Facebook link Linked In link You Tube link
Translational Immunology Update©
A quarterly publication of FOCiS
June 17, 2013

Editor: Andrew H. Lichtman, MD, PhD, Brigham & Women's Hospital
Editorial Board: Abul K. Abbas, MD, University of California, San Francisco | Carla J. Greenbaum, MD, Benaroya Research Institute 

Highlights in Recent Literature | Clinical Immunology Highlights | Immunphenotyping | Basic Immunology | Clinical TrialsPDF Version |Previous Issues

Highlights from Recent Literature

Th1-like Innate Lymphoid Cells Accumulate in the Intestinal Lesions of Patients with Crohn’s Disease

A review of Jochem H. Bernink, et al. Human Type 1 Innate Lymphoid Cells Accumulate in Inflamed Mucosal Tissues. Nature Immunology 14, 221–229 (2013). PMID: 23334791

Innate lymphoid cells (ILC) are recently described lymphoid cells that lack rearranged T cell receptors and are thought to participate in innate immune reactions and tissue remodeling. Subsets of ILC have been described that have cytokine production profiles similar to those of helper T cell subsets. In this article, the authors describe a type of ILC that resembles Th1 cells and show they accumulate in the intestines of patients with Crohn’s disease.

Studying cells from human tonsils, the authors identified a novel population of T helper type 1 cell (Th1 cell)-like ILC distinct from NK cells that produced IFNγ in response to IL-12. The authors designated these cells ILC1 cells.

Crohn’s disease is a type I mediated inflammatory disease with increased production of IFNγ, IL-12 and IL-18.

The authors compared ILC populations from healthy intestine and from the inflamed and non-inflamed intestines of patients with Crohn’s disease. They observed increased numbers of ILC1 cells and decreased numbers of Th17-like ILC3 cells in the intestinal lesions of Crohn’s disease vs. healthy controls.

Irradiated immunodeficient mice reconstituted with human hematopoietic stem cells had an ILC pool similar to non-inflamed human gut. Treatment of these mice with dextran sodium sulfate (DSS) led to gut inflammation that was not dependent on human immune cells but nonetheless was associated with increased numbers of resident ILC1 cells. These studies suggested that ILC1 cells accumulated as a result of inflammation.The authors observed that ILC populations were plastic, and that both immature ILCs and ILC3 cells could differentiate into ILC1 cells in vitro in the presence of IL-2 and IL-12.

The authors observed that distinct Treg subsets are present in human colon cancer with pro-inflammatory and anti-inflammatory phenotypes and functions. RORγt -expressing Tregs were pro-inflammatory, supported the development of pre-cancerous polyps, and reduced anti-polyp immune responses. IL-17 produced by these Tregs increased the formation of precancerous polyps but paradoxically impaired progression to invasive cancer. In summary, the authors find Tregs within cancer can be surprisingly diverse and that some of Tregs can directly promote carcinogenesis as well as suppressing anticancer responses.

The authors describe ILC that produce IFNγ and are distinct from previously identified ILC3 and ILC2 cells. These ILC1 cells accumulated in the mucosal lesions of human Crohn’s disease. Mouse studies showed these cells accumulated because of inflammation but did not necessarily cause inflammation. The authors observed functional plasticity among ILC in that both immature ILC and mature ILC3 cells could differentiated into ILC1 cells in vitro. In summary, the authors describe a novel population of Th1-like ILC that accumulate in the inflamed lesions of human Crohn’s disease. The role of these cells in healthy immunity and their contribution to pathology remain to be determined.

Reviewed by Rachael A. Clark, MD, PhD, Department of Dermatology, Brigham and Women's Hospital. 

T Follicular Helper-like Cells Appear in the Circulation and Lead to a Burst of Anti-viral Antibody Production by Memory T Cells Following Influenza Vaccination

A review of Salah-Eddine Bentebibel, et al. Induction of ICOS+CXCR3+CXCR5+ TH Cells Correlates with Antibody Responses to Influenza Vaccination. Science Translational Medicine 5,176ra32 (2013).  
PMID: 23486778

The immune events that lead to protective antibody production after influenza vaccination remain poorly characterized in humans. The authors studied circulating T cell subsets following influenza vaccination identified a particular type of T cell that correlates with the production of protective anti-influenza antibody by memory T cells.

