Outcome and feasibility of radiotherapy bridging in large B-cell lymphoma patients receiving CD19 CAR T in the UK.

Radiotherapy (RT) has potential synergistic effects with chimeric antigen receptor (CAR) T but is not widely used as bridging therapy due to logistical challenges and lack of standardised protocols. We analysed RT bridging in a multicentre national cohort of large B-cell lymphoma patients approved for 3L axicabtagene ciloleucel or tisagenlecleucel across 12 UK centres. Of 763 approved patients, 722 were leukapheresed, 717 had data available on bridging therapy. 169/717 (24%) received RT bridging, 129 as single modality and 40 as combined modality treatment (CMT). Of 169 patients, 65.7% had advanced stage, 36.9% bulky disease, 86.5% elevated LDH, 41.7% international prognostic index (IPI) ≥3 and 15.2% double/triple hit at the time of approval. Use of RT bridging varied from 11% to 32% between centres and increased over time. Vein-to-vein time and infusion rate did not differ between bridging modalities. RT-bridged patients had favourable outcomes with 1-year progression-free survival (PFS) of 56% for single modality and 47% for CMT (1-year PFS 43% for systemic bridging). This is the largest cohort of LBCL patients receiving RT bridging prior to CAR T reported to date. Our results show that RT bridging can be safely and effectively used even in advanced stage and high-risk disease, with low dropout rates and excellent outcomes.

Shutdown of immunological priming and presentation after in vivo administration of adenovirus.

Adenoviral (Adv) vectors are widely used in both experimental and clinical trials for vaccination and gene therapy. Recombinant Adv can evoke potent innate immune responses and adaptive immune responses to encoded antigens. However, how Adv infection affects the response to subsequently encountered antigens is poorly understood. We show that intravenously administered replication defective (E1 and E3 deleted) Adv educes functional changes in dendritic cells (DC) resulting in impaired priming of cytotoxic T lymphocytes (CTL) more than 7 days after Adv treatment. Generalized DC activation was indicated by transient upregulation of CD86 and reduced endocytosis of fluorescent beads. It is known that CD8+ DC are predominantly responsible for uptake and presentation (cross-presentation) of exogenous antigens to CD8+ CTL. Hence, impaired endocytosis in CD8+, but not CD8-, DC at 7 days after Adv administration provided an explanation for the impaired CTL response to antigen at this time. Shutdown of cross-presentation was confirmed using cytochrome c (cytc), an agent that selectively depletes cross-presenting DC. Adv-infection rendered CD8+ DC resistant to depletion by cytc. As the cross-presentation pathway underlies CD8 T-cell responses to many cancers and to vaccines or viruses that do not directly infect DC, systemic Adv administration may impair these responses.

Dietary protein restriction in artificially reared neonatal rats causes a reduction of insulin-like growth factor-I gene expression.

To determine in neonates the effects of protein restriction on growth, serum IGF-I, and IGF-I gene expression, we adapted a technique for rearing neonatal rats (days 6-18 of life) artificially by continuous infusion of milk through a gastrostomy. The artificially reared (AR) animals were given isocaloric diets containing 8%, 13.5%, or 18% lactalbumin protein. The AR rats were compared to rats reared by their mothers (MR) for 18 days. The growth of AR rats was related to the amount of dietary protein, with the pups given 18% protein having the best growth (25.55 g gained during the 12 days of the study) and those given 8% protein having the worst (13.42 g). The 13.5% protein-fed animals were intermediate in weight gain (18.39 g). The weight gains of the 18% and 8% protein-fed pups were significantly different from that of the MR animals (18.37 g). An identical pattern of tail length growth was noted among the groups. Mean serum IGF-I concentrations followed the same pattern (MR, 1.66 U/ml on day 18 of life; 18% AR, 2.53; 13.5% AR, 1.52; 8% AR, 1.31). Liver IGF-I mRNA was rank-ordered identically with weight gain and serum IGF-I [MR, 23.10 pg/micrograms poly(A+) RNA; 18% AR, 27.66; 13.5% AR, 21.02; 8% AR, 18.76]. Unexpectedly, the 7.5-kilobase IGF-I mRNA size class showed a 2- to 3-fold higher abundance in all groups of AR rats compared to that in MR controls (P less than 0.01), suggesting that this IGF-I size class is regulated independently of the other species. The reductions in serum IGF-I and IGF-I mRNA during protein restriction of neonatal rats suggest that these responses are mediated by decreased IGF-I gene expression at the level of transcription or RNA stabilization.

