Fc-MODIFIED BIOLOGICALS FOR LOCAL DELIVERY TO COMPARTMENT, IN PARTICULAR TO THE CNS

Abstract

A polypeptide comprising a crystallizable fragment (Fc) region of IgG for use in prevention or treatment of a disease, particularly a disease affecting the central nervous system. The polypeptide is administered locally to the affected compartment, in particular to the central nervous system. The Fc region bears a modification resulting in reduced affinity to the neonatal Fc receptor (FcRn), resulting in an increased brain to serum concentration of the polypeptide.

Claims

1-25. (canceled)

26. A polypeptide comprising: a Fc region of IgG, wherein said Fc region bears a modification resulting in reduced affinity to the neonatal Fc receptor (FcRn), and said Fc region is or comprises a sequence characterized by SEQ ID NO. 002 (IAQ), SEQ ID NO. 003 (AHQ), SEQ ID NO. 004 (NHQ), SEQ ID NO. 005 (AAQ), SEQ ID NO. 006 (NAQ), SEQ ID NO. 007 (AHH), SEQ ID NO. 008 (NHH), SEQ ID NO. 009 (AAH), SEQ ID NO. 010 (NAH), SEQ ID NO. 011 (NAA), SEQ ID NO. 012 (NAE), SEQ ID NO. 013 (AAA) or SEQ ID NO. 014 (AAE).

27. The polypeptide according to claim 26, wherein said polypeptide is selected from: a. a fusion protein comprising i. an effector polypeptide, and ii. said Fc region; or b. an antibody or antibody-like molecule comprising said Fc region.

28. The polypeptide according to claim 27, wherein said effector polypeptide is selected from hIL-12, IL-10, IL-2, IL-7, IFNα, IFNβ, IFNγ, IL-15, TNFα, CTLA-4, TGFβ, TGFβRII, GDNF, hIL-35, CD95, IL-1RA, IL-4, IL-13, SIRPα, G-CSF, GM-CFS, OX40L, CD80, CD86, GITRL, 4-1BBL, EphrinA1, EphrinB2, and EphrinB5, BDNF, C9orf72, NRTN, ARTN, PSPN, CNTF, TRAIL, IFNα, IFNβ, IL-4, IL-3, IL-1 IL-5, IL-8, IL-18, IL-21, CCL5, CCL21, CCL10, CCL16, CX3CL1, and CXCL16.

29. The polypeptide according to claim 27, wherein said antibody or antibody-like molecule is selected from an antibody or antibody-like molecule specifically binding to a molecule selected from PD-L1, TNFα, Histone, IFNγ, CXCL10, CTLA4, PD-1, OX40, CD3, CD20, CD22, CD25, CD28, TREM2, IL-6, CX3CR1, Nogo-A, CD27, IL-12, IL-12Rb1, IL-23, CD47, TGFβ, EGFR, EGFRvIII, Her2, PDGFR, TGFR, FGFR, IL-4, IL-4RA, TfR, LfR, IR, LDL-R, LRP-1, CD133, CD111, VEGFR, VEGF-A, Ang-2, IL-10, IL-10R, IL-13Rα2, α-synuclein, CSF1R, GITR, TIM-3, LAG-3, TIGIT, BTLA, VISTA, CD96, 4-1BB, CCL2, IL-1 or IL-1R, EphA2, EphA3, EphB2, EphB3, and EphB4, LINGO-1, L1 CAM, NCAM, C0147, SOD-1, SIGMAR-1, SIGMAR-2, TDP-43, Aβ, Tau, IFNα, IFNβ, TRPM4, ASIC1, VGCCs, CB.sub.1, TTR, HTT, JCV, and C9orf72.

30. The polypeptide according to claim 26, wherein said polypeptide is an antibody or antibody-like molecule comprising or linked to said Fc region.

31. The polypeptide according to claim 30, wherein said antibody or antibody-like molecule is a bispecific construct able to bind two antigens at the same time,

32. The polypeptide according to claim 31, wherein said bispecific antibody or antibody-like molecule binds to PD-L1 and IL-12 receptors in an agonistic manner.

33. A nucleic acid encoding the polypeptide according to claim 26.

34. A viral vector comprising the nucleic acid according to claim 33.

35. A method of treating a disease selected from brain cancer, stroke, dementia, Parkinson's disease, Alzheimer's disease, multiple sclerosis, epilepsy, and traumatic central nervous system injury, which comprises administering the polypeptide according to claim 26.

36. A method of treating a disease selected from uveal melanoma, uveitis, and wet macular degeneration, which comprises administering the polypeptide according to claim 26.

37. A method of treating a disease selected from rheumatoid arthritis, juvenile rheumatoid arthritis, gout, pseudogout, osteoarthritis, chronic hemophilic synovitis, psoriatic arthritis, and ankylosing spondylitis, which comprises administering the polypeptide according to claim 26.

38. A method of treating a disease selected from coronavirus disease 2019, diseased caused by severe acute respiratory syndrome coronavirus (SARS-CoV), severe acute respiratory syndrome, asthma, allergic asthma, severe uncontrolled asthma, fibrosis, cystic fibrosis, chronic obstructive pulmonary disease, influenza, lung oedema, sarcoidosis, lung cancer, tuberculosis, human orthopneumovirus, bubonic plague, pneumonic plague, anthrax, invasive fungal disease in lung, pulmonary fibrosis, respiratory syncytial virus, chronic rhinosinusitis with nasal polyps, interstitial lung disease, idiopathic pulmonary fibrosis, and pulmonary paracoccidioidomycosis, which comprises administering the polypeptide according to claim 26.

39. A method of preventing or treating a disease affecting a central nervous system (CNS), which comprises: administering to a brain a fusion polypeptide comprising IL-12 and a crystallizable fragment (Fc) region of IgG, wherein said Fc region bears a modification resulting in reduced affinity to the neonatal Fc receptor (FcRn).

40. The method according to claim 39, wherein a serum or plasma to a brain concentration ratio of said polypeptide is below a predetermined threshold selected from: a. at most ⅔ of the serum or plasma to the brain concentration ratio of the same polypeptide comprising a non-modified Fc region, or b. at most ⅛ of the serum or plasma to the brain concentration ratio of the same polypeptide neither comprising an Fc region nor peptide linkers, measurable 24 h after intracranial injection into the striatum of FcRn.sup.tg mice.

41. The method according to claim 39, wherein said reduced affinity of said polypeptide to FcRn is characterized by a dissociation constant (K.sub.D) selected from: a. a K.sub.D that is at least 2× increased compared to a K.sub.D characterizing binding of FcRn to the same polypeptide comprising a non-modified Fc region, and b. a K.sub.D that is at least 1.5× increased compared to a K.sub.D characterizing binding of FcRn to the same polypeptide comprising a differently modified Fc region, namely one mutant selected from IAQ and AAA.

42. The method according to claim 39, wherein said administration is effected by a method selected from: a. single, intermittent or continuous local infusion, including convection enhanced delivery (CED), b. intrathekal or intracerebroventricular administration, c. in situ production of said polypeptide, d. release from implanted slow release formulations, e. molecular transport into the central nervous system, f. cellular transport into the central nervous system, or g. transport to the central nervous system after intranasal application.

43. The method according to claim 39, wherein said disease affecting the central nervous system is a malignant disease.

44. The method according to claim 43, wherein said malignant disease is a glioma or a high grade glioma (HGG).

45. The method according to claim 39, wherein said Fc region is a human Fc region or a chimeric Fc region comprising a human amino acid sequence and bears a mutation at position 253.

46. The method according to claim 45, wherein said mutation at position 253 is I253A or I253N.

47. The method according to claim 39, wherein said Fc region is or comprises a sequence characterized by SEQ ID NO. 002 (IAQ), SEQ ID NO. 003 (AHQ), SEQ ID NO. 004 (NHQ), SEQ ID NO. 005 (AAQ), SEQ ID NO. 006 (NAQ), SEQ ID NO. 007 (AHH), SEQ ID NO. 008 (NHH), SEQ ID NO. 009 (AAH), SEQ ID NO. 010 (NAH), SEQ ID NO. 011 (NAA), SEQ ID NO. 012 (NAE), SEQ ID NO. 013 (AAA) or SEQ ID NO. 014 (AAE).

48. A pharmaceutical composition suitable for use as a medicament, comprising: a polypeptide comprising a crystallizable fragment (Fc) region of IgG, wherein said Fc region bears a modification resulting in reduced affinity to the neonatal Fc receptor (FcRn), and said Fc comprises mutations I253N and H435Q and an H at position 310.

49. The pharmaceutical composition according to claim 49, wherein said polypeptide further comprises IL-12.

50. The pharmaceutical composition according to claim 49, wherein said Fc region is or comprises a sequence SEQ ID NO. 004 (NHQ).

51. A method of preventing or treating a disease affecting the eye, which comprises: administering to the eye by intraocular administration a polypeptide comprising a crystallisable fragment (Fc) region of IgG, wherein said Fc region bears a modification resulting in reduced affinity to the neonatal Fc receptor (FcRn), and said Fc comprises mutations I253N and H435Q and an H at position 310.

52. The method according to claim 51, wherein said polypeptide further comprises IL-12 or a polypeptide binding to any one of VEGFR, Ang2, TNFα, IL-17, PD-1 or PD-L1.

53. A method of preventing or treating a disease affecting a joint, which comprises: administering to said joint by intraarticular administration a polypeptide comprising a crystallisable fragment (Fc) region of IgG, wherein said Fc region bears a modification resulting in reduced affinity to the neonatal Fc receptor (FcRn), and said Fc comprises mutations I253N and H435Q and an H at position 310.

