Lars Adamson

Myeloid-derived suppressor cells impair the quality of dendritic cell vaccines

Myeloid-derived suppressor cells (MDSC) are important regulators of the immune system and key players in tumor-induced suppression of T-cell responses. CD14+HLA-DR−/low MDSC have been detected in a great number of malignancies, including melanoma. MDSC are known to be impaired in their ability to differentiate along the myeloid lineage, e.g., into dendritic cells (DC). This is a concern for utilization of monocyte-derived DC for vaccination of patients with melanoma or other cancers exhibiting accumulation of CD14+ MDSC. When producing DC according to standard operating procedures of two currently ongoing clinical trials, we found that MDSC co-purified with monocytes isolated by elutriation. MDSC frequencies did not affect yield or viability of the produced DC, but induced a dose-dependent decrease in DC maturation, ability to take up antigen, migrate and induce T-cell IFNγ production. Changes in DC characteristics were most notable when ‘pathological’ frequencies of >50% CD14+HLA-DR− cells were present in the starting culture. The impaired DC quality could not be explained by altered cytokine production or increased oxidative stress in the cultures. Tracking of HLA-DR− cells throughout the culture period revealed that the observed changes were partially due to the impaired maturation and functionality of the originally HLA-DR− population, but also to their negative effects on HLA-DR+ cells. In conclusion, MDSC could be induced to differentiate into DC but, due to the impairment of overall DC vaccine quality when >50% HLA-DR− cells were present in the starting culture, their removal could be advisable.

FXIII induced gelation of human fibrinogen — An alternative thiol enhanced, thrombin independent pathway

Factor XIII induced gelation of human fibrinogen in the presence of calcium ions. At the end of this reaction between 95 and 100% of the fibrinogen was incorporated into the gel matrix. The gelation was dramatically enhanced by DTT. Cysteine and beta-mercaptoethanol also enhanced the reaction, but less efficiently. Thrombin activated factor XIII led to shortened gelation time and increased the rate of gelation. The reaction was inhibited by p-chloromercuribenzoate and iodoacetamide. Neither fibrinopeptide A, nor fibrinopeptide B were released during gelation, while quantitative release of FPA by thrombin was demonstrated from preformed gel matrices. SDS-PAGE showed the presence of gamma-dimers and alpha-polymers in the gel matrix. In the clot supernatants gamma-dimers were observed already before the gel point. We also observed that the clotting of fibrinogen by thrombin was perturbed by DTT. Preincubation of fibrinogen with calcium ions prevented this effect of DTT.

Factor XIII catalyzed formation of fibrinogen-fibronectin oligomers--a thiol enhanced process.

Fibrinogen and plasma fibronectin were shown to interact in the presence of factor XIIIa. The reaction was enhanced by dithiothreitol and was accompanied by an increase in the turbidity of the solution and the formation of particulate matter and gel structures. At a constant concentration of fibrinogen the turbidity increase was dependent on the fibronectin concentration and at a constant concentration of fibronectin, on the fibrinogen concentration. Kinetic experiments showed that an initial step in the reaction between fibrinogen and fibronectin was the formation of a transient intermediate containing 1 mole of fibrinogen and 1 mole of fibronectin. Transient intermediates of larger molecular weight and containing both fibrinogen and fibronectin were also formed. These heterooligomers eventually reached huge molecular sizes and at early times formed particulate matter that sedimented on centrifugation. The predominant molecular species formed in an equimolar mixture of fibrinogen and fibronectin were heteropolymers. Small amounts of homopolymers composed of fibrinogen and possibly also homopolymers of fibronectin were detected. The results are discussed in terms of reaction mechanism and potential importance of this novel oligomerization pathway in haemostasis, thrombosis and tissue repair.

A phase I clinical trial combining dendritic cell vaccination with adoptive T cell transfer in patients with stage IV melanoma

Adoptive transfer of in vitro-expanded tumor-infiltrating lymphocytes (TIL) has shown great clinical benefit in patients with malignant melanoma. TIL therapy itself has little side effects, but conditioning chemo- or radiotherapy and postinfusion interleukin 2 (IL-2) injections are associated with severe adverse advents. We reasoned that combining TIL infusion with dendritic cell (DC) vaccination could circumvent the need for conditioning and IL-2 support and thus represent a milder treatment approach. Eight patients with stage IV melanoma were enrolled in the MAT01 study, consisting of vaccination with autologous tumor-lysate-loaded DC, followed by TIL infusion. Six of eight patients were treated according to protocol, while one patient received only TIL and one only DC. Treatments were well tolerated with a single grade 3 adverse event. The small study size precludes analysis of clinical responses, though interestingly one patient showed a complete remission and two had stable disease. Analysis of the infusion products revealed that mature DC were generated in all cases. TIL after expansion were CD3+ T cells, dominated by effector memory CD8+ cytotoxic T cells. Analysis of the T cell receptor repertoire revealed presence of highly dominant clones in most infusion products, and many of these could be detected in the circulation for weeks after T cell transfer. Here, we report the first combination of DC vaccination and TIL infusion in malignant melanoma. This combined treatment was safe and feasible, though after evaluating both clinical and immunological parameters, we expect that administration of lymphodepleting chemotherapy and IL-2 will likely increase treatment efficacy.