Nucleic acid vaccines: tasks and tactics.

There are no adequate vaccines against some of the new or reemerged infectious scourges such as HIV and TB. They may require strong and enduring cell-mediated immunity to be elicited. This is quite a task, as the only known basis of protection by current commercial vaccines is antibody. As DNA or RNA vaccines may induce both cell-mediated and humoral immunity, great interest has been shown in them. However, doubt remains whether their efficacy will suffice for their clinical realization. We look at the various tactics to increase the potency of nucleic acid vaccines and divided them broadly under those affecting delivery and those affecting immune induction. For delivery, we have considered ways of improving uptake and the use of bacterial, replicon or viral vectors. For immune induction, we considered aspects of immunostimulatory CpG motifs, coinjection of cytokines or costimulators and alterations of the antigen, its cellular localization and its anatomical localization including the use of ligand-targeting to lymphoid tissue. We also thought that mucosal application of DNA deserved a separate section. In this review, we have taken the liberty to discuss these enhancement methods, whenever possible, in the context of the underlying mechanisms that might argue for or against these strategies.

The need for IgG2c specific antiserum when isotyping antibodies from C57BL/6 and NOD mice.

Isotyping and quantitation of murine IgG2a antibodies are widely performed with commercial monoclonal and polyclonal antisera raised against BALB/c IgG2a myeloma proteins. Recently it became evident that inbred mouse strains with the Igh1-b allele do not have the gene for IgG2a and instead express the IgG2c isotype. We show that commercial anti-IgG2a sera cross-react inadequately against IgG2c in immunoblot and ELISA and hence, are not suitable to detect and measure this subclass in mouse strains such as C57BL/6, C57BL/10 and NOD. We have used DNA immunization to generate polyclonal anti-IgG2c serum and demonstrated that it is essential to use IgG2c-specific antiserum to quantify accurately isotypic responses in mouse strains with the Igh1-b allele.

Additive efficacy of CTLA4Ig and OX40Ig secreted by genetically modified grafts.

BACKGROUND: The use of systemic immunosuppressive drugs have been paramount in the success in transplantation, but there are serious deleterious effects. Genetic modification of grafts to secrete immunomodulators locally may be a way to reduce the need for systemic immunosuppression. METHODS AND RESULTS: An insulinoma cell line, NIT, having the nonobese diabetic (NOD) genotype but also expressing the SV40 large T Ag, was transfected with CTLA4Ig or OX40Ig in an attempt to block signals in the costimulatory/adhesion pathways. The extracellular domains of these molecules have been fused to the Fc of IgG2c derived from the NOD mouse strain. This resulted in secreted and dimerized proteins. SV40 T Ag is potent at inducing graft rejection. Test and control transfectants were transplanted subcutaneously into young NOD mice to determine whether secretion of CTLA4Ig and OX40Ig would promote survival of the insulinoma graft. In immunodeficient mice, cell growth was similar for all transfectants. However, in immunocompetent NOD mice, the survival/growth of test grafts was significantly better than that of controls. By combining test transfectants, we found that graft survival was enhanced in an additive and significant fashion. In vitro, there was a significant reduction in immune responses-compared with control-when purified fusion proteins were added to mixed leukocyte reaction cultures. CONCLUSIONS: We conclude that blockade of individual costimulatory/adhesion signals by graft manipulation can contribute to transplantation success and that blockade of combinations of signals in these pathways enhances this success. Successful immunomodulation by the graft itself can be achieved.

