How predictive are in vitro assays for cytokine release syndrome in vivo? A comparison of methods reveals worrying differences in sensitivity and frequency of response

Ex vivo and in vitro Monocyte Responses Do Not Reflect in vivo Immune Responses and Tolerance

Cytokine production by ex vivo (EV)-stimulated leukocytes is commonly used to gauge immune function and frequently proposed to guide immunomodulatory therapy. However, whether EV cytokine production capacity accurately reflects the in vivo (IV) immune status is largely unknown. We investigated relationships between EV monocyte cytokine responses and IV cytokine responses in a large cohort of healthy volunteers using a highly standardized IV model of short-lived LPS-induced systemic inflammation, which captures hallmarks of both hyperinflammation and immunological tolerance. Therefore, 110 healthy volunteers were intravenously challenged with 1 ng/kg LPS twice: on day 0 to determine the extent of the IV (hyper)inflammatory response and on day 7 to determine the degree of IV endotoxin tolerance. Baseline EV monocyte cytokine production capacity was assessed prior to LPS administration. Short-term and long-term EV tolerance was assessed in monocytes isolated 4 h and 7 days after LPS administration, respectively. No robust correlations were observed between baseline EV cytokine production capacity and IV cytokine responses following LPS administration. However, highly robust inverse correlations were observed between IV cytokine responses and EV cytokine responses of monocytes isolated 4 h after IV LPS administration. No correlations between IV and EV tolerance were found. In conclusion, attenuated EV cytokine production capacity reflects ongoing IV inflammation rather than immune suppression. Results of EV assays should be interpreted with caution at the risk of improper use of immuno-stimulatory drugs.

Immunomodulatory peptides-A promising source for novel functional food production and drug discovery

Immunomodulatory peptides are a complex class of bioactive peptides that encompasses substances with different mechanisms of action. Immunomodulatory peptides could also be used in vaccines as adjuvants which would be extremely desirable, especially in response to pandemics. Thus, immunomodulatory peptides in food of plant origin could be regarded both as valuable suplements of novel functional food preparation and/or as precursors or possible active ingredients for drugs design for treatment variety of conditions arising from impaired function of immune system. Given variety of mechanisms, different tests are required to assess effects of immunomodulatory peptides. Some of those effects show good correlation with in vivo results but others, less so. Certain plant peptides, such as defensins, show both immunomodulatory and antimicrobial effect, which makes them interesting candidates for preparation of functional food and feed, as well as templates for design of synthetic peptides.

Humanized mouse models of immunological diseases and precision medicine

With the increase in knowledge resulting from the sequencing of the human genome, the genetic basis for the underlying differences in individuals, their diseases, and how they respond to therapies is starting to be understood. This has formed the foundation for the era of precision medicine in many human diseases that is beginning to be implemented in the clinic, particularly in cancer. However, preclinical testing of therapeutic approaches based on individual biology will need to be validated in animal models prior to translation into patients. Although animal models, particularly murine models, have provided significant information on the basic biology underlying immune responses in various diseases and the response to therapy, murine and human immune systems differ markedly. These fundamental differences may be the underlying reason why many of the positive therapeutic responses observed in mice have not translated directly into the clinic. There is a critical need for preclinical animal models in which human immune responses can be investigated. For this, many investigators are using humanized mice, i.e., immunodeficient mice engrafted with functional human cells, tissues, and immune systems. We will briefly review the history of humanized mice, the remaining limitations, approaches to overcome them and how humanized mouse models are being used as a preclinical bridge in precision medicine for evaluation of human therapies prior to their implementation in the clinic.

Meta-Analysis for Correlating Structure of Bioactive Peptides in Foods of Animal Origin with Regard to Effect and Stability

Amino acid (AA) sequences of 807 bioactive peptides from foods of animal origin were examined in order to correlate peptide structure with activity (antihypertensive, antioxidative, immunomodulatory, antimicrobial, hypolipidemic, antithrombotic, and opioid) and stability in vivo. Food sources, such as milk, meat, eggs, and marine products, show different frequencies of bioactive peptides exhibiting specific effects. There is a correlation of peptide structure and effect, depending on type and position of AA. Opioid peptides contain a high percentage of aromatic AA residues, while antimicrobial peptides show an excess of positively charged AAs. AA residue position is significant, with those in the first and penultimate positions having the biggest effects on peptide activity. Peptides that have activity in vivo contain a high percentage (67%) of proline residues, but the positions of proline in the sequence depend on the length of the peptide. We also discuss the influence of processing on activity of these peptides, as well as methods for predicting release from the source protein and activity of peptides.

