Abstract Background Anti-PD-1 antibodies have revolutionized cancer immunotherapy due to their ability to induce long-lasting complete remissions in a proportion of patients. Current research efforts are attempting to identify biomarkers and suitable combination partners to predict or further improve the activity of immune checkpoint inhibitors. Antibody-cytokine fusions are a class of pharmaceuticals that showed the potential to boost the anticancer properties of other immunotherapies. Extradomain A-fibronectin (EDA-FN), which is expressed in most solid and hematological tumors but is virtually undetectable in healthy adult tissues, is an attractive target for the delivery of cytokine at the site of the disease. Methods In this work, we describe the generation and characterization of a novel interleukin-7-based fusion protein targeting EDA-FN termed F8(scDb)-IL7. The product consists of the F8 antibody specific to the alternatively spliced EDA of FN in the single-chain diabody (scDb) format fused to human IL-7. Results F8(scDb)-IL7 efficiently stimulates human peripheral blood mononuclear cells in vitro. Moreover, the product significantly increases the expression of T Cell Factor 1 (TCF-1) on CD8+T cells compared with an IL2-fusion protein. TCF-1 has emerged as a pivotal transcription factor that influences the durability and potency of immune responses against tumors. In preclinical cancer models, F8(scDb)-IL7 demonstrates potent single-agent activity and eradicates sarcoma lesions when combined with anti-PD-1. Conclusions Our results provide the rationale to explore the combination of F8(scDb)-IL7 with anti-PD-1 antibodies for the treatment of patients with cancer. Keywords: Antibodies, Bispecific; CD8-Positive T-Lymphocytes; Combined Modality Therapy; Cytokines; Immune Checkpoint Inhibitors __________________________________________________________________ WHAT IS ALREADY KNOWN ON THIS TOPIC * Monoclonal antibodies targeting the PD-1/PD-L1 axis have demonstrated clinical activity in certain cancer types although a substantial proportion of patients still do not benefit from immune checkpoint blockade (ICB) treatment. From recent studies, expression of IL-7R and TCF-1 has emerged as predictive biomarkers associated with improved clinical outcomes for ICB treatment. WHAT THIS STUDY ADDS * Here, we describe the generation and characterization of a novel IL-7-based fusion protein targeting extradomain A (EDA-FN), termed F8(scDb)-IL7. * The product has shown upregulation of TCF-1 on human CD8+T cells in vitro and was able to selectively localize to solid tumors in vivo. * The combination of F8(scDb)-IL7 with PD-1 blockade was able to induce tumor regression or retardation in four distinct immunocompetent mouse models of cancer. HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY * The data presented in this study provide a rationale for exploring the combination of ICB with EDA-FN-targeted IL-7 in clinical trials. Background Monoclonal antibodies targeting the PD-1/PD-L1 axis have demonstrated clinical activity for the treatment of numerous cancer entities (including skin cancer, non-small cell lung cancer, Hodgkin lymphoma, head and neck cancer, renal cell carcinoma (RCC), urothelial carcinoma, colorectal cancer, hepatocellular carcinoma (HCC), esophageal carcinoma, gastric cancer, triple-negative breast cancer, malignant pleura mesothelioma[51]^1) and are broadly used, either alone or in combination with other therapeutic agents.[52]2,[53]4 PD-1 blockade functions by preventing the onset of the SHP-2 signaling pathway in activated T lymphocytes that occurs after the engagement of PD-1 with its ligands, PD-L1 or PD-L2.[54]^5 6 Yet, a substantial proportion of patients either do not respond (primary resistance) or respond only temporarily (acquired resistance) to immune checkpoint blockade (ICB) treatment.[55]^7 8 Primary resistance, which occurs in approximately 30% of advanced melanoma patients, may emerge as a result of immunosuppressive mechanisms in the tumor microenvironment (TME) distinct from the PD-1/PD-L1 axis.[56]^9 Furthermore, approximately 25% of responders develop acquired resistance which may be attributed to the onset of other immunosuppressive signaling receptors such as TIM-3 or LAG-3.[57]^10 These clinical findings have motivated efforts to discover predictive biomarkers for the efficacy of ICB.[58]^11 12 A study led by Caushi et al has identified phenotypic differences in tumor-specific tumor-infiltrating lymphocytes (TILs) of patients with resectable lung cancer after treatment with nivolumab. In patients enjoying a major pathological response (MPR) TILs not only had a lower expression of exhaustion markers (eg, PD-1, TIGIT, LAG-3) indicating a healthier, less terminally differentiated cell state but also showed a higher expression of memory markers such as interleukin-7 receptor (IL-7R) and transcription factor 1 (TCF1), compared with non-MPR patients.[59]^13 It has been hypothesized that molecular strategies aimed at increasing the concentration of interleukin-7 (IL-7) within the neoplastic mass may be associated with durable objective responses on ICB treatment.[60]14,[61]17 IL-7 is a member of the common gamma chain (γc) cytokine family involved in various processes that influence the differentiation of immune cells.