Safe eradication of large established tumors using neovasculature‐targeted tumor necrosis factor‐based therapies

• DOI: 10.15252/emmm.201911223
• Publication Date: 2020-01-08T09:52:51Z
• Authors: Leander Huyghe; Alexander Van Parys; Anje Cauwels; Sandra Van Lint; Stijn De Munter; Jennyfer Bultinck; Lennart Zabeau; Jeroen Hostens; An Goethals; Nele Vanderroost; Annick Verhee; Gilles Uzé; Niko Kley; Frank Peelman; Bart Vandekerckhove; Peter Brouckaert; Jan Tavernier
• Keywords: INTERFERON; EXPRESSION; Medicine (General); [SDV]Life Sciences [q-bio]; tumor necrosis factor; TNF; Mice, Transgenic; QH426-470; tumor necrosis; neovasculature; Mice; R5-920; Neoplasms; AMINOPEPTIDASE N; Medicine and Health Sciences; Genetics; cancer; Animals; Pharmacology & Drug Discovery; COMBINATION; IFN-GAMMA; RECEPTOR; Tumor Necrosis Factor-alpha; FACTOR; Biology and Life Sciences; FACTOR-ALPHA; Articles; ISOLATED LIMB PERFUSION; targeted therapy; ENDOTHELIAL-CELLS; 3. Good health; interferons; Mice, Inbred C57BL; factor; Immunotherapy

Article8 January 2020Open Access Source DataTransparent process Safe eradication of large established tumors using neovasculature-targeted tumor necrosis factor-based therapies Leander Huyghe Cytokine Receptor Laboratory, VIB Center for Medical Biotechnology, Department Biomolecular Medicine, Ghent University, Ghent, Belgium Search more papers by this author Alexander Van Parys BelgiumShared authorship Anje Cauwels Sandra Lint Stijn De Munter Clinical Chemistry, Microbiology and Immunology, Jennyfer Bultinck Lennart Zabeau Orionis Biosciences, Boston, MA, USA Jeroen Hostens Biomedical Molecular Biology, An Goethals Nele Vanderroost Annick Verhee Gilles Uzé CNRS UMR 5235, University Montpellier, France Niko Kley Frank Peelman Bart Vandekerckhove Peter Brouckaert Jan Tavernier Corresponding Author [email protected] orcid.org/0000-0002-7609-6462 Information Huyghe1, Parys1, Cauwels1, Lint1, Munter2, Bultinck1,7, Zabeau3, Hostens4,8, Goethals4,9, Vanderroost1, Verhee1, Uzé5, Kley3, Peelman6, Vandekerckhove2, Brouckaert4 *,1,3 1Cytokine 2Department 3Orionis 4Department 5CNRS 6VIB 7Present address: Oxyrane, 8Present PerkinElmer, Zaventem, 9Present Genae, Antwerp, *Corresponding author. Tel: [phone]; Fax: 9340; E-mail: EMBO Mol Med (2020)12:e11223https://doi.org/10.15252/emmm.201911223 See also: T Kammertoens et al (February 2020) PDFDownload PDF article text main figures. Peer ReviewDownload a summary the editorial decision including letters, reviewer comments responses to feedback. ToolsAdd favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Abstract Systemic toxicities have severely limited clinical application factor (TNF) as an anticancer agent. Activity-on-Target cytokines (AcTakines) are novel class immunocytokines with improved therapeutic index. A TNF-based AcTakine targeted CD13 enables selective activation neovasculature without any detectable toxicity in vivo. Upregulation adhesion markers supports enhanced T-cell infiltration leading control or elimination solid by, respectively, CAR cells combination therapy CD8-targeted type I interferon AcTakine. Co-treatment CD13-targeted II leads very rapid destruction complete regression large, tumors. As no needed, safe efficacious broad range types becomes feasible. Synopsis TNF IFN-γ great potential but due side-effects. In present study, endothelium is identified target cell their antitumor activity new biologics that allow targeting developed. By transgenic mouse technologies was effect IFN-γ. AcTakines were developed fusing inactivated cytokine mutants single domain antibody (resp. CD13-AFR CD13-AFN-II), allowing endothelium. synergized immunotherapies (CD8-AFN T-cells) immune-mediated killing tumors, Combined treatment CD13-AFN-II resulted apoptosis destruction, Introduction Tumor can cause hemorrhagic both animal models patients. Its use, however, impeded life-threatening