Development of a technology platform for large-scale clinical grade production of DC.

BACKGROUND Clinical studies require protocols where a sufficient number of well-characterized highly immunogenic DC are produced according to good manufacturing practice (GMP) guidelines. METHODS In the present study, using leukapheresis products from 10 cancer patients, we validated an elutriation technology for large-scale clinical grade production of monocyte-derived DC. RESULTS The elutriation method gave a very high purity (mean+/-SD) (86+/-5.3%) and recovery (66+/-10.4%) of monocytes. Specifically for the two monocyte-rich fractions (3 and 4,) the recovery was 42+/-13% of viable cells that could be further differentiated into immature DC in hydrophobic culture bags using GM-CSF and IL-4. The immature DC exhibited<1% CD83+ expression and >98% phagocytic activity. Maturation with TNF-alpha or poly I:C resulted in DC with expression of CD80+, CD86+ and HLA-DR+ (>99%) and CD83+ (80+/-11.9%), as well as producing IL-12p70 and lacking phagocytic activity (<5%). This cell product can be cryopreserved with cell viability >85% and cell recovery >80% after thawing. DISCUSSION The elutriation procedure, when optimized and if the monocyte content of the starting material exceeds 5%, does not require further selection or depletion using affinity approaches.

Generation of DC-based vaccine for therapy of B-CLL patients. Comparison of two methods for enriching monocytic precursors

BACKGROUND The generation of Ag-loaded DC under good manufacturing practice (GMP) conditions is logistically challenging and further compounded when the starting precursors need to be purified from B-CLL patients who have overwhelming numbers of circulating B-CLL cells and decreased numbers of monocytes. METHODS We have previously demonstrated that DC with endocytosed B-CLL apoptotic bodies are powerful stimulators of anti-leukemic T cells. In this study we compared counterflow elutriation and immunomagnetic separation for enriching monocyte precursors, and evaluated the feasibility of generating DC from B-CLL patients and the effects of cryopreservation. RESULTS Monocyte yield from a single leukapheresis product of a B-CLL patient varied from 1 x 108 to 10 x 108 total cells, from which 40-200 x 106 mature DC could be produced. Adequate numbers of monocytes could not be enriched from one patient with 0.2% monocytes in the leukapheresis product, and the target of 50 x 106 DC was barely achieved in another patient with 0.9% monocytes in the pheresed cells. These results suggested that successful production of DC is dependent on a minimum frequency of 1% CD14(+) monocytes in the leukapheresis product. Cryopreservation of tumor cell-loaded DC yielded a recovery rate of 86+/-4.4% upon thawing, with a total viability of 90+/-2.8%. Most importantly, cryopreserved Ag-loaded DC retained their morphology, phenotype and function. DISCUSSION The results demonstrate that adequate numbers of functional DC required for clinical therapy can be generated from patients who have >1% of CD14(+) monocytes in the leukapheresis product. Moreover, Ag-loaded DC can be cryopreserved and recovered without significant change in phenotype or function.

Dendritic cell regulation of NK-cell responses involves lymphotoxin-α, IL-12, and TGF-β

Dendritic cell (DC) vaccines induce T‐cell responses in cancer patients. However, there is a paucity of data regarding the role of DC vaccines in shaping natural killer (NK) cell responses. Here, we observe that NK cells are less activated following DC vaccination. In vitro, DC‐mediated inhibition of NK cells did not require cell‐to‐cell contact, but required increased Signal transducer and activator of transcription 3 (STAT3) phosphorylation (pSTAT3) in DCs. When phosphorylation of STAT3 was inhibited in DCs, we found that DCs did not suppress NK cells, and observed an increase in the production of lymphotoxin‐alpha (LTα) and interleukin‐12 (IL‐12) as well as reduced release of transforming growth factor beta (TGF‐β). The addition of recombinant LTα or IL‐12 to the DC‐NK‐cell cocultures restored NK‐cell activity, and neutralization of TGF‐β resulted in elevated production of LTα and IL‐12 from DCs. Compared with LPS, DCs matured with a cocktail of R848, poly I:C, and IFN‐γ showed reduced levels of pSTAT3 and higher levels of LTα and IL‐12 and did not inhibit NK‐cell activity. These results show that LTα, IL‐12, and TGF‐β are involved in the cross‐talk between NK cells and DCs. Our findings have important implications for the development of DC‐based vaccination strategies to potentiate NK‐cell responses in patients with cancer.