[62]^18 By binding to the IL-7R, a heterodimeric complex consisting of the IL-7Rα and the γc chains, IL-7 triggers the JAK1/JAK3 intracellular signaling pathway. This pathway ultimately modulates the activation and inhibition of target genes involved in the proliferation, and functional enhancement of immune cells, notably of CD8+T cells known for their central role in immune responses against cancer. IL-7 is also known for its crucial involvement in forming long-lived memory T-effector cells.[63]^19 The formation of memory T-effector cells is key for sustaining robust and durable immune responses.[64]^20 In this respect, IL-7 may be superior to other cytokines, such as IL-2 or IL-15, which have been shown to promote exhaustion and terminal differentiation of T-cells after repeated administration to patients with cancer.[65]^21 22 Moreover, IL-7 may promote the formation of organized immune centers called tertiary lymphoid structures, which positively correlate with prolonged overall survival for certain malignancies.[66]^23 Recombinant human interleukin-7 (rhIL-7) has been studied as a single agent in several clinical trials, involving patients with melanoma, RCC cancer, colon carcinoma osteogenic sarcoma, and other refractory solid tumors ([67]NCT00091338, [68]NCT00492440, [69]NCT00062049).[70]^24 A maximum tolerated dose for IL-7 therapeutics has not been established with doses ranging from 3 to 120 µg/kg for the recombinant cytokine[71]^24 and 60 to 1440 µg/kg for Fc-fused variants.[72]^25 No serious adverse events have been reported in clinical trials. The most common toxicities were injection site reactions such as urticaria, edema, or itching sensation while in a small proportion of patients fatigue and febrile symptoms were observed. Both for rhIL-7 and the Fc-fused variant, a marked increase in peripheral CD4+, and CD8+lymphocytes was observed in a dose-dependent manner in subjects receiving ≥10 mg/kg/dose. A general strategy to enhance the therapeutic index of cytokines consists of their fusion with tumor-homing antibodies to concentrate the payload at the site of the disease, sparing normal tissues. Our group and other groups have extensively studied the anti-cancer properties of several antibody cytokine fusions, also termed immunocytokines,[73]^26 27 and some of these products have moved to clinical trials with encouraging results.[74]^28 29 Components of the modified extracellular matrix, such as splice isoforms of fibronectin (FN) containing the extra domain A (EDA) or the extra domain B (EDB) are particularly attractive targets for pharmaco-delivery strategies, as these antigens are expressed in most solid malignancies while being undetectable in most adult healthy tissues. Nuclear medicine studies in rodents and in patients have demonstrated the selective localization of radiolabelled preparations of antibodies against these targets to neoplastic lesions. This validation supports their use for the targeted delivery of cytokine payloads and other therapeutic agents[75]30,[76]32 Here, we report the generation and characterization of a novel IL-7-based fusion protein termed F8(scDb)-IL7. The product consists of the F8 antibody specific to the alternatively spliced EDA of FN[77]^33 in the single-chain diabody (scDb) format fused to human IL-7. The human cytokine payload was used for preclinical studies as it cross-reacts with the murine cognate receptor.[78]^34 F8(scDb)-IL7 has been shown to efficiently stimulate human peripheral blood mononuclear cells (hPBMCs) in vitro and to increase the expression of TCF1 to a higher extent than L19IL2, an IL2-based immunocytokine currently being investigated in late-stage clinical trials ([79]NCT02938299) ([80]NCT03567889) ([81]NCT04362722) ([82]NCT05329792).[83]^35 36 In the WEHI-164 sarcoma model, F8(scDb)-IL7 halted tumor progression in monotherapy and induced complete cancer remissions when combined with PD-1 blockade. Similar findings were confirmed in two other subcutaneous carcinomas and in an orthotopic glioma model. The results suggest that F8(scDb)-IL7, or a similar targeted version of this cytokine, may represent ideal combination partners for PD-1 blockade. Methods Cell lines All cell lines were received between 2018 and 2023, expanded and stored as cryopreserved aliquots in liquid nitrogen. CHO-S, F9, WEHI-164, and MC38 cells were obtained from ATCC. Cells were grown according to the supplier’s protocol and kept in culture for no longer than 10 passages. PowerCHO Serum-Free medium (Lonza BE12-776Q)+4 mM L-glutamine (Lonza BEBP17-605E) was used for CHO-S cells. DMEM (Gibco 11965)+10% Fetal Bovine Serum (FBS, Gibco 10270106) was used for F9 and MC38. F9 cells were grown on 0.1% gelatin-coated flasks (Sigma-Aldrich G1393). RPMI (Gibco 11875093)+10% FBS (Gibco 10270106) was used for WEHI-164. GL-261 iRFP720 cells were generated and maintained as described.[84]^29 Authentication of the cell lines, including checks of postfreeze viability, growth properties, and morphology test for mycoplasma contamination, isoenzyme assay, and sterility test, was performed by the cell bank before shipment. Healthy donor buffy coats were acquired from the Zurich blood donation service (Blutspende Zurich, Switzerland). PBMCs were enriched by density gradient centrifugation (Ficoll-Paque Plus, GE Healthcare, Chicago, Illinois, USA). Purified T-cells were cryopreserved in freezing media consisting of 90% FBS supplemented with 10% Dimethylsulfoxide (DMSO) and stored in liquid nitrogen at −196°C. Cloning, expression, and protein purification The fusion protein F8(scDb)-IL7 contains the F8 antibody in single chain Diabody (scDb) format fused to human IL-7 to the C-terminus of the variable light (V[L]) chain by an 18 amino acid long linker (GGGGSGGGS)[2]. The gene encoding the F8 antibody in scDb format was PCR amplified between the NheI/EcoRI restriction sites while the gene encoding human IL-7 was PCR amplified between the EcoRI/NotI restriction sites. The genes encoding F8(scDb) and hIL-7 were PCR assembled and cloned into the mammalian expression vector pcDNA3.1(+) by a NheI/NotI restriction site. Human IL-7 is cross-reactive from mouse to man which guided our choice for preclinical studies with respect to the murine variant.