side-effects, i.e., systemic inflammatory response syndrome may lead shock organ failure. studies, doses up 50 times below estimated effective dose already caused serious close maximum tolerated produced only minimal (Roberts al, 2011). The current use therefore advanced cancers limbs via isolated limb perfusion (ILP). high rates achieved ILP (2–6 mg) melphalan clearly show drug (Eggermont 2003; Lejeune 2006; van Veenendaal 2017). exact mechanism underlying still poorly understood several lines evidence point critical role TNF-R1 receptor complex on (Van de Wiel 1989; Ruegg 1998; Stoelcker 2000; 2006). One strategy reduce fusions moieties (e.g., fragments, single-domain antibodies peptides) direct specific tissue, so index increased lowering (List Neri, 2013). Following strategy, fusion proteins NGR-TNF, marker aminopeptidase N/CD13, L19-TNF, oncofetal splice variant fibronectin, currently development mesothelioma melanoma, respectively (Danielli 2015a; Gregorc 2018). However, affinity binding, which intrinsic wild-type (wt) cytokines, inevitably entails risk especially case pleiotropic cytokines. On top, portion wt drugs will be captured ubiquitously expressed receptors before they reach (the so-called "sink effect" (Tzeng 2015)) and, TNF, also soluble circulation. To avoid such unwanted effects, we engineered (AcTakines), replaced mutant version strongly reduced its receptor, fused moiety binds cell-specific surface marker. Consequently, inactive way target, thereby eliminating unfavorable side sink regain full local avidity-driven binding (Garcin 2014). We previously reported remarkable efficacy dendritic cell-targeted (AcTaferon, AFN) multiple models, side-effects (Cauwels 2018a,b). now demonstrate vasculature-targeted AcTafactor, AFR; AcTaferon-II, AFN-II) selectively activate kill endothelial cells. AFR strikingly either CD8-AFN immunotherapy, human AFN-II therapy, mice toxicity. Results sufficient induce investigate importance made having conditional reactivation allele (p55cneo/cneo mice) expression restored upon Cre-mediated excision inhibitory floxed Neo cassette (Victoratos Crossing p55cneo/cneo Flk1Cre yielded express functional levels (Fig 1C). Daily perilesional (p.l.) C57BL/6 bearing B16Bl6 melanoma 7 μg mTNF within 2–3 days extensive 1B). similar observed deleter Cre (DelCre) p55lox/lox mice. Despite TNF-R1-expressing tumor, completely absent TNF-R1−/− Conversely, when carrying TNF-unresponsive (B16-dnTNF-R1) treated parental 1A). Together these data unequivocally host-mediated. Significantly, identical induced mice, indicating signaling through effect. Figure 1. vasculature not shock-inducing inoculated 6 × 105 dnTNF-R1 day 0 daily p.l. from 10. growth shown mean size (TSI) + SEM (n = 6). line under graph represents period. after C57BL/6J (WT), TNF-R1−/−, knockout (p55cneo/cneo), (Flk1Cre) all (DelCre). TSI WT Flk1Cre, 8 DelCre, 10 p55cneo/cneo). PCR analysis detection wild-type, (cneo) reactivated (lox) whole tissues, lung (LECs), single-cell suspension depleted CD31+ Toxicity i.v. bolus injection indicated mTNF. Mean rectal body temperature cumulative survival 4 μg; DelCre μg, 15 20 5 other groups). For continuity graphs, dead included 20°C. available online figure. Data 1 [emmm201911223-sup-0003-SDataFig1.xlsx] Download figure PowerPoint safety activity, used model. Intravenous (i.v.) lethal causes drastic drop temperature, inflammation, shock, eventually death 12–72 h. naive LD50 LD100 1D). expected, resistant sensitivity slightly 25 reflecting incomplete differences genetic background. stark contrast, TNF-induced 250 even causing temperature. focusing potentially therapy. AcTafactors target-specific delivery typically consists three parts: complex, flexible linker, VHH-type 2A). Given homotrimeric nature opted single-chain (scTNF), short C- N-terminal linkers between monomers (Krippner-Heidenreich 2008). This linkage precludes vivo monomer