The authors observed an increased number of CD4+ T cells expressing ICOS after vaccination with split influenza vaccines. These cells resembled T follicular helper (TFH) cells but lacked the Bcl-6 expression observed in true TFH found in secondary lymphoid organs. These cells were induced in patients of different ages and in response to different split influenza vaccine formulations.

Emergence of these cells correlated with increased global protective anti-influenza antibody titers against previously recognized strains but not against novel strains, suggesting these cells contributed to boosting pre-existing antibody titers but did not induce primary antibody responses.

A subset of these circulating ICOS+ TFH-like T cells was specific for influenza antigens. These cells produced IFNγ, IL-2, IL-10 and IL-21, efficiently helped memory B cells differentiate into plasma cells, and specifically assisted influenza specific B cells to differentiate into plasma cells making influenza specific antibodies.

However, these cells did not provide help to naïve B cells and therefore did not support the formation of primary anti-influenza immune responses. The authors demonstrate that the appearance of circulating ICOS+ TFH-like CD4+ T cells is correlated with an increase in pre-existing anti-influenza antibodies but not with the generation of primary antibody responses following vaccination with split influenza vaccines. These T cells induced influenza-specific B cells to differentiate into plasma cells that produced protective anti-influenza antibodies but they were unable to provide help to naïve T cells. In addition to clarifying the role of ICOS+ TFH-like CD4+ T cells, these results suggest that split influenza vaccines primarily boost production of existing antibodies and fail to induce strong primary responses. Alternative vaccine strategies or inclusion of adjuvants may be necessary to induce protective primary immune responses against different, newly encountered strains of influenza.

Reviewed by Rachael A. Clark, MD, PhD, Department of Dermatology, Brigham and Women's Hospital.

Open Access Systems Immunology and Vaccinology

A review of Obermoser, G. et al. Systems Scale Interactive Exploration Reveals Quantitative and Differences in Response to Influenza and Pneumococcal Vaccines.  Immunity (2013) 38: 831-844. 
PMID: 23601689

The application of high throughput profiling strategies to disease pathogenesis and progression has recently been extended to deepen our understanding of vaccines. We have been able to generate vaccines against many of the infectious diseases that stalked human populations (and still do in unvaccinated populations). However, we have been most successful in crafting vaccines to prevent invariant pathogens (like measles, rubella, etc) it has remained a monumental challenge to develop effective vaccines against HIV, TB, Hepatitis C virus and malaria, among others. Influenza vaccination, while effective, requires costly, time consuming yearly re-design of the immunization. There are many contributing factors to the challenges faced by vaccine researchers, from the ability of some pathogens to rapidly mutate their fundamental building blocks to escape detection and our defenses, to the humoral nature of the immunity elicited by our standard vaccine designs and the ability of many viruses to stealthily combat our immune system, yielding poor natural immunity to infection (let alone vaccination). In this context, many groups are trying to better understand the mechanism by which our best and our most commonly used vaccines function, from the mechanism of the limited set of clinically approved adjuvants to the implications of sub-unit component selection or vaccine site selection. In order to accomplish this goal, the application of an unbiased approach to gathering relevant data about the cell circuits that underlie protection (or lack thereof) in response to a vaccine, yield massive datasets, which can be impenetrable to researchers beyond the primary team. This work assesses the immune response in a small group of healthy patients to two commonly used non-live vaccines, those for influenza and pneumococcus. In addition, they crafted an intriguing user friendly interface to allow full immersion and consideration of the wealth of data they have collected.