Fibrinogen Aarhus and factor XIII induced polymerization and gel formation

Fibrinogen Aarhus is an abnormal fibrinogen for which the clotting time with thrombin is greatly prolonged both in plasma and in the isolated fibrinogen. The whole blood clotting time is only slightly prolonged. The patient with this fibrinogen has no bleeding tendency. In this report we have investigated fibrinogen Aarhus in two alternative, thrombin independent polymerization and gelation pathways. These pathways are the factor XIII dependent oligomerization and gelation of fibrinogen, and heteropolymer (fibrinogen‐fibronectin) formation which also is catalysed by factor XIII. Both of these reactions are qualitatively the same in fibrinogen Aarhus as in normal fibrinogen, but the rate of oligomerization is somewhat slower in fibrinogen Aarhus. This may depend on impaired association between factor XIII and fibrinogen Aarhus.

Immunotherapy in atypical teratoid-rhabdoid tumors: Data from a survey of the HGG-Immuno Group

BACKGROUND AIMS Atypical rhabdoid/teratoid tumors (AT/RT) are the most common brain tumors in infants and associated with a dismal prognosis. Although intensification of first-line therapy has resulted in improvement of overall survival, novel treatment strategies are needed. Because immunotherapy has resulted in remarkable results in several adult tumor entities, incorporation of immunotherapy into AT/RT treatment offers a novel alternative. METHODS We retrospectively analyzed data from 7 AT/RT patients from five countries treated within the HGG-Immuno Consortium. Two patients were ≤1 year and 4 patients were ≤2 years of age at diagnosis. All received immunotherapy with autologous, tumor-lysate-loaded dendritic cells (DCs) on a compassionate use basis using a schedule of three to four weekly DC vaccinations with up to 2 × 10(7) DCs per vaccine, followed by three lysate boosts each 1 month apart. RESULTS Monocyte collections (median age at apheresis 31.5, range 20-143 months) and vaccinations were uneventful without any severe adverse event related to the vaccine, demonstrating feasibility and safety in this very young age group. Two children received immunotherapy during their primary and the remaining five during second- or third-line therapy. Three of seven patients survived long term with a follow-up of 143, 138 and 46 months, with at least two of them harboring somatic mutations. One long-term survivor was vaccinated during primary treatment and the other two after first or second relapse/progression. Two analyzed patients showed positive CD8(+) T-cell responses after vaccination. DISCUSSION Our data demonstrate that anti-tumor immunotherapy with autologous DCs is feasible and safe in young children with ATRTs and that this approach warrants further investigation in controlled clinical trials.

Enhanced stimulation of human tumor-specific T cells by dendritic cells matured in the presence of interferon-γ and multiple toll-like receptor agonists

Dendritic cell (DC) vaccines have been demonstrated to elicit immunological responses in numerous cancer immunotherapy trials. However, long-lasting clinical effects are infrequent. We therefore sought to establish a protocol to generate DC with greater immunostimulatory capacity. Immature DC were generated from healthy donor monocytes by culturing in the presence of IL-4 and GM-CSF and were further differentiated into mature DC by the addition of cocktails containing different cytokines and toll-like receptor (TLR) agonists. Overall, addition of IFNγ and the TLR7/8 agonist R848 during maturation was essential for the production of high levels of IL-12p70 which was further augmented by adding the TLR3 agonist poly I:C. In addition, the DC matured with IFNγ, R848, and poly I:C also induced upregulation of several other pro-inflammatory and Th1-skewing cytokines/chemokines, co-stimulatory receptors, and the chemokine receptor CCR7. For most cytokines and chemokines the production was even further potentiated by addition of the TLR4 agonist LPS. Concurrently, upregulation of the anti-inflammatory cytokine IL-10 was modest. Most importantly, DC matured with IFNγ, R848, and poly I:C had the ability to activate IFNγ production in allogeneic T cells and this was further enhanced by adding LPS to the cocktail. Furthermore, epitope-specific stimulation of TCR-transduced T cells by peptide- or whole tumor lysate-loaded DC was efficiently stimulated only by DC matured in the full maturation cocktail containing IFNγ and the three TLR ligands R848, poly I:C, and LPS. We suggest that this cocktail is used for future clinical trials of anti-cancer DC vaccines.