[85]^34 37 All fusion proteins were expressed by transient gene expression (TGE) in CHO-S cells and purified with proteinA sepharose (Sino Biological) as described in Corbellari et al [86]^38 or by cOmplete His-Tag Purification Resin (Merck 5893682001) as described in Plüss et al.[87]^39 Cloning strategies for the L19-based prototypes are described in [88]online supplemental material. Biodistribution experiments In order to validate the specific accumulation of the immunocytokines in tumor tissues, 129/SvEv mice were inoculated subcutaneously into the right flank with 10^7 F9 tumor cells. When tumors reached a volume of 100–200 mm^3, 100 µg of L19(scDb)-IL7, F8(scDb)-IL7, L19-IgG[4]-IL7 or L19-IgG[4]-KIH-IL7 were radiolabeled with ^125I and chloramine T, filtered on a PD10 column and injected into the lateral tail vein.[89]^33 Mice were euthanized 24 hours after injection. Organs, blood, and tumors were weighed, and radioactivity was detected using a Packard Cobra gamma counter. The immunocytokine uptake in blood, organs, and tumors was calculated and expressed as the percentage of the injected dose per gram of tissue (%ID/g±SEM, n=4). Data were adjusted for tumor growth.[90]^40 Antigen expression study on human tissue microarray and on murine tumor models Immunofluorescence analysis was performed on human tissue microarray (TMA) purchased from Amsbio (catalog no T6235700–5, lot no B712100). Before staining, samples were fixed in ice-cold acetone. Sections were stained with F8(scDb)-IL7 (specific to EDA) and L19(scDb)-IL7 (specific to EDB) at a concentration of 5 µg/mL in 2% bovine serum albumin/PBS. A rat anti-human IL-7 IgG was used as a secondary antibody (BioLegend BVD10-40F6) and a donkey anti-rat-Alexa Fluor488 (Invitrogen; A21208) was used for detection. Tissue vasculature was visualized by a mouse anti-human-CD31 (Invitrogen; 14-0319-82) and detected using a goat anti-mouse Alexa Fluor594 (Invitrogen; A11032). Slides were mounted with a fluorescence mounting medium (Dako Agilent) and imaged with the Axio Scan.Z1 (Zeiss) slide scanner at 20×magnification using the ZEN blue software. Tumor region annotation and pixel quantification were performed with QuPath (0.3.0) using object classification.[91]^41 EDA and EDB expression was assessed on ice-cold acetone-fixed 10 µm cryostat sections of F9, WEHI-164, GL261 and MC38 tumors stained with F8 or L19 IgG1 and revealed with goat anti-human-Alexa Fluor488 (Invitrogen; A11013). Blood vessels were stained with a rat anti-mouse-CD31 (BD; 550274) and revealed with donkey anti-rat-Alexa Fluor549 (Invitogen; A21209). Slides were mounted with a fluorescent mounting medium (Dako) and analyzed with a wide-field Leica TIRF microscope using Leica LAS X Life Science Microscope Software. Proliferation assay on hPBMCs The biological activity of F8(scDb)-IL7 was evaluated by a proliferation assay on hPBMCs. After thawing, cells were transferred into RPMI 1640 medium with 10% FBS and 10 µg/mL PHA-P (sigma, L1668) and incubated for 4 days at 37°C 5% CO[2]. Precoating of the wells with EDA was not performed. After washing in RPMI 1640 medium with 10% FBS, cell density was adjusted to 1.0×10^5 per well. Serial dilutions ranging from 4 nM to 0.2 pM of either rhIL-7 (ACROBiosystems, Cat. No. IL7-H4219) or F8(scDb)-IL7 were prepared and added to the 96-well plate. Cells were incubated for 6 days at 37°C 5% CO[2]. At day 6, 100 mL of CellTiter-Glo Reagent (Promega, cat#: G7572) was added to each well, and cells were incubated for an additional 10 min to stabilize the luminescent signal. For each well, 180 µL was transferred to a 96-well black polystyrene high bind stripwell and luminescence was measured using a Tecan microplate reader (Infinite M200 Pro, Maennedorf, Switzerland). IFNγ release assay on hPBMCs To further characterize the biological activity of F8(scDb)-IL7 or KSF(scDb)-IL7, an IFNγ release assay was performed on hPBMCs. Precoating of the wells with 0.1 µM EDA was performed overnight at 4°C. The following day, 100 µL of cell suspension (1.0×10^5 cells/well) was incubated for 72 hours at 37°C and 5% CO2 with a serial dilution of F8(scDb)-IL7. Cultured supernatants were analyzed by a sandwich ELISA according to the manufacturer protocol ELISA MAX Deluxe Set Human IFNγ (Biolegend, 430104). Absorbance was measured at 450 nm and 570 nm. Relative absorbance was converted to IFNγ (pg/mL) with the respective calibration curve. Flow cytometry experiment on hPBMCs Freshly thawed hPBMCs were co-cultured in a 96-well plate with either F8(scDb)-IL7 or L19IL2 at high, medium, and low concentrations (1000 nM, 250 nM and 62.5 nM for F8(scDb)-IL7 and 100 nM, 50 nM and 10 nM for L19IL2) or in medium alone. Cells were incubated for 72 hours before staining. Precoating of the wells with EDA was not performed. The same experiment was repeated by coculturing freshly thawed hPBMCs with CD3/CD28 Dynabeads at a 1:1 ratio (ThermoFischer) and either rIL-7 or rIL2, in complete Advanced RPMI (1% P/S, 10%FBS). 