induces an i umo
esponses. Sci T ansl Med. 2014 Aug 13;6(249):249 a111. doi:
h ps://doi.o g/10.1126/sci anslmed.3008982. PMID: 25122639; PMCID: PMC4399712.
19. Filip Janku, Halle Huihong Zhang, Abdulmohammad Pezeshki, Sanjay Goel, Ra i Mu hy, And ea
Wang-Gillam, Dale R. Shepa d, Tho unn Helgason, Tyle Mas e s, Da id S. Hong, Sa ina A. Piha-
Paul, Daniel D. Ka p, Ma k Klang, S e en Y. Huang, Di ya Sakamu i, Anjali Raina, Jean
To isi, S ephen B. Solomon, Alice Weiss eld, E nes T e ino, Ga y DeC escenzo, Amanda
Collins, Ma ia Mille , Jenni e L. Sals om, Ronald L. Ko n, Linping Zhang, Sau abh Saha, Alexey A.
Leon o ich, Da id Tung, B en K eide , Ma y Va e asian, Khashaya sha Khazaie, M inal M. Gounde ;
In a umo al Injec ion o Clos idium no yi-NT Spo es in Pa ien s wi h T ea men - e ac o y Ad anced
Solid Tumo s. Clin Cance Res 1 Janua y 2021; 27 (1): 96–106. h ps://doi.o g/10.1158/1078-0432.CCR-
20-2065
20. S.G. Ahmed, G. Oli a, M. Shao, X. Wang, J.J. Mekalanos, & G.J. B enne , In a umo al injec ion o
schwannoma wi h a enua ed Salmonella yphimu ium induces an i umo immuni y and con ols umo
g ow h, P oc. Na l. Acad. Sci. U.S.A. 119 (24) e2202719119, h ps://doi.o g/10.1073/pnas.2202719119
(2022)
21. Ba ula S, Papas oi sis G, Kau man HL, e al. In a umo al aluminum hyd oxide–ancho ed IL-12
d i es po en an i umo ac i i y by emodeling he umo mic oen i onmen . JCI Insigh 2023; 8.
doi:h ps://doi.o g/10.1172/jci.insigh .168224
22. Aglia di, G., Liuzzi, A.R., Ho black, A. e al. In a umo al IL-12 deli e y empowe s CAR-T cell
immuno he apy in a p e-clinical model o glioblas oma. Na Commun 12, 444
(2021). h ps://doi.o g/10.1038/s41467-020-20599-x
23. Ta hini, A. A., E oglu, Z., Eljilany, I., Zage , J. S., Gonzalez, R. J., Sa naik, A. A., C use, C. W.,
Khushalani, N. I., De Aquino, D. B., Ab aham, E., Ace edo, D. M., Richa ds, A., Schell, M. J., Kalos, D.,
Chen, P. L., Messina, J. L., Can on, D. A., & Sondak, V. K. (2024). Neoadju an In a umo al Plasmid
IL-12 Elec o-Gene-T ans e and Ni olumab in Pa ien s wi h Ope able, Loco egionally Ad anced
Melanoma. Clinical cance esea ch : an o icial jou nal o he Ame ican Associa ion o Cance
Resea ch, 30(23), 5333–5341. h ps://doi.o g/10.1158/1078-0432.CCR-24-2768
24. Mackensen, A., Meidenbaue , N., Vogl, S., Laume , M., Be ge , J., & And eesen, R. (2006). Phase I
s udy o adop i e T-cell he apy using an igen-speci ic CD8+ T cells o he ea men o pa ien s wi h
me as a ic melanoma. Jou nal o clinical oncology : o icial jou nal o he Ame ican Socie y o Clinical
Oncology, 24(31), 5060–5069. h ps://doi.o g/10.1200/JCO.2006.07.1100
25. D o ako a, T., Finisgue a, V., Fo men i, M. e al. Enhanced umo esponse o adop i e T cell
he apy wi h PHD2/3-de icien CD8 T cells. Na Commun 15, 7789
(2024). h ps://doi.o g/10.1038/s41467-024-51782-z
26. Mele o, I., Ochoa, M. C., Molina, C., Sanchez-G ego io, S., Ga asa, S., Lu i-Rey, C., He as-S ubbs,
S., Casa es, N., Elizalde, E., Gomis, G., Ci ella, A., Be aondo, P., Teijei a, A., & Al a ez, M. (2023).