Local production of anti-CD4 antibody by transgenic allogeneic grafts affords partial protection.

BACKGROUND: Immunosuppressive drugs and anti-lymphocyte antibody are used clinically to suppress cellular rejection responses. However, these systemic regimens often led to general immunodeficiency and thus increased susceptibility to opportunistic infection and neoplasia. Immunosuppressive molecules delivered locally may be a way of inhibiting rejection responses, whereas systemic immunity is preserved. To achieve protective local immunosuppression, we produced a graft secreting its own immunomodulator, by deriving transgenic mice expressing a chimeric anti-CD4 antibody (GK2c) in the pancreas. METHODS AND RESULTS: Transgenic mice in bml genetic background expressing a modified anti-mouse CD4 antibody (GK2c) under two promoters have been produced. Tissue expression of GK2c was detected by immunoperoxidase staining. Under the cytomegalovirus promoter, there was abundant GK2c expression in pancreatic exocrine tissue. Under the rat preproinsulin II promoter, there was abundant GK2c expression in pancreatic endocrine tissue only. High-expression transgenic lines had 10-100 microg/ml GK2c in blood plasma. By flow cytometry, these transgenic mice were devoid of CD4+ cells in their peripheral lymphoid organs. To test transgenic mice as donors, fetal pancreata from transgenic mice were grafted into fully allogeneic CBA mice under the kidney capsule, transgenic grafts had prolonged survival compared with control non-transgenic grafts. Furthermore, GK2c transgenic grafts had reduced infiltration with an absence of CD4+ cells at the graft site without any effect on the cell composition in lymphatic tissues. CONCLUSION: Transgenic grafts that secrete anti-CD4 antibody can afford some protection against graft rejection, while only affecting the CD4 population at the graft site.

Local secretion of a chimeric anti-CD4 antibody protects against graft rejection in the NOD mouse.

BACKGROUND: Engineering a graft to secrete its own immunosuppressive antibodies may minimize the risks associated with current high dose systemic immunosuppression. METHODS AND RESULTS: A beta cell insulinoma cell line (NIT-1) was transfected with genes encoding a chimeric anti-CD4 antibody. The NIT-1 cells secreted functional chimeric anti-CD4 antibody that bound to the CD4 molecule on mouse thymocytes and inhibited in vitro proliferation of CD4+ve T cells. Both test and control transfected cell lines grew at a similar rate in immunodeficient mice. In immunocompetent NOD mice, NIT-1 cells are normally rejected by a cellular immune response against the SV40 T antigen. Although control transfected NIT-1 cells were rapidly rejected by NOD mice, anti-CD4 secreting NIT-1 cells grew significantly better and were able to form tumors at the site of injection. CONCLUSIONS: The local secretion of chimeric anti-CD4 antibody from transfected cells can contribute to graft survival in our transplantation model.

Protective effect of CTLA4Ig secreted by transgenic fetal pancreas allografts.