exchange wtTNF. generate (AFRs), evaluated panel scTNF VHH directed against CD20. Y86F mutation biological L929 cytotoxicity assay about 000-fold 2B). Strikingly, mCD20-AFR largely recovered assayed stably expressing mCD20. Such non-targeted BcII10-AFR. Similarly, generated Y87F hCD20. When MCF7 cells, hCD20-AFR displayed 280-fold reduction cytotoxic while MCF7-hCD20 nearly recovery 2C). physiological setting, next primary umbilical vein (HUVECs) hCD13, naturally target. HUVECs susceptible death, NF-κB-dependent IL-8 secretion readout. At concentrations ng/ml, had virtually HUVECs, hCD13-targeted significant 2D). Interestingly, at concentration hCD13-AFR unable level wtTNF, suggesting cascade initiated might differ wtTNF (see below). 2. AFRs mediate A. Schematic representation AFR, consisting 20xGGS linker (e-TNF) C-terminal tag. B, C. Viability mCD20-expressing 72-h stimulation sc BcII10 mCD20 (B) viability hCD20-expressing hTNF hCD20 (C). Cell measured ATP luminescence assay. Each replicates, error bars SEM. D. HUVEC supernatant 24-h stimulus, ELISA. Error ns, non-significant; **P < 0.01; ***P 0.001 one-way ANOVA Bonferroni's comparison test. 2 [emmm201911223-sup-0004-SDataFig2.xlsx] combine evaluate toxicity, first model BcII10-AFR effects. It Krippner-Heidenreich (2008) decreased compared trimeric Indeed, scTNF, led moderate 3A), circulating IL-6, sensitive 3B). Thirty-five microgram (1.75 mg/kg) characterized dramatic concomitant increase IL-6. BcII10-AFR, consecutive injections 200 (10 non-toxic did confirming untargeted inactive. 3. Sc mCD13-AFR active (↑) naïve C56BL/6 First i.v., i.p. ± 4). Plasma IL-6 Blood samples taken h Y86F. Values individual shown. non-significant Immunohistochemical staining PECAM-1 (red), ICAM-1 (green), DNA (blue) mCD13-AFR. Scale bar μm. Expression ICAM-1, E-selectin, VEGF-R2 treatment, qPCR RNA. *P 0.05; weight change (equimolar) mCD13-AFR, (BcII10-AFR) mCD13 (CD13-hIFNα2) TSI, 5). two-way Wortmannin (Wm), Birinapant (Bir), combinations thereof. given CD13-AFR. Wm Bir, groups 17 inoculation. *** 3 [emmm201911223-sup-0005-SDataFig3.zip] vasculature, (Curnis Pasqualini 2000). immunocytokine mCD13-targeted efficient than VHH. AFRs, performed immunohistochemical sections 3C). Six hours injection, clear induction vessels observed, pronounced 24 prolonged activation. Induction confirmed RNA 3D). Quite E-selectin repression injection. Since tumor-bearing tend sensitized toward 2018b), tested contrast loss apparent 3E). highly significant, neither nor protein hIFNα2 (which does bind IFNAR) higher doses, appeared vascular disruption same extent investigated whether inhibition pro-survival could Pretreatment Smac-mimetic broad-spectrum PI3K inhibitor wortmannin resulting necrosis, resembling 3F G). Taken together, suggest favored pro-inflammatory beneficial treatments chemo- immunotherapy. Vasculature-targeted synergizes immunotherapy AFR-induced would suffice synergize based has promising certain leukemias lymphomas, least partly poor (D'Aloia Therefore, mCD13-AFR-mediated facilitate model, hCD70+ SKOV3 ovarian cancer hCD70 4A) NSG Treatment itself growth, potentiated immediate stasis 4B C). hCD70-negative RL EV1D), underlining antigen specificity. FACS significantly (Figs 4D EV1A–C). 4. construct used, anti-hCD70 VHH, hCD8α hinge transmembrane (TM) domain, CD137 co-stimulatory CD3ζ domain. 106 12 (↑). 5) Flow cytometric hCD45+hCD3+eGFP+ 18 Mice injected PBS (↑ B). Two replicate EV1A gating strategy. E. 30 mCD8-AFN, F. 19 [emmm201911223-sup-0006-SDataFig4.xlsx] Click here expand EV1. increases specificity Gating flow PBMCs First, debris bulk gated out ("non-debris"), then, PI-positive removed ("live"), finally, doublets ("single cells"). 2.8 14 inoculation (C) 14, 15, 16, 25, 26, 28 43. Human hCD45 hCD3 double positive (upper panel). presence transgene eGFP (lower mCD8α-targeted AcTaferon (mCD8-AFN) mCD8-AFN itself, 4E