24 hours later, the beads were removed and the cells were placed in complete Advanced RPMI. Cells were incubated for 5 days at 37°C and 5% CO[2] prior to staining with the fusion proteins F8(scDb)-IL7 or L19IL2 or in medium alone. A detailed list of antibodies used for flow cytometry can be found in ([92]online supplemental table S1). The acquisition was performed on a BD FACSVerse Analyzer or LSR II Fortessa 4 L (BD), and data were analyzed with FlowJo (Tree Star). Subcutaneous tumor models Tumor cells were implanted subcutaneously into the right flank with 5×10^6 cells (WEHI-164) or 1×10^6 cells (MC38) or 1.5×10^7 cells (F9). When tumors reached a suitable volume (around 100 mm^3), mice were randomized and intravenously injected with 130 µg of the immuno-cytokine preparation dissolved in saline solution, also used as the negative control, every second day for three times. In the combination groups, mice first received 200 µg of mouse anti-PD-1 (BioXcell, clone 29F.1A12, cat BE0273), followed by 130 µg of F8(scDb)-IL7, 24 hours later. Survivors were reinoculated in the left flank for rechallenge studies with 5×10^6 cells WEHI-164. Subcutaneous tumors presenting ulceration and/or exceeding 15 mm in length and/or width were sacrificed as an endpoint. Splenomegaly and IFNγ levels in plasma For quantitative analysis of splenomegaly and IFNγ in plasma, WEHI-164 tumor-bearing mice were euthanized 24 hours after the last injection of F8(scDb)-IL7, anti-PD-1, both or saline solution. Immediately after sacrificing, the mouse blood was collected from the heart and spleens were harvested. For each group, spleens were weighted on a microscale and the value was collected for quantitative comparative assessment of splenomegaly. Blood was left to clot for 1 hour at room temperature in microtubes containing lithium heparin (BD Microtainer Tube). Plasma was collected after centrifugation and a sandwich ELISA was performed to quantify IFN-g levels in blood according to ELISA MAX Deluxe Set mouse IFNγ (Biolegend, 430804). Absorbance was measured at 450 nm and 570 nm. Relative absorbance was converted to IFNγ (pg/mL) with the respective calibration curve. Immunofluorescence infiltrate study For ex vivo infiltrate immunofluorescence analysis, mice were injected according to the therapy schedule and euthanized 24 hours after the last injection. Tumors were excised and embedded in a cryoembedding medium (ThermoScientific, Waltham, Massachusetts, USA) and the corresponding cryostat tissue sections (8–10 µm thickness) were stained using the reagents described in [93]online supplemental table S2. Slides were mounted with a fluorescent mounting medium (Dako Agilent, Carpinteria, California, USA) and analyzed with a wide-field Leica TIRF microscope using the Leica LAS X Life Science Microscope Software. Channels acquired were merged using Image J software.[94]^42 Proteomic analysis of spleens and tumors from treated WEHI164-bearing BALB/c mice WEHI-164 tumor-bearing mice were treated with either Saline, αPD-1 200 µg, F8(scDb)-IL7 130 µg three times every 48 hours or with αPD-1 200 µg in combination with F8(scDb)-IL7 130 µg and sacrificed 48 hours after the last injection. Spleens and tumors were harvested, snap-frozen, processed, and analyzed as presented in [95]online supplemental material. Intracranial orthotopic glioma models C57BL/6 (#SC-C57N-F) mice of 6–12 weeks of age were purchased from Janvier Labs and 4×10^4 GL-261 iRFP720 cells were stereo-tactically implanted into the right striatum. Mice were observed daily for the development of symptoms and treated as follows: F8(scDb)-IL7 (130 µg) or/and αPD-1 (200 µg) were intravenously injected every second day for three times. For the combination group, αPD-1 was injected first and 24 hours later an injection of F8(scDb)-IL7 followed. Mice reaching a BW loss ≥15% as a symptom attributed to the tumor burden were sacrificed as an endpoint. Experimental animals A total of 73 female BALB/c, 44 C57BL/6 and 44 129/SvEv mice, aged 8 weeks with an average weight of 20 g, were used in this work. Mice were purchased from Janvier (France) and raised in a pathogen-free environment with a relative humidity of 40%–60%, at a temperature between 18°C and 26°C and with daily cycles of 12 hours light/darkness according to GV-SOLAS; FELASA. Animals were kept in groups of 5 or fewer mice per cage and were reallocated in case of single housing to another cage. Blinding of the experimental groups was not performed; animals were enrolled in experimental groups according to their tumor volume (ie, when tumors reached a volume between 80 and 110 mm^3). Mice were monitored daily; tumor volume was measured with a caliper (volume=length×width×0.52). Results Production and characterization of IL7-based fusion proteins Both the alternatively spliced EDA and EDB domains of FN, recognized by the F8 and by the L19 antibodies,[96]^33 43 respectively, were considered for the production of tumor-homing fusion proteins. Three fusion proteins based on the L19 antibody were fused to human IL-7. The first molecule, referred to as L19(scDb)-IL7, the L19 antibody in single-chain diabody format was fused to human IL-7 through a 20-amino acid long linker at the C-terminus ([97]online supplemental figures S1A and S2A). A