54. The method according to claim 53, wherein said polypeptide further comprises IL-12 or a polypeptide binding to any one of TNFα, IL-1RA, IL-6R, IL-6, CD27, IL-22, IL-17 or CD27,

55. A method of preventing or treating a disease affecting the lungs, which comprises: delivering to the lungs, via inhalation, a polypeptide comprising a crystallisable fragment (Fc) region of IgG, wherein said Fc region bears a modification resulting in reduced affinity to the neonatal Fc receptor (FcRn), and said Fc comprises mutations I253N and H435Q and an H at position 310.

56. The method according to claim 55, wherein said polypeptide further comprises IL-12 or IL-10 or a polypeptide binding to any one of IL-4RA, TNFα, IL-5, IL-6R, PD-1, PD-L1, CTLA-4, IL-8, IL-21R, CD25, CD20 or NF-kB.

Description

[0424] The invention is further illustrated by the following examples and figures, from which further embodiments and advantages can be drawn. These examples are meant to illustrate the invention but not to limit its scope.

[0425] FIG. 1: Human IL-12Fc has better tissue retention than IL-12. A. Schematic structure of murine IL-12Fc. rhIL-12—recombinant human IL-12, hIL-12Fc—human IL-12Fc. IgG4 Fc—fragment crystallizable region of human IgG4. B. Schematic of the experiment. IL-12 of IL-12Fc was injected into the striatum of FcRn.sup.tg mice. After 24 hours the remaining amount of injected protein was assessed in the brain and compared to the amount present in serum. C. Ratio of serum to brain IL-12 amount as assessed by ELISA. ELISA measured hIL-12 as a generic measure of IL-12Fc. Unpaired Student's t-test. **p<0.005. Mean±SD.

[0426] FIG. 2: IL-12Fc is being exported from the brain in an FcRn-mediated fashion. A. Brain tumor-bearing wt and FcRn.sup.tg mice were implanted with osmotic pumps delivering 12.5 pg/kg/day of murine IL-12Fc directly into the tumor lesion. Murine IL-12 levels measured in serum using a bead-based array. Unpaired Student's t-test of groups wt mIL-12Fc vs FcRn.sup.tg mIL-12Fc. Mean±SD. One way ANOVA with Tukey's multiple comparison test. B. Mice treated like in FIG. 2 A. Amount of IL-12 present in the circulation 24 hours after the start of the treatment as measured in serum using a bead-based array. Mean±SD. C. Mice treated like in FIG. 2 A. Levels of IFNγ in the circulation 24 hours after the start of the treatment as measured in serum using a bead-based array. Mean±SD. D. IFN-γ levels at day 7, experiment in A, Mean±SD.

[0427] FIG. 3: Protein stability measured using thermal shift assay. Protein was incubated in PBS (A) or artificial cerebrospinal fluid (aCSF, B). Five measurements per IL-12Fc variant. Whiskers represent the minimum and maximum spread.

[0428] FIG. 4: Mutations in the Fc fragment of IL-12Fc do not affect the biological activity of IL-12. A. Bioactivity of IL-12 measured using HEK-BIue™ IL-12 assay. EC50—effective concentration leading to 50% of maximal signal from HEK-Blue™ IL-12 reporter cells stimulated with IL-12Fc in the range 0 to 50 ng/ml, two replicates per concentration. Measured by using the activity of the secreted alkaline phosphatase using a colorimetric method. Each point shows result from an independent experiment. Mean±SD. B. STAT-4 phosphorylation in peripheral blood mononuclear cells (PBMCs) stimulated for 1 h with 100 ng/ml anti-CD3 and 10 ng/ml of recombinant IL-12, IL-12Fc WT or three of the variants designed for reduced FcRn affinity. Mean±SD. C. IFNγ production by PBMCs stimulated for 24 h with 100 ng/ml anti-CD3 and indicated concentrations of recombinant IL-12, IL-12Fc WT or three of the variants designed for reduced FcRn Affinity.

[0429] FIG. 5: Human IL-12Fc variants have reduced FcRn affinity. A. Surface plasmon resonance (SPR) measurement of FcRn affinity with human recombinant FcRn immobilized on the surface and IL-12Fc variants in the liquid phase. Affinity measured at pH=6.0. Data normalized to IL-12Fc WT. B. IL-12Fc variants binding to human FcRn. Measured by ELISA at pH=6.0. Mean±SD.

[0430] FIG. 6. Ratios of the concentrations of IL-12Fc in the blood and in the injected hemisphere. A. 1 pg of IL-12Fc WT or NHQ variant were injected into the striatum of FcRn.sup.tg mice. After 24 hours the amounts of IL-12 were assessed in the injected brain hemisphere and in serum by ELISA, their ratios were calculated and normalized to those for IL-12Fc WT group. 4 mice per group. Unpaired Student's t-test. *p<0.05. Mean±SD. B. 1 pg of IL-12Fc WT, IAQ, AAA or NHQ were injected into the striatum of FcRn.sup.tg mice using convection enhanced delivery (CED). After 24 hours the amounts of IL-12 were assessed in the injected brain hemisphere and in plasma by ELISA, their ratios were calculated and normalized to those for IL-12Fc WT group. 7-8 mice per group. One-way ANOVA with Tukey's multiple comparison test. Mean±SD.

[0431] FIG. 7: Brain retention after local treatment with IL-12Fc variants. FcRn.sup.tg mice were injected with 1 μg of IL-12Fc WT, IAQ, AAA or NHQ into the striatum using convection enhanced delivery (CED). Amount of IL-12Fc remaining in the brain tissue was measured 6 hours after injection by ELISA and normalized to IL-12Fc WT. One-way ANOVA with Tukey's multiple comparison test. Outlier removal. Mean±SD.

[0432] FIG. 8: A. Schematic structure of native and rmIL-12, mIL-12hIgG4 wt, mIL-12hIgG4 NHQ and mIL-12hIgG1:anti-hPD-L1 NHQ. B. Bioactivity of murine IL-12 constructs measured using HEK-Blue™ IL-12 assay. HEK-BIue™ reporter cells stimulated with IL-12 or IL-12Fc variants in the range of 0 to 50 ng/mL, using 5 to 8 dilution steps, two replicates per concentration. Measured by using the activity of the secreted alkaline phosphatase using a colorimetric method. X-axis values: concentration plotted as the corresponding amount of IL-12 molecules in pmol/ml. Representative of two individual experiments. C. Binding to PD-L1 on cells compared to full anti-PD-L1 (Atezolizumab) antibody. GL-261:luc or PD-L1 deficient GL-261:luc (PD-L1 KO) cells, stimulated with murine interferon-gamma (IFNγ) to stimulate PD-L1 expression, stained with anti-PD-L1 antibody or h/mIL-12hFc:aPD-L1 NHQ variants. Detection of cell-bound antibodies using anti-human-IgG-PE secondary antibody. D. Affinity to FcRn of NHQ mutated variants compared to WT as measured by surface plasmon resonance (SPR). Human recombinant FcRn immobilized on the surface and PD-L1 binders in liquid phase. Affinity measured at pH=6. Affinity constant K.sub.D in nM.

[0433] FIG. 9: Optimized IL-12 Fc fusions for local therapy of brain cancer lead to reduced systemic exposure without affecting the therapeutic effect.

[0434] A. Experimental timeline in days post tumor injection. GL-261:luc Brain tumor bearing animals were systematically allocated to treatment groups of comparable tumor load via bioluminescent imaging (BLI) on day 20 and treated via convection enhanced delivery (CED) with buffer only (control) or 1 pg of rmIL-12, mIL-12hFc:anti-PD-L1 bifunctional molecule, mIL-12hFc WT or mIL-12hFc NHQ on days 21 and d28 post tumor implantation. Blood sampling for plasma on time points: 0, 6 h, 24 h, 72 h, 7 days post CED injections as well as 14 days after the second CED injection.

[0435] B. Tumor progression upon treatment monitored by bioluminescence imaging. Plotted average radiance (p/s/cm2/sr) from region of interest (ROI) of individual animals, grouped by treatment cohort. Treatment via CED indicated by dotted vertical lines.

[0436] C. Plasma levels of IL-12 (black lines, left Y axis) and IFNγ (gray lines, right Y axis) in response to treatment. Measured on given time points by bead-based cytokine array. Treatment via CED indicated by dotted vertical lines.

[0437] D. FcRn affinity dependent difference of plasma IL-12 levels 6 h after CED on day 21. Mice injected with mIL-12hFc WT and mIL-12hFc NHQ. Data from experiment shown in A-C.

[0438] E. Kaplan-Meyer analysis of survival of treated mice from A-D. 6-7 mice per group.

[0439] FIG. 10: Antibodies

[0440] A. Surface plasmon resonance (SPR) measurement of FcRn affinity with human recombinant FcRn immobilized on the surface and IgG1 variants in the liquid phase. Affinity measured at pH=6.0. Data normalized to an IgG1 antibody with a non-modified Fc (WT). Three additional IgG1 clinical grade antibodies with a non-modified Fc part were used as an additional reference (IgG1_01 ipilimumab, IgG1_02 atezolizumab, IgG1_03 rituximab). Mean±SD.

[0441] B. 1 μg of IgG1 WT, IAQ, AAA or NHQ variant were injected into the striatum of FcRn.sup.tg mice using Convection Enhanced Delivery (CED). After 24 hours the amounts of human IgG were assessed in the injected brain hemisphere and in plasma by ELISA, their ratios were calculated and normalized to those for IL-12Fc WT group. 5 mice per group. Mean±SD.

[0442] C. Surface plasmon resonance (SPR) measurement of FcRn affinity with human recombinant FcRn immobilized on the surface and IgG4 variants in the liquid phase. Affinity measured at pH=6.0. Data normalized to an IgG4 antibody with a non-modified Fc (WT). A second IgG4 antibody (nivolumab) with a non-modified Fc part was used as an additional reference (IgG4). Mean±SD.