  • They recruited 6 healthy volunteers who randomly received influenza vaccine, 23-valent pneumococcal vaccine or saline injections. These participants had a baseline blood draw pre-immunization (day -7) and then had blood drawn for transcriptome analysis at 0, 1, 3, 7, 10, 14, 21 and 28 days. In addition, this was correlated with serum antibody titers in response to the relevant vaccine at days 0 and 28.
  • Throughout the study, the immunological findings are intriguing, with creative and clarifying use of novel static and interactive figures. The ability to link gene expression results to flow cytometry and cytokine results is a powerful model going forward, and raises the bar for dissemination of complex datasets in a fashion that aims to spur detailed analysis of the datasets in labs across the world.
  • They performed a modular analysis, with modules defined by previous investigations. This allowed a granular analysis (in addition to a fine analysis) of the major coordinated gene expression circuits that fundamentally underlie these responses. By comparing two vaccines, they were able to show both similarities and differences in the responses elicited.
    • While a number of gene expression changes were shared, each vaccine had modules that were uniquely stimulated or repressed by one of the two studied vaccines. They group the gene expression changes linked to PCV23 as inflammatory, while the modules associated with influenza were related to innate immunity and interferon signaling.
  • Detailed analysis of early transcriptional events is made possible by a parallel study of 6 independent patients who were assayed frequently after immunization via finger stick blood draws (at -168, 0, 1.5, 3, 6, 9, 12, 15, 24, 36 and 48h) which were assessed for transcriptional changes in response to vaccination and then compared to the transcriptional changes in cells exposed to a variety of pathogens (as well as the saline control patients).
  • They focus on the interferon module and assess transcription in sorted cell subsets (neutrophil, CD4 and CD8 T cells and monocytes), implicating neutrophils and monocytes as the predominant source of interferon early in the vaccine response. Finally, they identified plasmablast responses by gene expression and flow cytometry at day 7, as seen in other studies.

This paper deeply investigates the response of the human immune system to two non-live vaccinations (influenza and pneumococcus), generating a deep and fascinating dataset. However, this study embraces the spirit of collaboration and takes on (rather than simply acknowledging) the practical challenges of this type of work, by creating and sharing an interface that allows both experts and novices to engage directly with the various high throughput profiling data sets in a coordinated and clear fashion. While the data set is drawn from two small samples of healthy adults, the assessment of both early and medium terms responses in a detailed fashion, with a saline control for comparison for both, provide an optimal assessment of those patients. This is a fascinating paper, adding to the vaccinology, systems immunology and computational biology fields simultaneously.

Reviewed by Sarah Henrickson, MD, PhD, Boston Children’s Hospital and Harvard Medical School.

Evaluation B Cell Tolerance Checkpoints of Multiple Sclerosis

A review of Kinnunen T., et al. Specific Peripheral B Cell Tolerance Defects in Patients with Multiple Sclerosis. Journal of Clinical  Investigation. 2013. 123(6):2737-41.  PMID: 23676463

Multiple sclerosis (MS) is an autoimmune disease that preferentially attacks myelin resulting in immune mediated demyelination and subsequent neurologic sequelae. Historically, MS had long thought to be a T cell mediated disease. It was thus surprising when MS patients improved clinically after B cell depletion with the anti-CD20 antibody Rituximab. While this highlights an important role for B cells in MS, how they contribute to disease pathogenesis has remained incompletely understood. Meffre and co-workers address this knowledge gap by comparing the B cell repertoire of treatment-naïve MS patients and healthy controls. Each day humans produce an excess of ~10 billion lymphocytes. Because generation of B and T cell receptors occurs through random recombination, the majority of these lymphocytes are autoreactive. In humans, autoreactive B cells are culled from the repertoire at two key checkpoints, one in the bone marrow (central tolerance) and one in the periphery (peripheral tolerance) prior to emergence of newly emigrant B cells into the mature naïve B cell compartment. To evaluate the reactivity of the B cell repertoire, the authors clone and produce recombinant antibodies from individual newly emigrant/transitional and mature naïve B cells and evaluate each B cell antibody’s reactivity to self antigens. Prior work by the Meffre laboratory utilizing similar techniques has demonstrated that both checkpoints are defective in patients with rheumatoid arthritis (RA) or type 1 diabetes (T1D). Using a similar approach, they analyze the autoreactivity of B cells at each of these checkpoints in MS patients. There are several interesting findings from this work.

  • The proportion of newly emigrant/transitional B cells that were polyreactive with specificities to self antigens were comparable between most MS patients and healthy controls. This indicates that the central B cell tolerance checkpoint is intact in the majority of MS patients.
  • This is in stark contrast to RA and TID patients that show a high frequency of polyreactive new emigrant/transitional B cells due to a defective central B cell tolerance checkpoint.
  • Similar to RA and TID patients, evaluation of mature naïve B cells revealed an increased frequency of polyreactive B cells in MS patients relative to healthy controls, indicating that the peripheral B cell checkpoint is perturbed in MS. Importantly, these mature naïve B cells from MS patients were reactive to white matter extracts from the brain.
  • Loss of peripheral tolerance is accompanied by increased activation and homeostatic proliferation of mature naïve B cells but not newly emigrant/transitional B cells.
  • The authors note that the loss of peripheral but not central B cell tolerance in MS mirrored the patterns seen in IPEX patients. IPEX patients suffer from impaired Treg function due to mutations in the FOXp3 gene. While functional studies were unfortunately not performed, the authors do demonstrate an abnormal accumulation of memory Tregs relative to healthy controls in MS patients similar to that seen in IPEX patients. Based on this observation, they postulate that loss of peripheral tolerance in MS could be due to defective B-T cell interactions.