second molecule, termed L19(IgG4)KIH-IL7, featured point mutations on the CH[3] heavy chain to favor heterodimerization through knob-into-hole interaction ([98]online supplemental figures S1B and S2B). In a third molecule, termed L19(IgG4)-IL7, human IL-7 was fused at the C terminus of the L19 antibody in IgG format, leading to a bivalent cytokine display ([99]online supplemental figures S1C and S2C). In order to allow for a comparison between EDA and EDB, F8(scDb)-IL7 ([100]online supplemental figure S1D) was generated by using the F8 moiety in the same single-chain diabody format used for L19(scDb)-IL7. KSF(scDb)-IL7, a fusion protein with an identical format but specific to hen egg lysozyme,[101]^44 was also produced and used as negative control of irrelevant specificity in the mouse ([102]online supplemental figures S1E and S2D). In vivo quantitative biodistribution for preclinical studies The tumor-homing properties of radioiodinated preparations of IL-7-based fusion proteins were evaluated by quantitative biodistribution experiments in immunocompetent mice bearing F9 teratocarcinomas. Among the various formats assessed, L19(scDb)-IL7 exhibited the most favorable profile in terms of percent injected dose per gram (%ID/g) of tumor, as well as tumor-to-organ and tumor-to-blood ratios ([103]figure 1A). L19(IgG4)KIH-IL7 revealed a comparable %ID/g to L19(scDb)-IL7, however, had a higher accumulation in the liver ([104]figure 1B). Similarly, L19(IgG4)-IL7 exhibited elevated liver values, along with poor tumor accumulation compared with other variants ([105]figure 1C). This experiment indicated that the single-chain format offered superior biodistribution profiles compared with IgG-based molecules. Consequently, this format was selected for further characterization. Figure 1. Tumor homing properties of IL-7-based immunocytokines. Quantitative biodistribution analysis of ^125I radio labeled L19(scDb)-IL7 (A), L19 IgG4 KIH-IL7 (B), L19 IgG4-IL7 (C), F8(scDb)-IL7 (D) and KSF(scDb)-IL7 (E) in immunocompetent mice bearing F9 teratocarcinoma tumors. About 10 µg of radio-labeled fusion protein was injected into the lateral tail vein and mice were sacrificed 24 hours after injection. Organs and tumor were excised, weighed and the radioactivity of each sample was measured. Results are corrected on tumor growth and expressed as a percentage of injected dose per gram of tissue (% ID/g±SEM), (n=5 for F8(scDb)-IL7; n=4 for L19 IgG4 KIH-IL7 and KSF(scDb)-IL7; n=3 for L19(scDb)-IL7 and L19 IgG4-IL7). A schematic representation of each fusion protein is depicted. (F) Microscopic fluorescence analysis of EDA and EDB expression (green) in F9, WEHI-164, MC38 and GL261 tumors. An anti-CD31 antibody was used to stain blood vessels (red). 20×magnification, scale bar=100 µm. EDA, extra domain A. [106]Figure 1 [107]Open in a new tab In the second experiment, F8(scDb)-IL7 was compared with KSF(scDb)-IL7, which was used as negative control. In keeping what had previously been reported for other immunocytokines directed to splice variants of FN, F8(scDb)-IL7 demonstrated a remarkable tumor uptake, with a tumor-to-blood ratio of 9.2 at 24 hours ([108]figure 1D). By contrast, the KSF homolog did not exhibit a preferential uptake in the neoplastic mass ([109]figure 1E). Antigen expression analysis on murine and human cancer sections EDA and EDB expression was evaluated in F9, WEHI-164, MC38 and GL-261 tumor sections. All tumor models expressed comparable levels of the two antigens ([110]figure 1F). To refine the selection of our clinical candidate, we conducted an extensive analysis of EDA and EDB expression using a human tissue microarray stained with F8(scDb)-IL7 and L19(scDb)-IL7. High level of EDA and EDB expression was detected in all analyzed malignancies while both antigens were almost undetectable in most normal adult tissues ([111]figure 2 and [112]online supplemental figures 3 and 4). Based on these findings, we focused our investigations on the fusion protein F8(scDb)-IL7. Figure 2. Antigen expression analysis of human cancer sections. Microscopic fluorescence analysis of EDA and EDB expression (green) on various human cancer sections using F8(scDb)-IL7 (aEDA) and L19(scDb)-IL7 (aEDB) fusion proteins. An anti-CD31 antibody was used to stain blood vessels (red). 20×magnification, scale bar=100 µm. EDA, extra domain A. [113]Figure 2 [114]Open in a new tab Biochemical characterization of F8(scDb)-IL7 The fusion protein F8(scDb)-IL7 ([115]figure 3A) was cloned and produced by transient gene expression in mammalian CHO cells. The immunocytokine could be produced and purified to homogeneity, as shown by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) ([116]figure 3B) and the size exclusion chromatogram ([117]figure 3C). Figure 3. Biochemical and functional characterization of F8(scDb)-IL7. (A) Schematic representation of F8(scDb)-IL7. (B) SDS-PAGE analysis on 12% gel in reducing (R) and non-reducing (NR) conditions of F8(scDb)-IL7. (C) Size exclusion chromatogram of F8(scDb)-IL7. (D) SPR of F8(scDb)-IL7 on EDA-coated CM5 sensor chip. (E) Differential scanning fluorimetry of F8(scDb)-IL7. (F) Activity assay based on hPBMCs proliferation by exposure to F8(scDb)-IL7 and rhIL-7. (G) IFNg release by hPBMCs exposed to titration of F8(scDb)-IL7 in coated EDA (+) and non-coated EDA (-) wells. (H) IFNγ release on hPBMCs exposed to rhIL-7. Results are expressed as a percentage proliferation (proliferation%±SEM) and