[0443] D. 1 pg of IgG4 WT, IAQ, AAA or NHQ variant were injected into the striatum of FcRn.sup.tg mice using Convection Enhanced Delivery (CED). After 24 hours the amounts of human IgG were assessed in the injected brain hemisphere and in plasma by ELISA, their ratios were calculated and normalized to those for IL-12Fc WT group. 5 mice per group. Mean±SD.

EXAMPLE 1: MATERIAL AND METHODS

[0444] Animals

[0445] C57BL/6J mice were obtained from Charles River. mFcRn.sup.−/−hFcRn.sup.tg(32) (FcRn.sup.tg) mice were obtained from The Jackson Laboratory (stock number 014565). All animals were kept in house according to institutional guidelines under specific pathogen-free (SPH) conditions at a 12 h light/dark cycle with food and water provided ad libitum. All animal experiments were performed according to institutional guidelines and approved by the Swiss Cantonal Veterinary Office (license number 246/2015).

[0446] Tumor Cell Lines

[0447] GL-261 cells were provided by A. Fontana, Experimental Immunology, University of Zurich, Zurich, Switzerland and cultured in DMEM supplemented with 10% heat inactivated fetal calf serum and L-glutamine (all from Thermo Fisher Scientific). The murine GL-261 brain tumor cell line (syngenic to C57BL/6), was stably transfected with pGI3-ctrl and pGK-Puro (Promega) and selected with puromycin (Sigma-Aldrich) to generate luciferase-stable GL-261 cells. To generate GL-261:luc PD-L1 KO tumor cells, cells were transiently transfected with a Streptococcus pyogenes Cas9 P2A GFP—single guide RNA (sgRNA) expression vector (pX458; Addgene) modified to express the following sgRNA, 5′-GTATGGCAGCAACGTCACGA-3′. 3 days after transfection, GFP positive, PD-L1 KO cells were purified by flow cytometry by gating on PD-L1 negative cells after 48 h of IFN-γ stimulation (10 ng/ml). A single clone was further amplified and loss of PD-L1 expression was re-confirmed via flow cytometry before use in experiments.

[0448] Surgical Procedures

[0449] For glioma inoculation, 6-10 week old mice were anesthetized using a mixture of fentanyl (Helvepharm AG), midazolam (Roche Pharma AG) and medetomidin (Orion Pharma AG). GL261 cells were injected intracranially (i. c.) in the right hemisphere using a stereotactic robot (Neurostar). Briefly, a blunt-ended syringe (Hamilton; 75 N, 26 s/2″/2, 5 μl) was placed 1.5 mm lateral and 1 mm frontal of bregma. The needle was lowered into the burr hole to a depth of 4 mm below the dura surface and retracted 1 mm to form a small reservoir. Injection was performed in a volume of 2 μl at 1 μl/min. After leaving the needle in place for 2 min, it was retracted at 1 mm/min. The burr hole was closed with bone wax (Aesculap, Braun) and the scalp wound was sealed with tissue glue (Indermil, Henkel). Anesthesia was interrupted using a mixture of flumazenil (Labatec Pharma AG) and Buprenorphine (Indivior Schweiz AG), followed by injection of atipamezol 20 minutes later (Janssen). Carprofen (Pfizer AG) was used for perioperative analgesia.

[0450] After 7 to 14 days osmotic pumps (model 2004, 0.25 μl/h; Alzet) were filled with murine IL-12Fc (12.5 pg/kg/24 h) or PBS alone and primed at 37° C. in PBS. Mice were anaesthetized as above and the previous burr hole of the glioma injection was located, the bone wax and periosteal bone was removed, and the infusion cannula was lowered through the burr hole 3 mm into the putative center of the tumor. Serum samples were collected every two days by blood sampling from the tail vein, starting from day −1 of the pump implantation using Vacutainer tubes and following manufacturer's instructions (Becton, Dickinson and Company).

[0451] For the comparison of IL-12 and IL-12Fc WT serum to brain concentration ratio after bolus injection, mice were anesthetized and intracranially injected in the right hemisphere using a stereotactic robot (Neurostar) as described above for tumor cell injection. Mice received 100 ng of recombinant human IL-12 (Prospec) or equivalent amount of IL-12Fc (69 ng/mouse). Dosage was calculated based on the HEK-Blue IL-12 bioactivity assay. Animals were sacrificed 24 hours later by controlled CO.sub.2 asphyxiation. Blood samples were collected by cardiac puncture and mice were perfused with 20 ml of ice cold PBS. Serum was isolated as described above and brain tissue was snap-frozen in liquid nitrogen.

[0452] For the comparison of IL-12 WT and IL-12Fc NHQ serum to brain concentration ratio after bolus injection, mice were anesthetized and intracranially injected in the right hemisphere using a stereotactic robot (Neurostar) as described above for tumor cell injection. Mice received 1 μg of human IL-12Fc WT or IL-12Fc NHQ. Animals were sacrificed 24 hours later by controlled CO.sub.2 asphyxiation. Blood samples were collected by cardiac puncture and mice were perfused with 20 ml of ice cold PBS. Serum was isolated as described above and brain tissue was snap-frozen in liquid nitrogen.

[0453] For the convection enhanced delivery (CED) of protein into the brain, mice were anesthetized and intracranially injected in the right hemisphere using a stereotactic robot (Neurostar) and catheters made using a 27 G blunt-end needle with a 1 mm step at the tip made of fused silica with internal diameter of 0.1 mm and wall thickness of 0.0325 mm. Briefly, a burr hole was made at position 1 mm anteroposterior and 2 mm mediolateral of bregma. The catheter was lowered into the burr hole to a depth of 3.5 mm below the dura surface. Injection was performed in a volume of 5 μl at 0.2 μl/min, then 2 μl at 0.5 μl/min and 2 μl at 0.8 μl/min. After leaving the needle in place for 2 min, it was retracted at 1 mm/min. Mice received 1 μg of recombinant human IL-12Fc WT, IL-12Fc IAQ, IL-12Fc AAA, IL-12Fc NHQ or rmIL-12, mIL-12hFc WT, mIL-12hFc HNQ, mIL-12hFc:PD-L1 NHQ, Flu HA3.1 WT, Flu HA3.1 IAQ, Flu HA3.1 AAA, Flu HA3.1 NHQ or Atezolizumab WT, Atezolizumab IAQ, Atezolizumab AAA, or Atezolizumab NHQ. Animals were sacrificed 6 hours later by controlled CO.sub.2 asphyxiation. Ipsilateral brain hemispheres were snap-frozen in liquid nitrogen.

[0454] In Vivo Bioluminescent Imaging

[0455] Tumor-bearing mice were injected with d-Luciferin (150 mg/kg body weight; XenoLight d-luciferin potassium salt; BioVision 7903-1G; 15 mg/mL in PBS). Animals were transferred to the dark chamber of a Xenogen IVIS Lumina III (PerkinElmer) imaging system and luminescence was recorded for 1 to 2 minutes, medium binning (4). Data were subsequently analysed using Living Image 4.7.1 software (PerkinElmer). A circular region of interest (ROI; 1.5 cm diam) was defined around the tumor site and photon flux of this region was read out and plotted.

[0456] BLI and Systematic Group Allocation

[0457] At d 20 after implantation of GL-261luc glioma cells, tumor-bearing animals were distributed into experimental groups of equal average BLI.

[0458] Blood Sampling

[0459] Blood samples were taken 10 min before CED or 6 h, 24 h, 72 h, 7 days after CED injections. 20 to 50 uL of blood were taken from the tail vein into a microtainer containing dried K2-EDTA (Becton, Dickinson and Company). After centrifugation for 5 min at 10′000 g, plasma was transferred to a fresh tube and frozen.

[0460] FcRn ELISA

[0461] IL-12Fc variants or a recombinant human IgG4 anti-GFP antibody (clone 515, AbD Serotec), which served as a control, were coated on a micro-well plate (Greiner Bio-One). Histidine-coupled FcRn (R&D Systems) was incubated at increasing concentration in ELISA diluent (Mabtech) at pH=6.0. FcRn was detected by a biotinylated anti-His-antibody (clone 13/45/31-2, Dianova), streptavidin-coupled horseradish peroxidase (Mabtech) and a colorimetric substrate (Chromogen-TMB, Thermo Fisher Scientific). Optical density at a wavelength of 450 nm was measured using a spectrophotometer (Molecular Devices).

[0462] Bead-Based Cytokine Array

[0463] Serum levels of mIL-12 and mIFNγ were measured using Legendplex Mouse Inflammation Panel (Biolegend) following manufacturer's instructions. Samples were acquired using LSRII Fortessa (Becton, Dickinson and Company). Data analysis was performed using FlowJo Version 10.6 (Tree Star).

[0464] HEK-Blue IL-12 Bioactivity Assay

[0465] HEK-Blue IL-12 cells (InvivoGen) were plated on a flat bottom 96 well plates (Corning) at a density of 50 000 cells/well in medium containing normocin (InvivoGen). Cells were incubated with increasing amounts of IL-12, IL-12Fc WT or variants designed for reduced FcRn affinity for 17 hours. Culture medium was collected and incubated for 2 hours in presence of Quanti-Blue detection reagent (InvivoGen). Absorbance was measured at 640 nm using a table top spectrophotometer (Molecular Devices).

[0466] Human IL-12 Detection in the Brain, Plasma and Serum after Injection into the Brain and Calculation of the Serum or Plasma to Brain Concentration Ratio

[0467] Samples were diluted in PBS containing 0.05% Tween-20 and 0.1% BSA and IL-12 levels were assessed by ELISA for hIL-12p70 (Mabtech). To calculate serum or plasma to brain concentration ratio, the concentration of IL-12 in serum or plasma was described in pg/ml, whereas concentration in brain was calculated by dividing the total amount of IL-12 extracted from the brain corrected for the efficacy of protein extraction by the weight of the hemisphere (pg/mg of brain tissue).