The observation that defective peripheral B cell tolerance is common to MS, TID, and RA while the central tolerance defect is absent in MS is intriguing. The authors speculate that this could potentially explain why B cell depletion strategies tend to result in a more sustained remission in MS relative to other autoimmune disorders. They reason that once autoreactive B cells are eliminated in the periphery due to therapy, newly generated B cells exiting the bone marrow should be properly counterselected in MS patients but not patients whose autoimmune disease is characterized by defective central tolerance.

Reviewed by Michelle L. Hermiston, MD, PhD, University of California, San Francisco.

Armed and Waiting: Role of CDαα+T Cells in Regulation of Human Herpes Virus Infection

A review of Zhu J., et al. Immune Surveillance by CD8αα+ Skin-resident T Cells in Human Herpes Virus Infection. Nature. 2013 May 23;497(7450):494-7. PMID: 23657257

Clinical studies have shown that most HSV-2 reactivations are subclinical and of short duration. The mechanisms that regulate rapid asymptomatic HSV-2 containment upon reactivation had been incompletely elucidated. Prior studies had shown that CD8+ T cells persist at the dermal-epidermal junction (DEJ) at the sites of neural release of reactivating virus in genital skin and mucosa. However, the mechanistic basis by which these cells control viral reactivation were poorly understood due to the inherent challenges of studying primary immune cells during infection in humans. Zhu, Peng, Corey and colleagues use cell-type specific laser capture microdissection, transcriptional profiling, and T-cell antigen receptor chain (TCR) sequencing on sequential genital skin biopsies from 10 patients to tackle this challenge. By comparing individual CD8+ T cells from the DEJ of infected areas (DEJ CD8) to control CD8+ T cells located near blood vessels (BV CD8) or at the DEJ of the contralateral HSV-2 unaffected genital tissue (control CD8), they provide important new insights into the CD8 T cell population responsible for containment of HSV-2 reactivation.

  • A unique population of CD8+ T cells persists at the DEJ for at least 8 weeks post healing in human HSV-2 infection. In contrast, traditional CD8 T cells, which are mostly absent in the previously infected DEJ regions, are present in biopsies obtained near the blood vessel and in uninfected DEJ regions.
  • In contrast to normal CD8 T cells, the DEJ-associated CD8+ T cells lack expression of the chemokine receptors CCR7, CCR8, CXCR4, CXCR6, CXCR7, and S1PR1 that are required for lymphocyte egress and trafficking. The authors hypothesize that this may aid in the long-term persistence of these cells at former sites of infection.
  • Transcriptional and protein expression analysis of the DEJ-associated CD8+ T cells demonstrated that these cells have an activated phenotype with upregulation of genes important for antiviral and cytolytic function, unlike the CD8 T cells in the control site biopsies.
  • Sequencing of the TCR repertoire in the DEJ CD8+ T cells reveals that these cells are oligoclonal with a diverse usage of TCR variable-genes and demonstrates that these cells are distinct from mucosa-associated invariant T cells and NK T cells.
  • Sequential analysis of the TCR repertoire in a subset of patients over 2.5 years revealed dominant clonotypes that overlapped among multiple recurrences, supporting the notion that the DEJ CD8+ T cells are a long-lasting tissue-resident population.

Taking their data together, the authors conclude that the DEJ CD8+ T cells may play a critical role in immune surveillance and initial containment of HSV-2 reactivation in human peripheral tissue. This work has important implications for vaccine strategies and other immunotherapeutic approaches focused on regulation of HSV2 infection and reactivation. This study also provides important precedence for dissecting the immune response in other types of mucosal infections.

Reviewed by Michelle L. Hermiston, MD, PhD, University of California, San Francisco.