Different players generate positive responses in two in vitro cytokine assay formats with aqueous and immobilized TGN1412 analog

To detect potential risk of severe cytokine release syndrome, in vitro assay formats with human cells have been developed. The two major testing platforms are a combination of whole blood with aqueous-phase test articles (whole blood cytokine assay, WBCA) and peripheral blood mononuclear cells with solid-phase articles (PBMC assay). Significant induction of cytokines was seen in both assays after treatment with a widely used control agent, TGN1412 or its analog CD28SA, but the WBCA cytokine profile differed from what was expected from clinical experience. In the WBCA, potential risk of CD28SA was detected by elevation of IL-8 whereas IL-2, a key cytokine after stimulation of CD28, was not induced in approximately 40% of donor samples. Therefore, further mechanistic understanding of the different responses in the in vitro assay was needed. In this study of donor samples treated with CD28SA, we compared the induction of cytokines and identified the cytokine-producing cells in the two assays. IL-2 was markedly elevated in all the donors in the PBMC assay but only in 1 of 3 donors in the WBCA. IL-8, the most sensitive biomarker in the WBCA, was produced by monocytes and granulocytes. T cells, the most relevant player in the PBMC assay with CD28SA, did not contribute to the positive response seen in two donors in the WBCA, which suggests that different players caused the positive cytokine responses to CD28SA in the two assays.

Is an in vitro whole blood cytokine assay useful to detect the potential risk of severe infusion reaction of monoclonal antibody pharmaceuticals?

After the life-threatening cytokine release syndrome (CRS) occurred in the clinical study of the anti-CD28 monoclonal antibody (mAb) TGN1412, in vitro cytokine release assays using human blood cells have been proposed for non-clinical evaluation of the potential risk of CRS. Two basic assay formats are frequently used: human peripheral blood mononuclear cells (PBMC) with immobilized mAbs, and whole blood with aqueous mAbs. However, the suitability of the whole blood cytokine assay (WBCA) has been questioned, because an unrealistically large sample size would be required to detect the potential risk of CRS induced by TGN1412, which has low sensitivity. We performed a WBCA using peripheral blood obtained from 68 healthy volunteers to compare two high risk mAbs, the TGN1412 analogue anti-CD28 superagonistic mAb (CD28SA) and the FcγR-mediated alemutuzumab, with a low risk mAb, panitumumab. Based on the cytokine measurements in this study, the sample size required to detect a statistically significant increase in cytokines with 90% power and 5% significance was determined to be n = 9 for CD28SA and n = 5 for alemtuzumab. The most sensitive marker was IL-8. The results suggest that WBCA is a practical test design that can warn of the potential risk of FcγR-mediated alemtuzumab and T-cell activating CD28SA but, because there was apparently a lower response to CD28SA, it cannot be used as a risk-ranking tool. WBCA is suggested to be a helpful tool for identifying potential FcγR-mediated hazards, but further mechanistic understanding of the response to CD28SA is necessary before applying it to T cell-stimulating mAbs.

Cytokine release: A workshop proceedings on the state-of-the-science, current challenges and future directions

In October 2013, the International Life Sciences Institute - Health and Environmental Sciences Institute Immunotoxicology Technical Committee (ILSI-HESI ITC) held a one-day workshop entitled, "Workshop on Cytokine Release: State-of-the-Science, Current Challenges and Future Directions". The workshop brought together scientists from pharmaceutical, academic, health authority, and contract research organizations to discuss novel approaches and current challenges for the use of in vitro cytokine release assays (CRAs) for the identification of cytokine release syndrome (CRS) potential of novel monoclonal antibody (mAb) therapeutics. Topics presented encompassed a regulatory perspective on cytokine release and assessment, case studies regarding the translatability of preclinical cytokine data to the clinic, and the latest state of the science of CRAs, including comparisons between mAb therapeutics within one platform and across several assay platforms, a novel physiological assay platform, and assay optimization approaches such as determination of FcR expression profiles and use of statistical tests. The data and approaches presented confirmed that multiple CRA platforms are in use for identification of CRS potential and that the choice of a particular CRA platform is highly dependent on the availability of resources for individual laboratories (e.g. positive and negative controls, number of human blood donors), the assay through-put required, and the mechanism-of-action of the therapeutic candidate to be tested. Workshop participants agreed that more data on the predictive performance of CRA platforms is needed, and current efforts to compare in vitro assay results with clinical cytokine assessments were discussed. In summary, many laboratories continue to focus research efforts on the improvement of the translatability of current CRA platforms as well explore novel approaches which may lead to more accurate, and potentially patient-specific, CRS prediction in the future.