[0468] Human IgG Detection in the Brain and Plasma after Injection into the Brain and Calculation of the Plasma to Brain Concentration Ratio

[0469] Samples were diluted in PBS containing 0.05% Tween-20 and 0.1% BSA and IgG levels were assessed by ELISA. Briefly, plates were coated with polyclonal donkey anti-human IgG (Jackson ImmunoResearch), blocked with PBS containing 0.05% Tween-20 and 0.1% BSA. Analyte was detected with a polyclonal goat anti-human IgG (Sigma-Aldrich) and amplified with a polyclonal donkey anti-goat HRP-conjugated antibody (Promega). For calculation of the plasma to brain ratio, the concentration of both human IgG in plasma and brain was described in pg/ml.

[0470] Production of Human IgG1 Variants, IgG4 Variants hIL-12hFc:aPD-L1 NHQ and mIL-12hFc:aPD-L1 NHQ.

[0471] IgG4 variants were expressed in transiently transfected human embryonic kidney (HEK) cell cultures. IgG1 variants, hIL-12hFc:aPD-L1 NHQ and mIL-12hFc:aPD-L1 NHQ were produced by transiently transfected chinese hamster ovary (CHO) or cell cultures. Briefly, culture supernatants were collected and protein was purified by affinity chromatography (Protein G). Protein was further purified by ion exchange (IEC) and size Exclusion Chromatography (SEC). Protein was concentrated using spin columns (Sartorius, 30 kDa cutoff). Protein was stored in 20 mM histidine, 150 mM NaCl, pH=6.0 buffer. Quality was assessed by gel electrophoresis (SDS-PAGE) followed by Coomasie staining according to standard protocols. Nivolumab, atezolizumab, ipilimumab and rituximab are commercially available.

[0472] IFN-γ Production by Lymphocytes Stimulated with IL-12Fc

[0473] Human peripheral blood mononuclear cells (PBMCs) were stimulated for 24 h with increasing concentrations of IL-12, IL-12Fc or IL-12Fc variants with reduced FcRn affinity in presence of 100 ng/ml anti-CD3 antibody. IFN-γ levels in supernatant were measured by ELISA following manufacturer's instructions (Mabtech).

[0474] Brain Protein Isolation

[0475] After euthanasia and careful removal of the skullcap the brain was isolated. Cerebellum and olfactory bulbs were removed, the hemispheres are separated along the midline and the injected (ipsilateral) hemisphere was snap frozen in liquid nitrogen. Brain lysates were prepared by homogenization in ice cold lysis buffer (Cell signaling) containing Halt protease inhibitor cocktail (Thermo Fisher Scientific). 0.1 ml of lysis buffer was added per 10 mg of brain tissue. Brain tissue was minced using scissors, then passed through a 20 G needle and finally sonicated for 20 seconds. Samples were spun down for 10 minutes at 15000 g at 4° C. and supernatants were transferred to fresh tubes. Protein concentration was measured using Pierce BCA assay kit (Thermo Fisher Scientific) and this data was used to correct for the protein extraction efficiency within each experiment. Protein expression and purification

[0476] All human and murine IL-12Fc variants were expressed in HEK239T. Variants that retained protein G affinity, were purified from the culture supernatant by affinity chromatography using protein G sepharose (Biovision) and overnight dialysis with PBS. Variants that lost the protein G affinity were purified by precipitation with ammonium sulfate (VI) at 50% saturation, followed by dissolving the precipitate with PBS and purifying over ceramic hydroxyapatite (CHT) column (type II, 40 μm Bio-Rad). After protein G or CHT chromatography, samples were further purified by ion-exchange chromatography using diethylaminoethanol-linked sepharose as anionite (HiTrap DEAE Sepharose FF columns, GE Healthcare) on an ÄKTA Pure chromatography system (GE Healthcare). Finally, all the IL-12Fc variants were purified via size exclusion chromatography using a prepacked Superose 6 Column (GE Healthcare) on an ÄKTA chromatography system (GE Healthcare). Dimeric fraction was concentrated using Vivaspin 2 ml spin columns with 30 kDa cutoff (GE Healthcare). Protein purity was validated by SDS-PAGE electrophoresis followed by staining with Coomassie Brilliant Blue (VWR Life Science). Protein concentration was measured using Pierce BCA assay kit (Thermo Fisher Scientific) and by ELISA for IL-12p70 (Becton, Dickinson and Company).

[0477] Phosphorylation of STAT-4 by Lymphocytes Stimulated with IL-12Fc

[0478] Human peripheral blood mononuclear cells (PBMCs) were stimulated for 1 h with 10 ng/ml of IL-12, IL-12Fc or IL-12Fc variants with reduced FcRn affinity in presence of 100 ng/ml anti-CD3. Cells were then lysed using Pierce RIPA buffer (Thermo Fisher Scientific). Samples were analyzed by SDS-Page electrophoresis followed by transfer using Trans-Blot Turbo Blotting system (Bio-Rad Laboratories, Inc.) and staining with anti-STAT4 pY693 (clone 38/p-Stat4, Becton, Dickinson and Company). Band visualization was performed using ECL clarity substrate (Bio-Rad Laboratories, Inc.) and detection on BioRad MPCD imager (Bio-Rad Laboratories, Inc.).

[0479] Surface Plasmon Resonance

[0480] SPR was performed using the ProteOn XPR36 System (Bio-Rad Laboratories, Inc.), coating human recombinant biotinylated FcRn (Immunitrack) on a ProteOn NLC sensor chip to approximately 80 response units (RU). IL-12Fc variants were ran in 10 mM sodium citrate buffer pH=6.0 with decreasing concentration from 729 nM to 9 nM in three fold steps. The dissociation time was 600 s. Analysis was performed using the ProteOn Manager software (Bio-Rad Laboratories, Inc.) using data normalization to the injection time, interspot background removal and build-in artefact removal function. Kd were calculated using the equilibrium analysis model.

[0481] Thermal Shift Assay

[0482] Briefly, 0.2 mg/ml of protein samples were mixed with Sypro Orange Protein stain diluted to 1:1000 (Sigma-Aldrich) and ran on a CFX384 thermocycler (Biorad) with 0.2° C. temperature increase every 30 s from 20° C. to 95° C. with fluorescence as readout. Temperature of denaturation was determined as a first derivative of fluorescence over temperature. Experiment was performed in PBS and artificial cerebrospinal fluid (aCSF; 125 mM NaCl, 26 mM NaHCO.sub.3, 1.25 mM NaH2PO.sub.3, and 2.5 mM KCl) as solvents.

[0483] Statistical Analysis

[0484] Statistical analysis was performed using Graphpad Prism 5 software. Outliers were removed from the final analysis according to the Grubb's test (49). Two groups were compared using unpaired Student's t-test. More than two groups were compared using One-way ANOVA with Tukey's multiple comparison test.

[0485] Flow Cytometry PD-L1 Binding Assay

[0486] GL261:lucE9 or GL261:lucE9:PD-L1K0 cells were cultured over-night with addition of murine interferon-gamma in a final concentration of 20 ng/mL. The next day, cells were washed with DPBS. Trypsin-EDTA (Invitrogen 25300-054) was added to the flask and removed immediately again. Cells were left to detach from the flask for 2-5 min. They were washed with culture medium and centrifuged at 350×g 4° C. 5 min. Subsequently, cells were plated into a round-bottom 96-well plate at 100′000 cells per well and washed with DPBS twice.

[0487] Staining was performed in PBS, 25 μL per well containing Zombie Aqua fixable viability kit (BioLegend) diluted 1:200 and either human anti-PD-L1 (Atezolizumab) or m/hIL-12hFc:aPD-L1 NHQ at a final concentration of 0.1 mg/mL. Cells were stained for 20 min at 4° C. in the dark. Following a washing step with PBS, cells were incubated with secondary antibody anti-human IgG-Fc-PE (Biolegend, cat. nr. 409304, lot B260868) at 0.2 mg/mL or anti-mouse-PD-L1-BV421 (Biolegend, cat. nr. 124315; lot B228149) control antibody (data not shown), in PBS for 30 min at 4° C. in the dark. Cells were washed twice with PBS, filtered through a 40 μm mesh and acquired using LSRII Fortessa flow cytometer (BD). Data analysis was performed using FlowJo Version 10.6 (Tree Star).

[0488] Survival Analysis

[0489] Tumour-bearing animals were checked for neurological symptoms and weighed weekly until day 21 after tumour cell implantation. From day 21 onwards, monitoring frequency was increased to daily checks and weekly bioluminescence imaging (BLI). Animals were taken euthanized via controlled CO.sub.2 asphyxation upon reaching predefined withdrawal criteria (weightloss over 20% of peak weight and/or moribund) according to cantonal veterinary authorities (ZH 194/19).