Identification of T Regulatory Type 1 Cell-specific Surface Markers

A review of Gagliani N., et al. Coexpression of CD49b and LAG-3 Identifies Human and Mouse T Regulatory Type 1 Cells. Nature Medicine. 2013. Apr., 28; doi: 10.1038/nm.3179 PMID: 23624599

Multiple types of T regulatory cells (Tregs) are involved in maintaining immune tolerance and preventing autoimmunity, including natural Tregs (nTregs) and inducible Tregs (iTregs). Unlike nTregs, which develop in the thymus, iTregs are induced in the periphery and consist of several subtypes that utilize different suppressive mechanisms and can be distinguished by the cytokines they produce. The different subsets of Tregs also express varying levels of Foxp3, the major transcription factor responsible for driving nTreg differentiation. T regulatory type 1 (Tr1) cells are a type of iTreg that express low levels of Foxp3 and produce high levels of the cytokine IL10. Tr1 cells have been shown to be highly immunosuppressive and capable of restoring tolerance in several immune-mediated diseases. Elevated Tr1 cell levels have also been associated with improved outcomes following hematopoietic stem cell transplantation (HSCT). However, the study and clinical evaluation of Tr1 cells has been hindered by the lack of specific cell surface markers for isolating and tracking them. In this study, Nicola Gagliani and colleagues may have solved this problem. They demonstrate that the coexpression of two markers on CD4+ T cells, lymphocyte activation gene 3 (LAG-3) and CD49b, the 2 integrin subunit of very late activation antigen-2 (VLA-2), identifies Tr1 cells in both humans and mice.

  • The authors first compared gene expression profiles between human Tr1 cell clones and undifferentiated CD4+ TH0 cells to identify T cell markers differentially expressed between the two subsets. They identified 17 differentially expressed genes, including the genes encoding LAG-3 and CD49b. Differential expression of the three markers was also confirmed at the protein level by FACS analysis.
  • CD49b and LAG-3 coexpressing CD4+ T cells isolated from human peripheral blood were shown to possess the key phenotypic features of Tr1 cells. Specifically, they were Foxp3 low, produced large amounts of IL10, and suppressed the proliferation of effector T cells in vitro.
  • In a previous study, the authors had shown that Tr1 cells accumulate in the small intestine of mice following treatment with CD3-specific monoclonal antibodies. In this study, the authors showed that CD4+CD49b+LAG-3+ T cells isolated from intestines and spleens from these mice produce high amounts of IL10, and suppress T cell proliferation in vitro. They also showed that the adoptive transfer of CD4+CD49b+LAG-3+ T cells from these mice could reduce colitis in a model of inflammatory bowel disease (IBD) indicating that CD49b and LAG-3 coexpression also identified mouse Tr1 cells.
  • The authors then showed that CD49b and LAG-3 were specifically coexpressed on Tr1 cells, and not TH1, TH2, TH17 or Foxp3+ Treg isolated from mice.
  • Human T cells cultured under Tr1-polarizing conditions were also shown to upregulate coexpression of CD49b and LAG-3. In addition, FACS sorted CD4+CD49b+LAG-3+ T cells were shown to be more suppressive and produce more IL10 than unsorted bulk Tr1 cultures indicating that CD49b and LAG-3 could be used to purify Tr1 cells following in vitro polarization.
  • Higher percentages of circulating CD4+CD49b+LAG-3+ T cells were measured in subjects with B-thalassemia who demonstrated persistent mixed chimerism following allogeneic HSCT, which has been shown to be associated with increased levels of circulating Tr1 cells.

This study identified two selective markers for Tr1 cells, in both humans and mice, that will facilitate the investigation of these cells in a variety of clinical and experimental settings. Ultimately, use of these markers should make it possible to isolate and track Tr1 cells, and could lead to novel therapeutic strategies for transplantation, cancer, and autoimmune diseases.