BACKGROUND: Pancreas allotransplantation offers a cure for insulin-dependent diabetes mellitus. Systemic immunosuppression used to prevent immune destruction of the graft has side-effects, including increased susceptibility to infection and neoplasia. These unwanted effects may be limited by engineering the graft to secrete immunomodulatory molecules, to achieve local immunosuppression. Several studies have shown that transient local CTLA4Ig results in partial protection of allogeneic grafts. Our intent has been to determine whether sustained secretion of transgenic CTLA4Ig from pancreatic islets is able to protect against allograft rejection. METHODS AND RESULTS: Mouse CTLA4 (test=CTLA4Ig) or CD5 leader sequence (control=CD5LIg) was fused to the Fc of mouse IgG2c, and expressed transgenically under the control of the rat insulin promoter in C57BL/6 mice carrying the bml mutation of H-2K(b) (B6.C-H-2(bm1)). This resulted in expression in pancreatic islets. We used ELISA quantification of transgene products secreted into the supernatants of cultured fetal pancreata to select high (CTLA4Ig(hi)) and low (CTLA4Ig(lo)) expresser transgenic mice. Cultured fetal pancreata were transplanted under the kidney capsule of wholly allogeneic CBA recipient mice. CTLA4Ig(hi) but not CTLA4Ig(lo) expresser grafts showed enhanced survival compared with control CD5LIg grafts at 6 weeks posttransplant, provided the recipient mice were transiently depleted of CD4 T cells (by a single low-dose injection of GK1.5) before transplantation. CONCLUSIONS: Sustained local secretion of CTLA4Ig from transgenic grafts in combination with transient systemic CD4 T-cell depletion can enhance allograft acceptance.

Secretion of CTLA4Ig by an SV40 T antigen-transformed islet cell line inhibits graft rejection against the neoantigen.

In a model of transplantation rejection, we tested whether a graft manipulated to secrete an immunomodulator could protect itself from immune destruction, thus waiving the need for administration of exogenous immunosuppressants to the recipient. An insulinoma cell line, NIT, having the nonobese diabetic (NOD) genotype but also expressing the SV40 large T antigen, was transfected with CTLA4Ig in an attempt to block the CD28/B7 costimulatory pathway between antigen-presenting calls and T lymphocytes near the site of the graft. The SV40 T antigen is potent at inducing graft rejection. NIT.CTLA4Ig and control transfectants were transplanted subcutaneously into young NOD mice to determine whether CTLA4Ig secretion would abet the survival of the insulinoma graft. CTLA4Ig protein was secreted abundantly in vitro (3-5 microg/ml) and this phenotype was maintained in vivo. Tumor growth was monitored visibly, by palpation, by measuring blood glucose levels, and by death of the host from hypoglycemia caused by unregulated insulin production of the growing insulinoma. Cell growth was similar for NIT.CTLA4Ig7 and control transfectants in immunodeficient mice (nude, irradiated, or SCID mice), indicating that there was no intrinsic growth advantage of the NIT.CTLA4Ig cells. In immunocompetent NOD mice however, the survival/growth of the NIT.CTLA4Ig graft was significantly better than that of the controls. Histopathology was consistent with this finding. Donor-specific second-set grafts were acutely rejected, indicating that tolerance was not induced. CTLs were generated even when the graft secreted CTLA4Ig; there was no clear difference in in vitro immune responses generated by NIT.CTLA4Ig and control cells. We conclude that blockade of the B7 costimulation pathway by graft manipulation can contribute to transplantation success.

Prolonged allograft survival in anti-CD4 antibody transgenic mice: lack of residual helper T cells compared with other CD4-deficient mice.