as IFNγ measured in pg/mL±SEM (n=3). (**p<0.01, ***p<0.001). EDA, extra domain A; hPBMCs, human peripheral blood mononuclear cells; SDS-PAGE, sodium dodecyl sulfate–polyacrylamide gel electrophoresis. [118]Figure 3 [119]Open in a new tab Binding kinetics of the F8 antibody to its cognate antigen (EDA) was confirmed by surface plasmon resonance analysis ([120]figure 3D). Differential scanning fluorimetry revealed a first transition at 49.9°C attributed to the unfolding of IL-7, a second and third transition at 54.5°C and 60.1°C may be attributed to the opening of the pairing diabodies and the unfolding of the variable (V[H]V[L]) domains, respectively ([121]figure 3E). Bioactivity assessment of F8(scDb)-IL7 The biological activity of F8(scDb)-IL7 was evaluated by a proliferation assay conducted on hPBMCs. The fusion protein had an IL-7 activity that closely matched that of the rhIL-7 control, with EC50 values of 116 pM and 99 pM, respectively ([122]figure 3F). Additionally, an IFNγ release assay was performed on hPBMCs ([123]figure 3G) in the presence or absence of the target antigen EDA, immobilized in a solid support to mimic the TME. IL-7 activity was persistent at low concentrations in the presence of EDA. The EC[50] of F8(scDb)-IL7 in this assay was 62 nM. IFNγ levels released in response to rhIL-7 are shown in [124]figure 3H. IFNγ levels released in response to KSF(scDb)-IL7 are reported in [125]online supplemental figure S5 and were not dependent on the presence of EDA in the wells. Flow cytometry on hPBMCs cocultured with immunocytokine fusion proteins We studied cell surface markers to assess the effect of F8(scDb)-IL7 on hPBMCs, following in vitro coincubation. In inactivated hPBMCs exposed to F8(scDb)-IL7, a concentration-dependent increase in CD69 expression, an early activation marker of lymphocytes, on CD8+T cells was observed. CD44, a late activation memory marker of lymphocytes, levels were also increased ([126]figure 4A). Similar findings were observed using L19IL2, a clinical-stage fusion protein consisting of the L19 antibody in diabody format fused to IL2[127]^28 ([128]figure 4B). Figure 4. Fluorimetry analysis on hPBMCs exposed to F8(scDb)-IL7 reveals increased expression of TCF1 compared with L19IL2. Expression levels of CD69 and CD44 on inactivated CD8+T cells exposed to F8(scDb)-IL7 (A) or L19IL2 (B). Expression levels of CD69 and CD44 on activated CD8+T cells exposed to F8(scDb)-IL7 (C) or L19IL2 (D). Percentage of CD3+ and CD4+ T cells exposed to F8(scDb)-IL7 (E) or L19IL2 (F). Contour plot displaying expression levels of Ki67 (x-axis) and TCF1 (y-axis) on CD8+CD44+ T cells exposed to F8(scDb)-IL7 at 62.5 nM (G) or L19IL2 at 50 nM (H). Expression of TCF1+Ki67+ on CD8+ CD44+ exposed to F8(scDb)-IL7 (blue) or L19IL2 (green) at three distinct concentrations (1000 nM, 250 nM and 62.5 nM for F8(scDb)-IL7 and 100 nM, 50 nM and 10 nM for L19IL2) (I). Results are expressed as marker positivity percentage (% ±SEM). (*p<0.05, **p<0.01, ***p<0.001, ****p<0.0001). hPBMCs, human peripheral blood mononuclear cells. [129]Figure 4 [130]Open in a new tab When activated hPBMCs were incubated with F8(scDb)-IL7 or L19IL2, we observed a reduction of CD69 expression and a concentration-dependent increase in CD44 expression on CD8+T cells ([131]figure 4C,D). Furthermore, we noticed a consistent increase in the CD4+T cell population within the CD3+T cell subset across all concentrations of F8(scDb)-IL7 and L19IL2 ([132]figure 4E,F). TCF1 levels in CD8+T cells did not change on exposure to F8(scDb)-IL7 at various concentrations ([133]figure 4G, [134]online supplemental figure S6) while L19IL2 did not upregulate this marker ([135]figure 4H, [136]online supplemental figure S6). A greater percentage of CD8+Ki67+ TCF1+ was observed following incubation with F8(scDb)-IL7 over L19IL2 ([137]figure 4I). Additionally, F8(scDb)-IL7 induced an increase of IL7R expression and an expansion of CD8+CD62L+CD45RO− and of CD8+CD62L−CD45RO+ while L19IL2 induced an expansion of CD8+CD62L+CD45RO+ ([138]online supplemental figures S7 and S8). Therapy experiments on subcutaneous tumor models We studied the therapeutic activity of F8(scDb)-IL7, KSF(scDb)-IL7, and L19IL2 in immunocompetent BALB/c mouse-bearing WEHI-164 sarcomas ([139]figure 5A). F8(scDb)-IL7 displayed superior activity compared with the untargeted KSF(scDb)-IL7 counterpart. Moreover, the therapeutic efficacy of F8(scDb)-IL7 was similar to the one of L19IL2 in this model. L19IL2 was included as a positive control in our study since IL2 shares part of its receptor with IL7 (common g chain) and L19IL2 is a clinical-grade molecule, currently being investigated in phase III clinical trials ([140]NCT02938299 and [141]NCT03567889). Dose-limiting toxicity was not reached, as evidenced by the absence of body weight loss in all treatment groups. Figure 5. Therapeutic performance of F8(scDb)-IL7 in subcutaneous tumor models. (A) WEHI-164 tumor-bearing BALB/c mice received either Saline, KSF(scDb)-IL7 130 µg, F8(scDb)-IL7 130 µg and L19IL2 100 µg three times every 48 hours (black arrow; n=4 per group). (B) WEHI-164 tumor-bearing BALB/c mice received either Saline, αPD-1 200 µg, F8(scDb)-IL7 130 µg three times every 48 hours (black arrow) or αPD-1 200 µg (red arrow) in combination with F8(scDb)-IL7 130 µg (black arrow) 24 hours later. (C) MC38 tumor-bearing C57BL/6 mice received either Saline, αPD-1 200 µg, F8(scDb)-IL7 130 µg three times every 48 hours (black arrow) or αPD-1 200 µg (red arrow) in combination with F8(scDb)-IL7 130 µg (black arrow) 24 hours later.