Reviewed by Elizabeth Jaffee, MD, Johns Hopkins Institute for Clinical and Translational Research and Eric Lutz, PhD, Johns Hopkins University, Sidney Kimmel Cancer Center

Potential Vaccination Approach for Diverting Natural Immunodominance and Targeting More Diverse T Cell Epitopes

A review of Hansen S.G. et al. Cytomegalovirus Vectors Violate CD8+ T Cell Epitope Recognition Paradigms. Science 2013; 340, 1237874. PMID: 23704576

CD8+ T cells detect intracellular pathogens through T cell receptor (TCR)-mediated recognition of short 8-10 amino acid long pathogen-derived peptides (epitopes) loaded onto and presented by MHC class I (MHC-I) on the surface of infected cells. Although thousands of epitopes can potentially be generated by a pathogen, T cell responses are generally restricted to relatively few pathogen-derived epitopes (immunodominant epitopes). Despite this, most pathogens can be controlled by T cell responses directed against narrow pools of immunodominant epitopes. However, other pathogens, such as HIV and its simian relative SIV, can evade immunodominant T cell responses and are unable to be controlled by naturally induced immune responses. In this study, Scott Hansen and colleagues demonstrate that it may be possible to genetically manipulate vaccine vectors in order to direct T cell responses against a broader noncanonical repertoire of epitopes. They show that this vaccination strategy can induce protective immune responses against SIV.

  • In a prior study, the authors showed that vaccination of rhesus macaques with SIV protein-encoding rhesus cytomegalovirus (RhCMV) vectors resulted in protective immune responses in approximately 50% of treated animals. However, virologic control was not associated with enhanced T cell responses to canonical SIV epitopes.
  • In this study, the authors used intracellular cytokine analysis to compare SIV epitopes recognized in rhesus macaques with naturally controlled SIV infections, or after vaccination with conventional SIV vaccines or RhCMV vectors.
  • As previously shown, RhCMV vectors failed to elicit CD8+ T cell responses against canonical epitopes associated with SIV infection and conventional SIV vaccination. Instead, RhCMV vaccination induced responses to a broader (> 3 times greater number of epitopes) repertoire of highly promiscuous unconventional epitopes, over half (63%) of which were restricted to MHC class II (MHC-II).
  • CD8+ T cells recognizing these unconventional SIV epitopes were capable of recognizing SIV-infected cells indicating that the processing and presentation of these unusual epitopes are CMV-independent. However, CMV gene expression was required for priming T cell responses directed against them.
  • To determine which RhCMV genes were responsible for redirecting T cell responses, vectors with different RhCMV gene deletions were tested. The RhCMV-encoded Rh189 gene (human CMV US11 ortholog) was shown to be responsible for suppressing canonical CD8+ T cell responses, and T cell responses against unconventional epitope were only induced in the absence of Rh157.5, Rh157.4 and Rh157.6 (human CMV UL128, UL130 and UL131 orthologs).

These data suggest that hierarchies of CD8+ T cell responses are more flexible than had been thought, and may primarily be controlled by immunoregulation during priming rather than by limitations in antigen processing and presentation or the composition of the T cell repertoire. These data also identified several CMV genes capable of regulating the profile of epitopes recognized by CD8+ T cells, providing a novel approach for selectively programming CD8+ T cell responses with CMV vector –based vaccines.

Reviewed by Elizabeth Jaffee, MD, Johns Hopkins Institute for Clinical and Translational Research and Eric Lutz, PhD, Johns Hopkins University, Sidney Kimmel Cancer Center

Back to top.

Human Immunophenotyping Update

Detection and Analysis of Reate Events by Flow Cytometry

J. Philip McCoy, Jr., National Heart, Lung and Blood Institute (NHLBI) and the Center for Human Immunology, National Institutes of Health

There is an ever-increasing demand for those using flow cytometry to detect and quantify rare events, such as dendritic cell subsets, circulating tumor cells (CTCs), circulating endothelial cells (CECs) and endothelial progenitor cells (EPCs), to name a few. Whereas the enumeration of relatively abundant T cell subsets such as CD4 and CD8 is straightforward, the accurate detection of rare events demands rigorous attention to elimination of potential artifacts. There is no frequency of events that defines what is ‘rare’, but it is generally accepted that cellular populations constituting less than 1% of the total population require special measures to ensure the accuracy of immunophenotyping. In this article, we will review sources of potential artifact in rare event analysis and strategies to cope with these pitfalls, regardless of the rare population to be detected. Specific marker combinations for various rare populations will not be addressed, as these are often controversial and would require more discussion than is permitted by current space.