BACKGROUND: Investigations of the role of CD4 T lymphocytes in allograft rejection and tolerance have relied on the use of mouse models with a deficiency in CD4 cells. However, in mice treated with depleting monoclonal antibody (mAb) and in MHC class II knockout (KO) mice, there are residual populations of CD4 cells. CD4 KO mice had increased CD4- CD8-TCRalphabeta+ helper T cells, and both strains of KO mice could reject skin allografts at the normal rate. In this study, transgenic mice with no peripheral CD4 cells were the recipients of skin and heart allografts. Results were compared with allograft survival in CD4 and MHC class II KO mice. METHODS: GK5 (C57BL/6 bml mice transgenic for a chimeric anti-CD4 antibody) had no peripheral CD4 cells. These mice, and CD4 and class II KO mice, received BALB/c or CBA skin or cardiac allografts. Some GK5 mice were treated with anti-CD8 mAb to investigate the role of CD8 cells in rejection. CD4 and CD8 cells were assessed by FACS and immunohistochemistry. RESULTS: BALB/c skin on GK5 mice had a mean survival time +/- SD of 24+/-6 days, compared with 9+/-2 days in wild-type mice. Anti-CD8 mAb prolonged this to 66+/-7 days. BALB/c skin survived 10+/-2 days on class II KO and 14+/-2 days on CD4 KO, both significantly less than the survival seen on GK5 recipients (P<0.001). BALB/c hearts survived >100 days in GK5 recipients and in wild-type recipients treated with anti-CD4 mAb at the time of grafting, in contrast to a mean survival time of 10+/-2 days in untreated wild-type mice. Immunohistochemistry revealed that long-term surviving heart allografts from the GK5 recipients had CD8 but no CD4 cellular infiltrate. These hearts showed evidence of transplant vasculopathy. CONCLUSIONS: The GK5 mice, with a complete absence of peripheral CD4 cells, provide the cleanest available model for investigating the role of CD4 lymphocytes in allograft rejection. Prolonged skin allograft survival in these mice compared with CD4 and MHC class II KO recipients was clearly the result of improved CD4 depletion. Nevertheless, skin allograft rejection, heart allograft infiltration, and vascular disease, mediated by CD8 cells, developed in the absence of peripheral CD4 T cells.

Predominant transgene expression in exocrine pancreas directed by the CMV promoter.

The enhancer/promoter of the human cytomegalovirus gene encoding the major immediate-early protein (CMVp) is reputed to be one of the strongest and most promiscuous regulatory elements for directing transcription of heterologous genes in vitro. However, transgene expression under the promoter in adult transgenic mice is often more restricted. We selected a CMVp segment from position -350 to +59 to control expression of transgenes for two secretory fusion proteins. Expression was analyzed by immunohistology staining and quantified by Northern blot, Western blot, and ELISA of secretions from explanted tissues. In all six lines of transgenic mice, the highest expression of transgenes at the mRNA and protein level was observed in the exocrine tissue of the pancreas, although the levels of expression varied among the lines. The results indicate not only that CMVp is not a universal promoter in vivo but indeed that it can be relatively specific for the exocrine pancreas, where expression of the gene it controlled was consistently very high.

Inhibitory effect of lipopolysaccharide on immune response after DNA immunization is route dependent.

The DNA prepared from E. coli contained high levels of lipopolysaccharide (LPS). When antigen-encoding DNA was injected into mice, toxicity and increased IgM responses were observed. A method for purifying high yields of DNA (up to 12 mg/L of broth culture) with very low levels of LPS (0.05 ng/mg) was developed. When this purified DNA was used for immunization studies, the toxicity and increased IgM responses were abrogated. Thus, LPS was added to DNA in order to examine its influence on the IgG and cytotoxic T lymphocyte (CTL) response after intramuscular (i.m.) or intradermal (i.d.) DNA immunization. The IgG response to DNA-encoded antigen was inhibited in a dose-dependent manner by the i.d., but not the i.m., route of immunization. Surprisingly, no effect on the CTL response was observed. Therefore, the ability to produce high yields of plasmid DNA with very low levels of endotoxin contamination is advantageous for DNA immunization studies, not only for toxicologic but also for immunologic considerations. Furthermore, these results provide further evidence that immune induction occurs via different mechanisms after i.m. and i.d. DNA immunization.

Transgenic anti-CD4 monoclonal antibody secretion by mouse segmental pancreas allografts promotes long term survival.