Cytokine release assays for the prediction of therapeutic mAb safety in first-in man trials--Whole blood cytokine release assays are poorly predictive for TGN1412 cytokine storm

The therapeutic monoclonal antibody (mAb) TGN1412 (anti-CD28 superagonist) caused near-fatal cytokine release syndrome (CRS) in all six volunteers during a phase-I clinical trial. Several cytokine release assays (CRAs) with reported predictivity for TGN1412-induced CRS have since been developed for the preclinical safety testing of new therapeutic mAbs. The whole blood (WB) CRA is the most widely used, but its sensitivity for TGN1412-like cytokine release was recently criticized. In a comparative study, using group size required for 90% power with 5% significance as a measure of sensitivity, we found that WB and 10% (v/v) WB CRAs were the least sensitive for TGN1412 as these required the largest group sizes (n = 52 and 79, respectively). In contrast, the peripheral blood mononuclear cell (PBMC) solid phase (SP) CRA was the most sensitive for TGN1412 as it required the smallest group size (n = 4). Similarly, the PBMC SP CRA was more sensitive than the WB CRA for muromonab-CD3 (anti-CD3) which stimulates TGN1412-like cytokine release (n = 4 and 4519, respectively). Conversely, the WB CRA was far more sensitive than the PBMC SP CRA for alemtuzumab (anti-CD52) which stimulates FcγRI-mediated cytokine release (n = 8 and 180, respectively). Investigation of potential factors contributing to the different sensitivities revealed that removal of red blood cells (RBCs) from WB permitted PBMC-like TGN1412 responses in a SP CRA, which in turn could be inhibited by the addition of the RBC membrane protein glycophorin A (GYPA); this observation likely underlies, at least in part, the poor sensitivity of WB CRA for TGN1412. The use of PBMC SP CRA for the detection of TGN1412-like cytokine release is recommended in conjunction with adequately powered group sizes for dependable preclinical safety testing of new therapeutic mAbs.

An autologous endothelial cell:peripheral blood mononuclear cell assay that detects cytokine storm responses to biologics

There is an urgent unmet need for human tissue bioassays to predict cytokine storm responses to biologics. Current bioassays that detect cytokine storm responses in vitro rely on endothelial cells, usually from umbilical veins or cell lines, cocultured with freshly isolated peripheral blood mononuclear cells (PBMCs) from healthy adult volunteers. These assays therefore comprise cells from 2 separate donors and carry the disadvantage of mismatched tissues and lack the advantage of personalized medicine. Current assays also do not fully delineate mild (such as Campath) and severe (such as TGN1412) cytokine storm-inducing drugs. Here, we report a novel bioassay where endothelial cells grown from stem cells in the peripheral blood (blood outgrowth endothelial cells) and PBMCs from the same donor can be used to create an autologous coculture bioassay that responds by releasing a plethora of cytokines to authentic TGN1412 but only modestly to Campath and not to control antibodies such as Herceptin, Avastin, and Arzerra. This assay performed better than the traditional mixed donor assay in terms of cytokine release to TGN1412 and, thus, we suggest provides significant advancement and a definitive system by which biologics can be tested and paves the way for personalized medicine.

Cytokine release assays: current practices and future directions

As a result of the CD28 superagonist biotherapeutic monoclonal antibody (TGN 1412) "cytokine storm" incident, cytokine release assays (CRA) have become hazard identification and prospective risk assessment tools for screening novel biotherapeutics directed against targets having a potential risk for eliciting adverse pro-inflammatory clinical infusion reactions. Different laboratories may have different strategies, assay formats, and approaches to the reporting, interpretation, and use of data for either decision making or risk assessment. Additionally, many independent contract research organizations (CROs), academic and government laboratories are involved in some aspect of CRA work. As a result, while some pharmaceutical companies are providing CRA data as part of the regulatory submissions when necessary, technical and regulatory practices are still evolving to provide data predictive of cytokine release in humans and that are relevant to safety. This manuscript provides an overview of different approaches employed by the pharmaceutical industry and CROs, for the use and application of CRA based upon a survey and post survey follow up conducted by ILSI-Health and Environmental Sciences Institute (HESI) Immunotoxicology Committee CRA Working Group. Also discussed is ongoing research in the academic sector, the regulatory environment, current limitations of the assays, and future directions and recommendations for cytokine release assays.