Sources of artifacts and potential solutions:

  1. Dead cells. Dead cells are a key source of non-specific staining as these often bind, or trap, various fluorochrome-conjugated antibodies. This can lead to dead cells being mis-identified as the rare cells of interest. Dead cells are seldom a problem when immunophenotyping abundant populations in fresh peripheral blood of healthy donors, as the overwhelming majority of leukocytes are live in these specimens if processed promptly. However, when attempting to detect rare events such as CTCs or CECs, which may have a frequency of 1 cells per million leukocytes (or less), even an occasional dead cell in the sample can be a problem. To address this problem one can use of a viability stain to identify the dead cells and exclude them from analysis. A number of stains are available for this purpose, ranging from propidium iodide and other nuclear dyes to the newer live/dead stains that are available with various emission spectra. By carefully selecting the appropriate viability stain, these can be incorporated into most antibody staining panels without a problem.
  2. Cell doublets. Cytometers recognize events, or particles, passing through the laser beam, but do not recognize cells per se. Thus two cells (or more), or one cell and debris, stuck together will be recognized as only a singular event. If these cells have different phenotypic characteristics, the detected event will appear as a composite of these phenotypes, resulting in a misinterpretation of the actual staining of each cell. This can lead to a myriad of confusing phenotypes. Exclusion of cell doublets can be performed by pulse width analysis, and by simply acquiring both area and height measurements of forward and side scatter, plotting these parameters against one another, and excluding outliers from further analysis.
  3. Nucleation. With the exception of mature erythrocytes and platelets, all cells in the human peripheral blood are nucleated. A least one group has reported that clumps of platelets can be misinterpreted as CECs (ref), and it is likely clumps of cellular debris could also be misinterpreted. The inclusion of nuclear stain such as Hoechst, which can stain nuclei in viable cells, will ensure that the events identified are nucleated and therefore cellular. It should be emphasized that this staining is distinct from the viability stain described above, which identifies dead cells, sometimes by staining nucleic acids. Those stains are not membrane permeant whereas dyes such as Hoechst are membrane permeant.
  4. Collection of sufficient events. Rare events are often found at frequencies of 1 event per hundred thousand or per million leukocytes. It should be evident therefore that the collection of a 10,000 or 20,000 event file will not suffice. Even file sizes as large as 100,000 or 1 million events may not be sufficient to gather statistically significant, and reproducible, data(1). In planning these analyses one should look at the expected frequencies of the rare events and plan to acquire files large enough to contain an appropriate number of rare events that will satisfy a consulting statistician. These files may contain millions of events and could require a lengthy period of time to acquire (perhaps even an hour per file).
  5. Carry-over in the cytometer. An often overlooked source of artifacts in flow cytometry is the carry-over of cells from previous experiments. The small diameter tubing in cytometers can harbor cells after analysis, even though virtually all modern cytometers have systems to backflush these events out. Nonetheless there can be a small number of residual events that are not flushed out, but can be carried over into the next experiment. In most routine analysis, this contamination is of such low magnitude that it is doesn’t interfere with analysis of the data. In rare events analysis, where acquisition may take a long time, perhaps even an hour for one file, these rare contaminants can be problem. An easy test is to run a particle-free solution as a sample on the cytometer for 30 minutes and note the number of events collected. Removal of the contaminants can generally be accomplished by a thorough cleaning of the instrument (30 minutes or more). Efficacy of cleaning can be monitored by repeating the test above and looking for a decrease in the number of events collected.
  6. Interruption in the fluidics. Flow cytometry is dependent upon a stable fluidic system delivering the cells in a focused stream though points of laser excitation. While these fluidic systems are generally stable, minor fluctuations can occur, particularly when large data files are acquired over a long period of time. Such fluctuations have been demonstrated to directly affect to fluorescence detected on the cells(2). Brief changes in fluorescence could lead to false positive events. Causes of these fluctuations can include small bubbles, vibrations, and air supply interruptions, among others. Perturbations in the fluidics can easily be monitored in each experiment by including “time” as a parameter. This simply lets one display any parameter (usually a fluorescence channel) against the relative time at which it was acquired. By plotting this, any fluctuations in the fluidics become readily apparent and can be gated out from further analysis.
  7. Non-specific binding of antibodies. A well-known source of artifact in flow immunophenotyping is the non-specific binding of antibodies to irrelevant cells. The most common solution for this is the use of heat-inactivated normal serum to pretreat the cells and take up these binding sites, although FcBlock (a cocktail of CD16+CD32) is also used for this purpose. When staining human cells with mouse monoclonal antibodies there is some debate whether it is best to use mouse serum or human serum. Strong arguments can be made for either. Our laboratory uses mouse serum because of easy availability and reduced biohazard concerns.
  8. Exclusion of cells not of interest. (“Eliminate all other factors, and the one which remains must be the truth.” – Sherlock Holmes). For extremely rare events it is often quite helpful to gate out more common events that express markers not found on the rare cells. For CTCs and CECs, this is usually CD45, as it is present on all leukocytes but neither CTCs nor CECs. In other situations, a lineage mixture (such as CD19+CD3+CD33+CD14 all in the same fluorophore) might be used if it is suspected that the target cells might have dim CD45 expression. By gating out the predominant population of cells, it now becomes an easier task to use specific markers to identify the rare populations.