To compare the effectiveness of transgenic and systemic monoclonal antibody therapy for pancreas transplantation, vascularised segmental pancreas allografts from wild-type or transgenic pancreatic tissue that secreted monoclonal anti-CD4 were placed in CBA recipients in which diabetes had been induced chemically by streptozotocin (STZ, non-autoimmune diabetes). In untreated CBA recipients, wild-type BALB/c or C57BL/6 bml pancreas transplants were rejected in a mean survival time (MST) of 27 and 30 days, respectively. BALB/c and C57BL/6 graft survival improved when recipients were given a short course of T cell depleting monoclonal anti-CD4 antibody, (GK 1.5, 2 mg total on days -1, 0, 1, 2 with grafting on day 0) with MST +/- S.D. of 71 +/- 29 and 44 +/- 36 days, respectively. Thus, transient depletion of CD4 was effective in delaying pancreas allograft rejection in these strain combinations. The use of C57BL/6 bml mice transgenic for a rat anti-CD4 antibody (GK5 mice) as pancreas donors provided allografts that secreted sufficient anti-CD4 antibody to cause CD4 T cell depletion in the recipients (CD4 cells decreased from 30 to < 5% of small lymphocytes). This degree of depletion was not sustained and the CD4 recovery inversely correlated with graft survival. Mice with > 20% CD4 cells in the splenic lymphocyte population 4 weeks post-transplant rejected their grafts (3 of 10 mice). However, in 7 of 10 mice CD4 cells remained low (< 15%) and allografts survived for > 80 days. The GK5 allografts survived significantly longer than those from non-transgenic bml controls (MST 83 +/- 32 days, compared with 30 days, P < 0.0005). This survival time was similar to that of BALB/c allografts in CBA recipients treated with a high dose of anti-CD4 antibody. Thus, transgenic secretion of anti-CD4 antibody by the pancreas allograft was very effective in prolonging its survival.

Enhanced survival of grafts genetically endowed with the ability to block CD2 and B7.

In a model of transplantation rejection, we have tested whether a graft manipulated to secrete immunomodulators could protect itself from immune destruction. An insulinoma cell line having the NOD genotype but also expressing the neoantigen, SV40 T antigen, was transfected with CTLA4Ig or LFA3Ig to block signals in the co-stimulatory/adhesion pathways. This neoantigen is potent at inducing graft rejection. Secretion of CTLA4Ig and LFA3Ig by transfectants promoted survival of the insulinoma graft in young NOD mice. In immunodeficient mice, cell growth was similar for all transfectants. However, in immunocompetent NOD mice the survival/growth of test grafts was significantly better than that of the controls. Graft survival was enhanced additively, when the two test transfectants were cotransplanted. Endowing the graft the ability to secrete immunomodulators that block individual co-stimulatory/adhesion signals can contribute to transplantation success. Blockade of two signals (CD2 and CD28) in these pathways enhances this success.

Protection of xenografts by a combination of immunoisolation and a single dose of anti-CD4 antibody.

Immunoisolation is the separation of transplanted cells from cells of the immune system using a semipermeable membrane. Using one such immunoisolation capsule-the TheraCyte device-we have assessed the survival of encapsulated xenogeneic tissue in vivo as well as the contribution of CD4+ve T cells to encapsulated xenograft rejection. The foreign body reaction to the TheraCyte capsule in vivo was assessed by transplanting empty capsules into normal mice. These capsules elicit a foreign body response by the host animal. Encapsulated CHO, NIT-1, and PK-15 cells were placed in culture and in immunodeficient mice to investigate their growth characteristics in the TheraCyte device. These cell lines survive both in culture and in immunodeficient SCID mice. Xenogeneic PK cells were also transplanted into normal C57BL/6 mice. These cells do not survive in normal mice despite the absence of direct contact between infiltrating and encapsulated cells. In addition, the survival of encapsulated cells in mice treated with a single dose of anti-CD4 antibody was examined. This was assessed using two systems: 1) histological analysis of capsule sections; 2) a quantitative luciferase reporter system using PK cells transfected to express luciferase. In both cases, anti-CD4 antibody contributed to prolonged encapsulated xenogeneic cell survival. Encapsulated xenogeneic cells survive in immunodeficient mice but not normal mice. Treatment of normal mice with anti-CD4 antibody results in prolonged survival of xenogeneic cells that can be measured using a luciferase reporter system. These results highlight the contribution of CD4+ve T cells to encapsulated xenograft rejection.