EXAMPLE 2: INTRACRANIAL INJECTION OF HUMAN IL-12 HAS HIGHER SYSTEMIC LEAKAGE THAN HIL-12FC

[0490] IL-12Fc for the local treatment of brain tumors is very promising. However, for use in clinical trials a human version of IL-12Fc is required that needs to show similar properties. To obtain a human analogue to murine IL-12IgG3 the inventors fused single chain human IL-12 to the crystallisable fragment (Fc) of human immunoglobulin G4 (hIgG4) (FIG. 1A). Similar to mIgG3, hIgG4 does not support antibody dependent cell-mediated cytotoxicity (ADCC) and does not activate the complement system. To test for leakage and stability of human IL-12Fc (hIL-12Fc) vs recombinant human IL-12 (rhIL-12), the inventors injected a single bolus into the striatum of transgenic mice that express the human FcRn on a murine FcRn-deficient background (FcRntg) (Postow et al., 2015, N Engl J Med 372:2006-2017; Kamran et al., 2016, Expert Opin Biol Ther 16:1245-1264). After 24 hours, the inventors analysed the concentration of human IL-12 in the lysate of the ipsilateral hemisphere and in the serum to learn more about the stability and retention at the site of injection (residual concentration) and the rate of leakage into the bloodstream (FIG. 1B). For each mouse, the inventors calculated the ratios of concentrations in the serum vs the concentration at the injection site as an estimate for tissue retention. Comparing serum levels to local concentration at the injection site, hIL-12Fc showed superior tissue retention over rhIL-12, as the inventors observed considerably lower ratios (FIG. 1C). For local GB treatment, a human IL-12Fc fusion cytokine seems to be a superior compound compared to its natural counterpart due to its higher tissue retention, stability and solubility.

EXAMPLE 3: FCRN BINDING LEADS TO SYSTEMIC ACCUMULATION OF IL-12FC

[0491] The neonatal Fc receptor (FcRn)-based endosomal recycling system in endothelial cells and red pulp macrophages prevents rapid degradation and clearance of IgG. After pinocytosis, facilitated by the acidic pH of endosomes, FcRn binds IgG and recycles it to the cell surface, where neutral pH induces its release. When injected locally, IL-12Fc can leak from the brain in an FcRn-mediated fashion due to its Fc tag. Leakage from the brain leads to IL-12Fc serum accumulation and may ultimately reach toxic levels. To test whether FcRn-based recycling indeed promotes accumulation of hIL-12Fc in the serum, the inventors utilized transgenic mice that express the human FcRn on a murine FcRn-deficient background (FcRntg). Since human FcRn has a weak affinity to murine IgG, but promotes normal albumin recycling, only murine IgG recycling is impaired in this mouse model. Murine IL-12Fc should therefore bind considerably less to FcRn in these FcRn humanized mice. Thus, the inventors compared serum mIL-12 levels of glioma-bearing wild type (wt) and FcRntg mice that were being treated with local murine IL-12Fc via osmotic minipumps. Indeed, after a week the inventors observed an increase in IL-12 levels in wt mice that was less pronounced in FcRntg mice (FIG. 2A) followed by an increase in IFN-γ levels (FIG. 2D). The inventors even observed elevated levels of IL-12 in the sera of some mice as early as day 1 after pump implantation (FIG. 2B). Consequently, this led to noticeable increase of IFN-γ serum levels in wt mice, which was not the case in the FcRntg cohort (FIG. 2C). Similar to murine IL-12Fc, human IL-12Fc will most likely also leak and accumulate, posing a threat for systemic side effects. Moreover, IFN-γ is one of the main mediators of the IL-12 related side effects (Leonard et al., 1997, Blood 90:2541-2548) and its persistent systemic presence can be toxic (Weiss et al., 2007, Expert Opin Biol Ther 7:1705-1721). Taken together, the inventors conclude that the leakage of even minute amounts of IL-12Fc from the treatment site is sufficient to trigger detectable serum IFN-γ levels.

EXAMPLE 4: GENERATION OF HUMAN IL-12FC VARIANTS DESIGNED FOR IMPROVED TISSUE RETENTION

[0492] The observation that reduced FcRn binding potentially abrogates export from the brain and leads to dramatically reduced recycling upon leakage out of the brain can be exploited to increase the safety margin of hIL-12Fc. The inventors therefore set out to reduce the binding of the Fc portion of hIL-12Fc to human FcRn. By increasing the positive charge at the FcRn binding interface of the Fc part this interaction at acidic pH—and hence the recycling—can be abrogated, which was shown to decrease serum half-life of immunoglobulins. The inventors introduced a number of mutations into hIL-12Fc (Table 1), at the FcRn binding site of hIL-12Fc with the aim of decreasing its serum half-life in case of leakage.

[0493] The inventors have generated three IL-12Fc variants with mutations analogous to the previously published antibodies with reduced FcRn affinity called IAQ, AHH and AAA. Furthermore, the inventors substituted the isoleucine at position 253 not to alanine, which represents a simple shortening of the sidechain, but changed it to asparagine instead (I253N). Asparagine is a polar amino acid, whose sidechain has a similar length as isoleucine. The inventors have also modified histidine at position 310 to alanine and at position 435 to glutamine, alanine or glutamic acid.

[0494] All the variants were expressed in human embryonic kidney 293T cell (HEK293T) cultures. Expression levels for all the variants were similar.

EXAMPLE 5: HUMAN IL-12FC VARIANTS HAVE SIMILAR PROTEIN STABILITY

[0495] First, the inventors have validated if the changes introduced to the Fc influence the overall protein stability. To this end the inventors have measured the denaturation temperature for each of the variant in a Thermal Shift Assay performed in PBS as well as in artificial cerebrospinal fluid (aCSF). The denaturation temperature for all the variants oscillated around 60° C. (FIG. 3 A). Measurements performed in aCSF confirmed that all the variants have similar stability, although the overall denaturation temperature was lower—approximately 57° C. (FIG. 3 B).

EXAMPLE 6: HUMAN IL-12FC VARIANTS SUSTAIN THEIR BIOLOGICAL ACTIVITY

[0496] Even though the inventors aimed to reduce the binding of hIL-12Fc to FcRn the inventors could not exclude that those changes had influences on the IL-12 biological activity. This was tested first using a HEK cell line stably transfected with IL-12 signaling components and a downstream enzyme catalyzing a colorimetric reaction. Only the NAQ variant showed approximately 2× reduced activity compared to IL-12Fc, whereas all the others had an EC50 in the range of IL-12Fc WT (FIG. 4 A). Importantly, IL-12Fc had comparable activity in vitro as rIL-12.

[0497] To further validate activity of IL-12Fc variants, the inventors performed activation of peripheral blood mononuclear cells (PBMCs) with three different hIL-12Fc variants, namely IAQ, AHQ and NHQ, followed by analysis of STAT-4 phosphorylation (FIG. 4 B). More importantly, this STAT-4 phosphorylation translated into robust production of IFN-γ (FIG. 4 C) 24h later, indicating that all of the variants retained the activity of rhIL-12.

EXAMPLE 7: HUMAN IL-12FC VARIANTS DIFFER IN BINDING TO PROTEIN G

[0498] Protein A and G affinity chromatography are among the standard methods used for purification of recombinant antibodies and Fc-fusion proteins. Modifying the interface between Fc and FcRn is known to abrogate Protein A binding, an observation that the inventors also confirmed with IL-12Fc variants. In order to confirm feasibility of production in a scaled-up process the inventors decided to validate the possibility of purifying IL-12Fc variants via Protein G affinity columns. The majority of the inventors' variants retained affinity to Protein G, but to the inventors' surprise, all the variants containing both I253N together with the H310A mutations were not suitable for protein G purification (Table 2). This effect was independent of additional mutations on position 435. For further studies the inventors have focused their attention on the variants with retained Protein G affinity.

EXAMPLE 8: HUMAN IL-12FC VARIANTS HAVE REDUCED FCRN AFFINITY

[0499] In order to validate the affinity of the IL-12Fc variants to FcRn the inventors used surface plasmon resonance (SPR), a label-free method to characterize protein-protein interactions. The inventors immobilized human FcRn and measured the binding of IL-12Fc variants at lysosomal pH ranges (pH=6.0) in various concentrations (43). As shown on FIG. 5 A, the majority of the modified IL-12Fc variants have lowered affinity to human FcRn, with the NHQ variant showing the strongest reduction (approximately 8× lower). The inventors used a commercially available human monoclonal anti-GFP IgG4 antibody as a control.

[0500] Furthermore, the inventors corroborated these data with ELISA data for the NHQ construct using IL-12Fc WT, anti-GFP IgG4 and the published variant IAQ as references. Both IAQ and NHQ showed reduced binding, with the NHQ having the lowest affinity (FIG. 5 B). The inventors thus conclude that the NHQ combination of substitutions seemed to reduce the binding to FcRn most dramatically. This is in contrast to results from Kenanova and colleagues (Kenanova et al., 2005, Cancer Research 65:622-631) who suggested that the combined mutations H310A and H435Q are responsible for the strongest reduction of binding to FcRn at low pH.

EXAMPLE 9: INTRODUCTION OF NHQ MUTATION REDUCES THE SYSTEMIC EXPOSURE TO LOCALLY DELIVERED HIL-12FC

[0501] The inventors hypothesized that reducing the FcRn affinity will increase the retention of hIL-12Fc in the CNS and at the same time prohibit its systemic accumulation. This was addressed in a similar fashion to the comparison of hIL-12Fc WT and recombinant human IL-12 (FIG. 1 B). FcRn.sup.tg mice were injected with a single dose of 1 pg of IL-12Fc WT or the NHQ variant and IL-12 was measured by ELISA in the ipsilateral brain hemisphere and in serum. Mice injected with the NHQ variant showed lowered serum-to-brain ratios compared to mice injected with hIL-12Fc WT (FIG. 6 A). The inventors postulate that this can be an effect of both increased retention in the CNS as well as attenuated systemic accumulation due to the FcRn-mediated recycling.

[0502] Furthermore, using CED instead of bolus injection, the inventors have compared the concentrations of hIL-12Fc WT, IAQ, AAA and NHQ in plasma and the injected hemisphere 24 h after CED and observed that the NHQ variant exhibits the most significantly reduced plasma to brain ratio (FIG. 6 B) also in optimized delivery settings compared to bolus injection. Such increased CNS retention merged with lowered systemic exposure could potentially improve the safety profile of local IL-12Fc therapy.