In a umo al co-injec ion o NK cells and NKG2A-neu alizing monoclonal an ibodies. EMBO molecula
medicine, 15(11), e17804. h ps://doi.o g/10.15252/emmm.202317804
27. Sakamo o, N., Ishikawa, T., Koku a, S. e al. Phase I clinical ial o au ologous NK cell he apy using
no el expansion me hod in pa ien s wi h ad anced diges i e cance . J T ansl Med 13, 277
(2015). h ps://doi.o g/10.1186/s12967-015-0632-8
28. Willsmo e, Z. N., Coumbe, B. G. T., C escioli, S., Reci, S., Gup a, A., Ha is, R. J., Chenowe h, A.,
Chauhan, J., Bax, H. J., McC aw, A., Cheung, A., Osbo n, G., Ho mann, R. M., Nakamu a, M., Laddach,
R., Geh, J. L. C., MacKenzie-Ross, A., Healy, C., Tsoka, S., Spice , J. F., … Ka agiannis, S. N. (2021).
Combined an i-PD-1 and an i-CTLA-4 checkpoin blockade: T ea men o melanoma and immune
mechanisms o ac ion. Eu opean jou nal o immunology, 51(3), 544–
556. h ps://doi.o g/10.1002/eji.202048747
29. Ro e, A. Combina ion o CTLA-4 and PD-1 blocke s o ea men o cance . J Exp Clin Cance
Res 38, 255 (2019). h ps://doi.o g/10.1186/s13046-019-1259-z
30. Zhao, H., Huang, S., Wu, J., Lu, Y., Zou, Y., Zeng, H., Li, C., Wang, J., Zhang, X., Duan, S., & Liang,
W. (2025). E icacy and sa e y o i s -line PD-1/PD-L1 inhibi o in combina ion wi h CTLA-4 inhibi o
in he ea men o pa ien s wi h ad anced non-small cell lung cance : a sys emic e iew and me a-
analysis. F on ie s in immunology, 16, 1515027. h ps://doi.o g/10.3389/ immu.2025.1515027
31. P o enzano, P. P., Cue as, C., Chang, A. E., Goel, V. K., Von Ho , D. D., & Hingo ani, S. R. (2012).
Enzyma ic a ge ing o he s oma abla es physical ba ie s o ea men o panc ea ic duc al
adenoca cinoma. Cance cell, 21(3), 418–429. h ps://doi.o g/10.1016/j.cc .2012.01.007
32. Jacobe z, M. A., Chan, D. S., Neesse, A., Bapi o, T. E., Cook, N., F ese, K. K., Feig, C., Nakagawa,
T., Caldwell, M. E., Zecchini, H. I., Lolkema, M. P., Jiang, P., Kul i, A., Thompson, C. B., Mane al, D.
C., Jod ell, D. I., F os , G. I., Shepa d, H. M., Skeppe , J. N., & Tu eson, D. A. (2013). Hyalu onan
impai s ascula unc ion and d ug deli e y in a mouse model o panc ea ic cance . Gu , 62(1), 112–120.
h ps://doi.o g/10.1136/gu jnl-2012-302529
33. Di Modugno, F., Colosi, C., T ono, P., An onacci, G., Ruocco, G., & Nis icò, P. (2019). 3D models in
he new e a o immune oncology: ocus on T cells, CAF and ECM. Jou nal o expe imen al & clinical
cance esea ch : CR, 38(1), 117. h ps://doi.o g/10.1186/s13046-019-1086-2
34. Nelson, Blessie Elizabe h; Janku, Filip; Fu, Siqing; Dumb a a, Eca e ina E.; Hong, Da id S.; Ka p,
Daniel D.; e al. (2025). Da a om Phase Ib S udy o Pemb olizumab in Combina ion wi h In a umo al
Injec ion o Clos idium no yi-NT in Pa ien s wi h Ad anced Solid Tumo s. Ame ican Associa ion o
Cance Resea ch. Collec ion. h ps://doi.o g/10.1158/1078-0432.c.8034696