(D) F9 tumor-bearing 129/SvEv mice received either Saline, αPD-1 200 µg, F8(scDb)-IL7 130 µg three times every 48 hours (black arrow) or αPD-1 200 µg (red arrow) in combination with F8(scDb)-IL7 130 µg (black arrow) 24 hours later. Curves were halted on day 13 because animals were sacrificed earlier in the study due to small ulcer formation, in compliance with our animal license. Data represent mean tumor volume and body weight change % (±SEM). (*p<0.05, ****p<0.0001). CR, complete response. [142]Figure 5 [143]Open in a new tab In a second experiment, we explored the combination with PD-1 blockade in the WEHI-164 model. Notably, the coadministration of αPD-1 and F8(scDb)-IL7 resulted in complete tumor remission in 86% of the mice, whereas both monotherapy groups induced one complete response each (14%) ([144]figure 5B). Dose-limiting toxicity was not reached. Cured mice were rechallenged with WEHI-164 cells 50 days after primary tumor implantation and were found to have acquired protective immunity ([145]online supplemental figure S9). In a third experiment, coadministration of αPD-1 and F8(scDb)-IL7 was evaluated in C57BL/6 immunocompetent mice bearing MC38 carcinoma tumors. All treatment groups exhibited substantial tumor growth inhibition compared with the saline control, with statistically significant differences evident by day 16. Notably, the combination group demonstrated a positive trend, marked by a more pronounced tumor growth inhibition ([146]figure 5C), although complete responses were not observed. A fourth therapy experiment was performed in F9 teratocarcinoma implanted in 129/Sv immunocompetent mice. In this setting, only the combination of the F8(scDb)-IL7 combined with PD-1 blockade gave a detectable tumor growth inhibition with one observed complete response ([147]figure 5D). Biomarkers for immunotoxicity Spleen and blood samples were collected from mice in each group 24 hours after the last treatment to explore potential mechanisms underlying complete remission in the sarcoma model. Remarkably, mice treated with F8(scDb)-IL7, either alone or in combination with αPD-1, displayed visibly enlarged spleens compared with the saline or αPD-1 monotherapy groups ([148]figure 6A). To quantify this effect, spleens were weighted, revealing a statistically significant difference in mice treated with F8(scDb)-IL7 ([149]figure 6B). Moderate IFNγ levels were observed in the blood of mice treated with either the combination or F8(scDb)-IL7 alone ([150]figure 6C). Figure 6. Mechanism of action of F8(scDb)-IL7 and aPD-1 combination in BALB/c immunocompetent mice. (A) Photograph of spleens harvested from BALB/c tumor-bearing mice treated either with either saline, αPD-1 200 µg, F8(scDb)-IL7 130 µg three times every 48 hours or with αPD-1 200 µg in combination with F8(scDb)-IL7 130 µg. Scale bar=1 cm. (B) Weight comparison of spleens from BALB/c treated mice. (C) IFNγ levels in plasma of BALB/c treated mice. (D) Ex vivo immunofluorescence analysis on WEHI-164 sarcoma 24 hours after the third injection of saline, αPD-1, and F8(scDb)-IL7 alone or in combination with 200 µg of αPD-1. Markers specific for Tregs (Foxp3), NK cells (NCR1), CD4+T cells (CD4), and CD8+T cells (CD8) were used (green). Blood vessels were stained with an anti-CD31 antibody (red). Magnification: ×20; scale bars=100 µm. (E) Principal component analysis (PCA) of spleens proteomics data. (F) Pathway enrichment analysis was used to identify significantly impacted biological pathways after αPD-1 + F8(scDb)-IL7 combination treatment. (*p<0.05, **p<0.01, ****p<0.0001). [151]Figure 6 [152]Open in a new tab Immune cell infiltrate study on WEHI-164 sarcoma A microscopic analysis of tumor sections obtained from tumors harvested 24 hours after the last injection was conducted for each group ([153]figure 6D). Typically, Tregs, NK cells and CD4+T cells were not detectable in the various treatment groups. However, infiltration of CD8+T cells was evident in tumors from all treatment groups, except for mice which had received the saline solution. An enhanced density of CD8+T cells was observed in the combination group (see also [154]online supplemental figure S10). Proteomic analysis from treated WEHI164-bearing BALB/c mice A spleen proteomic analysis was carried out to better characterize the differences between the experimental groups. Principal component analysis allowed clustering of the different treatments based on principal component 1 (PC1, explaining 43% of variance) and PC 2 (PC2, explaining 7% of variance), suggesting that the proteomic landscapes were distinct ([155]figure 6E). Biologically relevant pathways involved in the combination treatment group were identified, these include enhanced proliferation, immune system activation, and inflammatory response ([156]figure 6F). The proteomic analysis allowed us to identify more than 2800 proteins in the spleen samples. Proteins statistically upregulated and downregulated are represented in [157]online supplemental figure S11A, B. Immunologically relevant upregulated proteins are represented in [158]online supplemental figure S11C. A detailed list of proteins can be found in [159]online supplemental tables S3 and S4. A tumor proteomic analysis revealed increased levels of Granzyme B, proteins associated with immune activation and infiltration in the groups