EXAMPLE 10: IL-12FC VARIANT NHQ HAS HIGHER BRAIN TISSUE RETENTION THAN OTHER LOW AFFINITY VARIANTS

[0503] Finally, the inventors measured the tissue retention after intracranial delivery of protein. To this end, the inventors injected 1 μg of unmodified IL-12Fc WT, the two previously published variants with reduced FcRn affinity, namely IAQ and AAA as well as NHQ, a variant with the lowest FcRn affinity according to the inventors' measurements (FIG. 5 A). In order to ensure maximal perfusion of the brain hemisphere, instead of a bolus injection of the protein solution, the inventors have used a CED protocol with a step catheter and a ramp-up injection regimen. To study the effect of different modifications at the FcRn binding interface in a most physiological setting, the inventors employed FcRn.sup.tg mice. As discussed earlier, FcRn is important for both the export from CNS and accumulation of Fc-containing molecules in the serum. In an attempt to decouple the two effects and focus solely on preventing the transport from CNS, the inventors have measured the amount of protein left in the brain 6 hours after the CED. Mice were euthanized, perfused with PBS, total protein in the ipsilateral hemisphere was isolated and hIL-12 was measured by ELISA. As shown on FIG. 7, IL-12Fc NHQ has superior tissue retention compared to IL-12Fc WT. Importantly, it was also better than IAQ and AAA, the two other variants with reduced FcRn affinity. Surprisingly, IAQ and AAA were not significantly different than IL-12Fc WT.

EXAMPLE 11: ANTI-TUMOR EFFECT IN VIVO

[0504] Human IL-12 is only poorly crossreactive with the murine IL-12 receptor. This implies, that for studying anti-tumor effect in vivo in a murine model, a surrogate molecule has to be used. In order to test the effects of the reduced affinity to FcRn, the inventors have fused single chain murine IL-12 to the same human IgG4 Fc as for hIL-12Fc (FIG. 8A).

[0505] IL-12 induces expression of IFNγ in target cells such as T cells and NK cells (Tugues et al., Cell death and differentiation (2015), 22:237-246)). IFNγ in turn can lead to upregulation of PD-L1 on myeloid cells and tumor cells in a process called adaptive resistance (O'Rourke et al., Sci. Transl. Med. (2017), (9), eaaa0984.). The inventors reasoned that PD-L1 would therefore serve as an induced anchor to further increase IL-12 tissue retention.

[0506] To assess the efficacy of IL-12Fc in combination with locally applied anti-PD-L1 antibody therapy, a bispecific Fc-fusion molecule was generated. It combines mIL-12hFc with an anti-PD-L1 half-antibody with a hIgG1 Fc containing NHQ mutation. The knobs-into-holes method was used for heterodimeric heavy chain assembly (Ridgway et al., Protein Eng (1996), 9:617-621). The anti-PD-L1 half-molecule is derived from atezolizumab, a clinically approved antibody and cross-reactive with murine and human PD-L1 (U.S. Pat. No. 8,217,149 B2) (FIG. 8A).

[0507] The inventors confirmed bioactivity of mIL-12hFc:aPD-L1 NHQ molecule in vitro: For IL-12 functionality, an IL-12-sensitive reporter cell line was used, IL-12 leads to secreted alkaline phosphatase, which in turn catalyzes a colorimetric reaction (FIG. 8B). Binding to cell bound PD-L1 was confirmed by flow-cytometry to detect binding of heterodimeric bifunctional constructs to PD-L1 on the surface of cells (FIG. 8C). The bifunctional heterodimeric constructs harbor the NHQ variant in their C.sub.H2 and C.sub.H3 domains and therefore FcRn binding is abrogated as confirmed by surface plasmon resonance and a comparably high K.sub.D value compared to the unmodified anti-PD-L1 antibody (FIG. 8D).

[0508] Following the in vitro characterization, the inventors continued to measure its properties in vivo. Using a murine glioma model GL-261, anti-tumor effects and systemic distribution were monitored in vivo. Briefly, tumour-bearing mice received two intracranial injections via CED with rmIL-12, mIL-12hFc:aPD-L1 NHQ, mIL-12hFc WT or NHQ or vehicle control (injection buffer only) (FIG. 9 A). Changes in tumor size were monitored using bioluminescent imaging, clinical impact was monitored via clinical scoring (FIG. 9B). To assess leakage and export upon CED, systemic IL-12 and IFNγ levels were measured in the blood plasma at various points in time (FIG. 9C). While animals receiving rmIL-12 or mIL-12hFc wt exhibited a sharp increase of systemic IL-12 signal closely followed by IFNγ upon CED, animals receiving mIL-12hFc NHQ or mIL-12hFc:aPD-L1 NHQ showed strongly reduced systemic IL-12 signals which rapidly returned to baseline and markedly reduced IFNγ signals (FIG. 9C). The difference in tissue retention between mIL-12hFc wt and mIL-12hFc NHQ already 6 hours after CED1 leads lower systemic IL-12 signals (FIG. 9D). Regarding the clinical course of treated animals, all groups receiving IL-12 constructs showed a marked increase in survival (FIG. 9E) compared to the control group, even at exceptionally late timepoint of intervention when disease was far progressed, 3 weeks after tumor inoculation. Of note, treatment response in groups receiving NHQ constructs (mIL-12hFc NHQ or mIL-12hFc:aPD-L1 NHQ) showed a severely reduced systemic IL-12 and IFNg albeit responding at least equally well to treatment when compared to groups receiving mIL-12hFc wt or rmIL-12.

EXAMPLE 12: AFFINITY MEASUREMENTS OF IL-12FC AND IGG VARIANTS TO HFCRN

[0509] To further evaluate the impact of low FcRn affinity on favorably influencing plasma to brain ratio upon local delivery to the CNS the IAQ, AAA and NHQ variants were compared to unmodified antibodies (FIG. 10). The inventors chose a human IgG1 directed against PD-L1 (FIG. 10A and FIG. 10B, atezolizumab,) and a human anti-Influenza A IgG4 antibody (FIG. 10C and FIG. 10D, Flu HA3.1, US2014/0370032A1).

[0510] The finding that hIL-12Fc is functional, has higher tissue retention than rhIL-12 and that abrogation of systemic recycling can increase the safety margin in case of leakage has potentially wide implications for the local administration of any Fc containing molecule. These modifications enable safe and efficacious local delivery of any antibody or Fc-fusion molecule for the local treatment of neurologic diseases.

[0511] Administration of therapeutics into the CNS via the systemic route (either per os or i.v.) is challenging mainly because of the BBB and—compared to the rest of the body—only a small selection of today's therapeutics actually reaches the brain. Unfortunately, antibodies and Fc containing biologics, particularly Fc-fusion proteins, do not readily cross the BBB and in addition are actively exported. Enabling transport of antibodies over the BBB into the brain parenchyma is being extensively studied, e.g. by exploiting receptor mediated transcytosis of transferrin. Cytokines have a short half-life in circulation and bear a high risk of adverse effects, which narrows their therapeutic opportunity window. Cytokines can be linked to antibodies homing to tumors, where they will accumulate, particularly NHS-IL-12. Even after subcutaneous dosing, these antibodies induce an IFNγ response as they travel to the tumor via the bloodstream. Initially, systemic delivery of IL-12 was assessed for treatment of non-brain cancers. However, these clinical trials had to be prematurely terminated, since—at effective doses—intravenous application led to serious adverse events, including deaths. One of the main reasons seems to have been the induction of IFNγ by IL-12.

[0512] The serum half-life and solubility of protein therapeutics can be improved by direct fusion of the therapeutic moiety with the crystallizable fragment (Fc) of antibodies. For direct local application in anatomically distinct locations this can lead also to less desirable effects. One of these can be FcRn-mediated export of Fc-containing molecules from immune privileged anatomical sites, particularly the brain and their serum accumulation analogous to IgG recycling.

[0513] The inventors have observed that local administration of an IL-12Fc fusion cytokine into the brain triggers FcRn dependent export of IL-12Fc through the BBB into the circulation. IL-12Fc accumulates in the blood and triggers potentially dangerous IFNγ production.

[0514] The inventors found that IL-12Fc with reduced FcRn affinity is functional and has higher tissue retention than recombinant IL-12, as well as unmodified IL-12Fc. When compared in a brain tissue retention experiment, the NHQ mutant was the only one with improved retention over IL-12Fc WT. Surprisingly, IAQ and AAA, two variants reported to have dramatically reduced FcRn binding were not different than unmodified IL-12Fc, suggesting that in order to obtain biological difference, the FcRn affinity must be reduced over a given threshold, that only the NHQ modification reaches. Alternatively, it cannot be ruled out that the NHQ mutations introduce other features that improve tissue retention in an FcRn-independent way.

[0515] This translates into an improved safety profile and broadens the therapeutic window for the IL-12Fc therapy of brain tumors. Moreover, the inventors' findings can be translated to any Fc containing therapeutics, primarily therapeutic antibodies where there is a strong rationale for local intracranial administration. Such an application route would be preferred due to weak efficacy when given systemically, potentially an effect of poor crossing through the BBB, or because the desired therapeutic effect should be contained locally. Local therapy with biologics optimized for such delivery should preclude the systemic toxicity and thus improve the safety profile of the drug.