where F8(scDb)-IL7 was enriched ([160]online supplemental figure S12). Therapy experiment on orthotopic glioblastoma model The combination of F8(scDb)-IL7 and PD-1 blockade was also explored in an orthotopic GL-261 glioma model according to the schedule indicated in [161]figure 7A. The survival curve indicated a beneficial effect of the combination group with a statistical significance compared with the saline and the αPD-1 monotherapy groups ([162]figure 7B,C). This finding was supported by a lower tumor volume observed in the combination group compared with the other groups, as indicated by tumor fluorescence measurements on day 18 ([163]figure 7D,E). Figure 7. Therapeutic performance of F8(scDb)-IL7 in GL-261 orthotopic glioma model. (A) Treatment schedule and schematic overview of the therapeutic agents investigated in C57BL/6 orthotopic GL-261 glioma-bearing mice. (B) Survival studies on GL-261 orthotopically implanted C57BL/6 female mice. Treatments started on day 5 post-tumor implantation as shown in the scheme. (C) Body weight change (%) of treated mice. (D) FMT of mice treated either with saline, αPD-1 200 μg, F8(scDb)-IL7 130 µg, or combination. (E) Tumor fluorescence signal at day 18 post-tumor implantation (n=3). Survival data are presented as Kaplan-Meier plots. P values were calculated with the log-rank test (treatment vs other treatments: **p<0.01). [164]Figure 7 [165]Open in a new tab Discussion In this work, we described the generation and characterization of novel IL-7-based fusion proteins. We assessed five different immunocytokines in in vivo quantitative biodistribution analysis from which the fusion proteins featuring the scDb format emerged as the best-performing ones, based on the tumor:organ ratio profile. The product based on the F8 antibody showed potent antitumor activity as monotherapy and potentiated PD-1 blockade in four different immunocompetent cancer mouse models. Several lines of research indicate that the expression of IL-7R may represent a prognostic biomarker in melanoma and other malignancies.[166]^16 45 Micevic et al reported that IL-7R^hi tumor-specific CD8+T cells had enhanced cytotoxic properties and reduced expression of exhaustion markers. As this immune population appears to be critical for the generation of antitumor memory, strategies aimed at increasing the IL-7 concentration at the tumor site should have therapeutic potential. Interestingly, IL-7R^hi CD8+T cells were characterized by high expression of TCF1[167]^46 in analogy to our findings in this article. Coexpression of stem-like or memory markers, such as TCF1 and IL-7R, on TILs was shown to be associated with improved clinical outcomes with ICB therapies. Siddiqui et al[168]^47 reported that TCF1+PD-1+ CD8+ intratumoral T-cells actively contribute to tumor growth inhibition during ICB treatment. To support their findings, the group generated transgenic mice lacking TCF1. These animals bearing B16-gp33 tumors did not respond to PD-1 blockade while the corresponding wild-type mice benefited from the treatment, suggesting a critical role of TCF1 in antitumor immune responses.[169]^9 A clinical study conducted in metastatic melanoma patients treated with adoptive cell immunotherapy revealed that complete durable responses were dependent on the phenotype of the ex vivo expanded tumor resident CD8+T cells. Krishna et al compared the phenotype of TILs expanded from complete responders (CRs) to those from non-responders and found that CR’s TILs lacked CD39 and CD69 expression but also co-expressed inhibitory receptors such as PD-1 or stem-cell-like markers such as TCF1.[170]^45 The discovery of this CD8 T cell population featuring high levels of IL-7R and TCF1 has sparked considerable interest in exploring the potential of IL-7-based pharmaceuticals. BiCKI-IL7, a fusion protein based on an αPD-1 antibody connected to an IL-7 variant, was designed to selectively target the so-called progenitor-exhausted CD8 T-cells which simultaneously express PD-1 and IL-7R.[171]^48 In preclinical studies, BiCKi-IL7 has been shown to significantly increase survival in an orthotopic HCC model compared with αPD-1 alone. NT-I7, a rhIL-7-Fc-hybrid fusion protein with a prolonged half-life,[172]^49 had shown promising results in combination with CAR-T cell therapy in mouse models of cancer.[173]^50 This product is currently being studied in clinical trials for the treatment of different types of malignancies, both in monotherapy or in combination with checkpoint inhibitors, CAR-T, chemotherapy, and radiotherapy. NT-I7 has also been used in a compassionate study program in recurrent glioblastoma patients, showing a favorable safety profile at high dosages (ie, 1440 µg/kg).[174]^25 As most patients ith cancer are still not cured by ICB, there is a medical need for novel effective combination immunotherapies. F8(scDb)-IL7 emerges as a promising candidate to boost tumor-specific memory T-cells in combination with anti-PD1 antibodies. As EDA FN is expressed in the majority of human solid malignancies, the product may find broad applicability. Based on our results, the combination of ICB with EDA-FN-targeted IL-7 represents a promising strategy that deserves to be explored in the clinic. supplementary material online supplemental file 1 [175]jitc-12-8-s001.pdf^ (6.5MB, pdf) DOI: 10.1136/jitc-2023-008504 online supplemental file 2 [176]jitc-12-8-s002.png^ (587.3KB, png) DOI: 10.1136/jitc-2023-008504 Acknowledgements