TABLE-US-00001 TABLE 1 List of the mutations introduced to the Fc part of IL-12Fc. Amino acid positions numbered according to EU numbering system (Edelman et al. Proceedings of the National Academy of Sciences of the United States of America (1969) 63(1):78-85). Name SEQ ID NO. Position 253 Position 310 Position 435 Fc WT SEQ ID NO. 001 I H H IAQ SEQ ID NO. 002 I A Q AHQ SEQ ID NO. 003 A H Q NHQ SEQ ID NO. 004 N H Q AAQ SEQ ID NO. 005 A A Q NAQ SEQ ID NO. 006 N A Q AHH SEQ ID NO. 007 A H H NHH SEQ ID NO. 008 N H H AAH SEQ ID NO. 009 A A H NAH SEQ ID NO. 010 N A H NAA SEQ ID NO. 011 N A A NAE SEQ ID NO. 012 N A E AAA SEQ ID NO. 013 A A A AAE SEQ ID NO. 014 A A E

TABLE-US-00002 TABLE 2 List of IL-12Fc variants and their ability to bind to Protein G. Retained affinity to the Protein G Name SEQ ID NO. chromatography bead: Fc wt SEQ ID NO. 001 + IAQ SEQ ID NO. 002 + AHQ SEQ ID NO. 003 + NHQ SEQ ID NO. 004 + AAQ SEQ ID NO. 005 + NAQ SEQ ID NO. 006 − AHH SEQ ID NO. 007 + NHH SEQ ID NO. 008 + AAH SEQ ID NO. 009 + NAH SEQ ID NO. 010 − NAA SEQ ID NO. 011 − NAE SEQ ID NO. 012 − AAA SEQ ID NO. 013 + AAE SEQ ID NO. 014 +

TABLE-US-00003 TABLE 3 List of sequences of molecules, which combination makes a bispecific antibody or antibody-like molecule binding to human or mouse IL-12 receptor, in particular in an agonistic manner, and human or mouse PD-L1. Name SEQ ID NO. mIL-12Fc-IgG4 NHQ Hole SEQ ID NO. 015 hIL-12Fc-IgG4 NHQ Hole SEQ ID NO. 016 mIL-12Fc-IgG1 short NHQ Hole SEQ ID NO. 017 hIL-12Fc-IgG1 short NHQ Hole SEQ ID NO. 018 mIL-12Fc-IgG1 long NHQ Hole SEQ ID NO. 019 hIL-12Fc-IgG1 long NHQ Hole SEQ ID NO. 020 anti-PD-L1 scFv-IgG4 NHQ Knob SEQ ID NO. 021 anti-PD-L1 scFv-IgG1 NHQ Knob SEQ ID NO. 022 anti-PD-L1 HC-IgG1 NHQ Knob SEQ ID NO. 023 anti-PD-L1 LC-IgG1 SEQ ID NO. 024

[0516] A combined bispecific molecule may consist of molecules described as sequence SEQ ID NO 15 with SEQ ID NO 21, SEQ ID NO 16 with SEQ ID NO 21, SEQ ID NO 17 with SEQ ID NO 22, SEQ ID NO 18 with SEQ ID NO 22, SEQ ID NO 17 with SEQ ID NO 23 and SEQ ID NO 24, SEQ ID NO 18 with SEQ ID NO 23 and 24, SEQ ID NO 19 with SEQ ID NO 22, SEQ ID NO 20 with SEQ ID NO 22, SEQ ID NO 19 with SEQ ID NO 23 and SEQ ID NO 24, SEQ ID NO 20 with SEQ ID NO 23 and SEQ ID NO 24.

[0517] Items [0518] 1. A polypeptide comprising a crystallizable fragment (Fc) region of IgG, for use as a medicament (use in medicine), wherein [0519] said Fc region bears a modification resulting in reduced affinity to the neonatal Fc receptor (FcRn) and [0520] said polypeptide is delivered by local administration to the tissue affected by the disease. [0521] 2. The polypeptide for use as a medicament according to item 1, wherein the polypeptide is delivered [0522] by intracranial administration, [0523] by intrathecal administration, or [0524] by intraocular administration. [0525] 3. The polypeptide for use as a medicament according to any one of the above items, wherein the polypeptide is administered by intracranial administration and the serum or plasma to brain concentration ratio of said polypeptide is below a predetermined threshold selected from [0526] a. at most ⅔ of the serum or plasma to brain concentration ratio of the same polypeptide comprising a non-modified Fc region, [0527] b. at most ⅛ of the serum or plasma to brain concentration ratio of the same polypeptide neither comprising an Fc region nor peptide linkers, [0528] measurable 24 h after intracranial bolus injection into the striatum of FcRn.sup.tg mice. [0529] 4. The polypeptide for use as a medicament according to any one of the above items, wherein the polypeptide is administered by intracerebroventricular or intrathecal administration and the serum or plasma to CSF concentration ratio of said polypeptide is below a predetermined threshold selected from [0530] c. at most ⅔ of the serum or plasma to CSF concentration ratio of the same polypeptide comprising a non-modified Fc region, [0531] d. at most ⅛ of the serum or plasma to CSF concentration ratio of the same polypeptide neither comprising an Fc region nor peptide linkers, [0532] measurable 24 h after intracerebroventricular or intrathecal injection of FcRn.sup.tg mice. [0533] 5. The polypeptide for use as a medicament according to any one of the above items, wherein said reduced affinity of said polypeptide to FcRn is characterized by a dissociation constant (K.sub.D) selected from [0534] a. a K.sub.D that is at least 2×, particularly at least 3×, more particularly at least 4×, even more particularly at least 5× increased compared to a K.sub.D characterizing binding of FcRn to the same polypeptide comprising a non-modified Fc region, and [0535] b. a K.sub.D that is at least 1.5×, particularly at least 2× increased compared to a K.sub.D characterizing binding of FcRn to the same polypeptide comprising a differently modified Fc region, namely one mutant selected from IAQ (bearing the mutations H310A and H435Q) and AAA (bearing the mutations I253A, H310A and H435A). [0536] 6. The polypeptide for use as a medicament according to any one of the above items, wherein intracranial delivery is effected by a method selected from [0537] a. single, intermittent or continous local infusion, including convection enhanced delivery (CED), [0538] b. intrathekal administration, [0539] c. in situ production of said polypeptide, [0540] d. release from implanted slow release formulations, [0541] e. molecular transport into the CNS, [0542] f. cellular transport into the CNS, or [0543] g. transport to the CNS after intranasal application. [0544] 7. The polypeptide for use as a medicament according to any one of the above items, for treatment or prevention of a disease that affects the central nervous system. [0545] 8. The polypeptide for use as a medicament according to item 7, wherein said disease affecting the central nervous system is a malignant disease, particularly a glioma, more particularly a high grade glioma (HGG). [0546] 9. The polypeptide for use as a medicament according to any one of the above items, wherein said Fc region is a human Fc region or a chimeric Fc region comprising a human amino acid sequence and bears a mutation at position 253, in particular I253A or I253N, more particularly I253N. [0547] 10. The polypeptide for use as a medicament according to item 9, wherein said Fc region does not bear a mutation at position 310. [0548] 11. The polypeptide for use as a medicament according to any one of the above items, wherein said Fc region comprises [0549] the mutations H310A and H435Q (IAQ); [0550] the mutations I253A and H435Q, and an H at position 310 (AHQ); [0551] the mutations I253N and H435Q, and an H at position 310 (NHQ); [0552] the mutations I253A, H310A and H435Q (AAQ); [0553] the mutations I253N, H310A and H435Q (NAQ); [0554] the mutation I253A and an H at position 310 and 435 (AHH); [0555] the mutation I253N and an H at position 310 and 435 (NHH); [0556] the mutations I253A and H310A, and an H at position 435 (AAH); [0557] the mutations I253N and H310A, and an H at position 435 (NAH); [0558] the mutations I253N, H310A and H435A (NAA); [0559] the mutations I253N, H310A and H435E (NAE); [0560] the mutations I253A, H310A and H435A (AAA); or the mutations I253A, H310A and H435E (AAE). [0561] 12. The polypeptide for use as a medicament according to any one of the above items, wherein said Fc region comprises [0562] the mutations I253N and H435Q, and an H at position 310 (NHQ); [0563] the mutations I253A, H310A and H435Q (AAQ); [0564] the mutations I253N, H310A and H435Q (NAQ); [0565] the mutations I253N, H310A and H435E (NAE); or [0566] the mutations I253A, H310A and H435E (AAE), [0567] particularly said Fc region comprises the mutations I253N and H435Q, and an H at position 310 (NHQ). [0568] 13. The polypeptide for use as a medicament according to any one of the above items, wherein said Fc region is or comprises a sequence characterized by SEQ ID NO 002 (IAQ), SEQ ID NO 003 (AHQ), SEQ ID NO 004 (NHQ), SEQ ID NO 005 (AAQ), SEQ ID NO 006 (NAQ), SEQ ID NO 007 (AHH), SEQ ID NO 008 (NHH), SEQ ID NO 009 (AAH), SEQ ID NO 010 (NAH), SEQ ID NO 011 (NAA), SEQ ID NO 012 (NAE), SEQ ID NO 013 (AAA) or SEQ ID NO 014 (AAE). [0569] 14. The polypeptide for use as a medicament according to any one of the above items, wherein said Fc region is or comprises a sequence characterized by SEQ ID NO 004 (NHQ), SEQ ID NO 005 (AAQ), SEQ ID NO 006 (NAQ), SEQ ID NO 012 (NAE) or SEQ ID NO 014 (AAE). [0570] 15. The polypeptide for use as a medicament according to any one of the above items, wherein said polypeptide is selected from [0571] a. a fusion protein comprising [0572] i. an effector polypeptide and [0573] ii. said Fc region; or [0574] b. an antibody or antibody-like molecule comprising said Fc region. [0575] 16. The polypeptide for use as a medicament according to any one of the above items, wherein said polypeptide is selected from [0576] a. a bispecific, trispecific or multispecific antibody or antibody-like molecule, particularly a bispecific antibody or antibody-like molecule specifically binding to [0577] i. CD3 and a tumor-associated antigen, [0578] ii. Histone and IL12 receptor in an agonistic manner; [0579] iii. PD-L1 and 4-1BB, [0580] iv. PD-L1 and CD28, [0581] v. PD-L1 and IL12 receptor in an agonistic manner, or [0582] vi. a tumor-associated antigen and IL12 receptor in an agonistic manner, [0583] b. an armed antibody or antibody-like molecule comprising an effector polypeptide, or [0584] c. a tumor conditional or tissue conditional antibody or antibody-like molecule comprising a shielding domain and a cleavable protease sensitive linker peptide. [0585] 17. The polypeptide for use as a medicament according to item 16, wherein said polypeptide is a bispecific, trispecific or multispecific antibody or antibody-like molecule, particularly a bispecific antibody or antibody-like molecule specifically binding to PD-L1 and comprising [0586] i. an effector polypeptide, [0587] ii. IL-12Fc, [0588] iii. a combination of a molecule being characterized by a sequence selected from SEQ ID NO. 015-016 and a molecule being characterized by a sequence of SEQ ID NO. 021, [0589] iv. a combination of a molecule being characterized by a sequence selected from SEQ ID NO. 017-020 and a molecule being characterized by a sequence of SEQ ID NO. 022; or [0590] v. a combination of a molecule being characterized by a sequence selected from SEQ ID NO. 017-020, a molecule being characterized by a sequence of SEQ ID NO. 023 and a molecule being characterized by a sequence of SEQ ID NO.024. [0591] 18. The polypeptide for use as a medicament according to item 16, wherein the polypeptide is administered by intracranial administration and the serum or plasma to brain concentration ratio of said polypeptide is below a predetermined threshold selected from [0592] a. at most ⅛ of the serum or plasma to brain concentration ratio of the same polypeptide comprising a non-modified Fc region, [0593] b. at most 1/20 of the serum or plasma to brain concentration ratio of the same polypeptide neither comprising an Fc region nor peptide linkers, measurable 24 h after intracranial bolus injection into the striatum of FcRn.sup.tg mice. [0594] 19. The polypeptide for use as a medicament according to any one of items 15 to 18, wherein said effector polypeptide is selected from hIL-12, hIL-10, hIL-2, hIL-7, IFNα, IFNβ, IFNγ, hIL-15, TNFα, CTLA-4, TGFβ, TGFβII, GDNF, hIL-35, CD95, hIL-1RA, hIL-4, hIL-13, SIRPα, G-CSF, GM-CSF, OX40L, CD80, CD86, GITRL, 4-1BBL, EphrinA1, EphrinB2, EphrinB5, BDNF, C9orf72, NRTN, ARTN, PSPN, CNTF, TRAIL, IL-4, IL-3, IL-1, IL-5, IL-8, IL-18, IL-21, CCL5, CCL21, CCL10, CCL16, CX3CL1, and CXCL16, in particular said effector polypeptide is hIL-12. [0595] 20. The polypeptide for use as a medicament according to any one of items 15 to 18, wherein said antibody or antibody-like molecule is selected from an antibody or antibody-like molecule specifically binding to PD-L1, TNFα, Histone, IFNγ, CXCL10, CTLA4, PD-1, OX40 CD3, CD25, CD28, TREM2, IL-6, CX3CR1, CD25, Nogo-A, CD27, IL-12, IL-12Rb1, IL-23, CD47, TGFβ, EGFR, EGFRvIII, Her2, PDGFR, TGFR, FGFR, IL-4RA, TfR, LfR, IR, LDL-R, LRP-1, CD133, CD111, VEGFR, VEGF-A, Ang-2, IL-10, IL-10R, IL-13Rα2, α-synuclein, CSF1R, GITR, TIM-3, LAG-3, TIGIT, BTLA, VISTA, CD96, 4-1BB, CCL2, IL-1 or IL-1R, EphA2, EphA3, EphB2, EphB3, EphB4, LINGO-1, L1CAM, NCAM, SOD-1, SIGMAR-1, SIGMAR-2, TDP-43, Aβ, Tau, IFNα, IFNβ, TRPM4, ASIC1, VGCCs, CB.sub.1, TTR, HTT, JCV, or C9orf72, in particular said antibody or antibody-like molecule is an antibody specifically binding to PD-L1, OX40, CD47 or Nogo-A. [0596] 21. An antibody or antibody-like molecule specifically binding to OX40 in an agonistic fashion comprising a crystallizable fragment (Fc) region of IgG, for use in prevention or treatment of a disease affecting the central nervous system, wherein [0597] said Fc region bears a modification resulting in reduced affinity to the neonatal Fc receptor (FcRn) and [0598] said antibody or antibody-like molecule is administered to the brain. [0599] 22. The antibody or antibody-like molecule for use in prevention or treatment of a disease affecting the central nervous system according to item 21, wherein the serum or plasma to brain concentration ratio of said antibody or antibody-like molecule is below a predetermined threshold is selected from [0600] a. at most ⅔ of the serum or plasma to brain concentration ratio of the same polypeptide comprising a non-modified Fc region, [0601] b. at most ⅛ of the serum or plasma to brain concentration ratio of the same polypeptide neither comprising an Fc region nor peptide linkers, measurable 24 h after intracranial bolus injection or CED into the striatum of FcRn.sup.tg mice. [0602] 23. The antibody or antibody-like molecule for use in prevention or treatment of a disease affecting the central nervous system according to any one of items 21 to 22, wherein said reduced affinity of said antibody or antibody-like molecule to FcRn is characterized by a dissociation constant (K.sub.D) selected from [0603] a. a K.sub.D that is at least 2×, particularly at least 3×, more particularly at least 4×, even more particularly at least 5× increased compared to a K.sub.D characterizing binding of FcRn to the same antibody or antibody-like molecule comprising a non-modified Fc region, and [0604] b. a K.sub.D that is at least 1.5×, particularly at least 2× increased compared to a K.sub.D characterizing binding of FcRn to the same antibody or antibody-like molecule comprising a differently modified Fc region, namely one mutant selected from IAQ (bearing the mutations H310A and H435Q) and AAA (bearing the mutations I253A, H310A and H435A) [0605] 24. The antibody or antibody-like molecule for use in treatment or prevention of a disease affecting the central nervous system according to any one of items 21 to 23, wherein said intracranial delivery is effected by a method selected from [0606] a. single, intermittent or continuous local infusion, including convection enhanced delivery (CED), [0607] b. in situ production of said polypeptide, [0608] c. intrathekal or intracerebroventricular administration, [0609] d. release from implanted slow release formulations, [0610] e. molecular transport into the cns, [0611] f. cellular transport into the CNS, or [0612] g. transport to the CNS after intranasal application. [0613] 25. The antibody or antibody-like molecule for use in treatment or prevention of a disease affecting the central nervous system according to any one of items 21 to 24, wherein said disease affecting the central nervous system is a malignant disease, particularly a glioma, more particularly a high grade glioma (HGG). [0614] 26. The antibody or antibody-like molecule for use in prevention or treatment of a disease affecting the central nervous system to any one of items 21 to 25, wherein said Fc region is a human Fc region or a chimeric Fc region comprising a human amino acid sequence and bears a mutation at position 253 [Kabat numbering system], in particular I253A or I253N, more particularly I253N. [0615] 27. The antibody or antibody-like molecule for use in prevention or treatment of a disease affecting the central nervous system according to item 26, wherein said Fc region does not bear a mutation at position 310. [0616] 28. The antibody or antibody-like molecule for use in prevention or treatment of a disease affecting the central nervous system according to any one of items 21 to 27, wherein said Fc region comprises [0617] the mutations H310A and H435Q (IAQ); [0618] the mutations I253A and H435Q, and an H at position 310 (AHQ); [0619] the mutations I253N and H435Q, and an H at position 310 (NHQ); [0620] the mutations I253A, H310A and H435Q (AAQ); [0621] the mutations I253N, H310A and H435Q (NAQ); [0622] the mutation I253A and an H at position 310 and 435 (AHH); [0623] the mutation I253N and an H at position 310 and 435 (NHH); [0624] the mutations I253A and H310A, and an H at position 435 (AAH); [0625] the mutations I253N and H310A, and an H at position 435 (NAH); [0626] the mutations I253N, H310A and H435A (NAA); [0627] the mutations I253N, H310A and H435E (NAE); [0628] the mutations I253A, H310A and H435A (AAA); or [0629] the mutations I253A, H310A and H435E (AAE). [0630] 29. The antibody or antibody-like molecule for use in prevention or treatment of a disease affecting the central nervous system according to any one of items 21 to 28, wherein said Fc region comprises [0631] the mutations I253N and H435Q, and an H at position 310 (NHQ); [0632] the mutations I253A, H310A and H435Q (AAQ); [0633] the mutations I253N, H310A and H435Q (NAQ); [0634] the mutations I253N, H310A and H435E (NAE); or [0635] the mutations I253A, H310A and H435E (AAE). [0636] 30. The antibody or antibody-like molecule for use in prevention or treatment of a disease affecting the central nervous system according to any one of items 21 to 29, wherein said Fc region is or comprises a sequence characterized by SEQ ID NO 002 (IAQ), SEQ ID NO 003 (AHQ), SEQ ID NO 004 (NHQ), SEQ ID NO 005 (AAQ), SEQ ID NO 006 (NAQ), SEQ ID NO 007 (AHH), SEQ ID NO 008 (NHH), SEQ ID NO 009 (AAH), SEQ ID NO 010 (NAH), SEQ ID NO 011 (NAA), SEQ ID NO 012 (NAE), SEQ ID NO 013 (AAA) or SEQ ID NO 014 (AAE). [0637] 31. The antibody or antibody-like molecule for use in prevention or treatment of a disease affecting the central nervous system according to any one of items 21 to 30, wherein said Fc region is or comprises a sequence characterized by SEQ ID NO 004 (NHQ), SEQ ID NO 005 (AAQ), SEQ ID NO 006 (NAQ), SEQ ID NO 012 (NAE) or SEQ ID NO 014 (AAE).