In addition to the principles above, investigators undertaking rare event analysis should also consider how they will report their findings. The unit of measure should be stated (events per ml of blood, per 1 million mononuclear cells?) and it should be clearly stated how these calculations were made. The investigator should also be cautious about stating sensitivity levels of their assays based on spiking experiments with cell lines, as cell lines often have different antigen intensities, cell size, and autofluorescence, than do the actual rare events. Consideration should also be given to the precision of the assay – if the same sample is run ten times, what is the variance of the results? Finally, there should be strong consideration to validation of the assays. How does one know that the rare cell identified is actually the cell type of interest and not an artifact? Can that cell be sorted and used for single cell PCR to validate the phenotype?

Rare event analysis by flow cytometry is tedious and filled with potential artifact, but it is hoped that the principles outlined above will be of use for laboratories attempting these assays, and ultimately result in more reproducible data. Listed below are suggested references for further reading.

References and Suggested Reading:

  1. Roederer M: How Many Events Is Enough? Are You Positive? Cytometry Part A. 73A (2008) 384385.
  2. Seamer L, Sklar LA: Time as a flow cytometric parameter. Methods Cell Biol. (2001) 63:169-83.
  3. Khan et al: Detection of Circulating Endothelial Cells and Endothelial Progenitor Cells by Flow Cytometry. Cytometry Part B (Clinical Cytometry) (2005) 64B:1–8.
  4. Donnenberg & Donnenberg. Rare-Event Analysis in Flow Cytometry. Clin Lab Med 27 (2007) 627–652.
  5. Tibbe et al: Statistical Considerations for Enumeration of Circulating Tumor Cells. Cytometry Part A (2007) 71A 154-162.
  6. Coumans et al: Challenges in the Enumeration and Phenotyping of CTC. Clin Cancer Res (2012) 18:5711-5718
  7. Hedley and Keeney: Technical issues: flow cytometry and rare event analysis. Int J Lab Hematol (2013) 35:344-350

Back to top.

Concepts in Basic Immunology: Principles and Therapeutic Applications

Innate Lymphoid Cells: New Awareness of Some Old Players in Immunity and Disease

Andrew H. Lichtman, MD, PhD, Brigham and Women's Hospital

Progress in understanding the innate immune system, which is phylogenetically older and arguably more essential for day to day survival than the adaptive immune system, had been shadowed for several decades by the brisk pace of advances in the molecular and cellular basis of adaptive immunity beginning in the 1980’s. The discovery of immune function of a fly Toll like receptor (TLR) in 1996 and characterization of mammalian TLRs soon after heralded the beginning of a new phase in innate immunity research, in which the molecular basis of pathogen associated pattern recognition and responses has been elucidated in great detail. Nonetheless, it appears that we remain in a catch-up phase in which fundamental knowledge of the cellular effectors of innate immune responses are still being sorted out. A prime example of this is the emerging knowledge of subsets of innate lymphoid cells (ILCs) that secrete different cytokines typically attributed to helper T cell subsets. These ILC subsets, some of which were not even known to exist a few years ago, appear to be of central importance in constitutive barrier immunity, innate responses to infection, and the pathogenesis of inflammatory diseases. The majority of research on ILCs has been done in murine models, which allow fate mapping studies and elimination of adaptive immune sources of cytokines, both of which have been essential for discovery in this field. Human ILCs comparable to the newly identified murine cells have also been found, and their roles in human disease are now being actively studied.

Distinguishing Features of Innate Lymphoid Cells
ILCs are bone marrow derived, and have the morphology of lymphocytes, but they do not rearrange or express TCR or Ig genes, nor do they express cell surface markers typical of myeloid cells. ILCs do arise from the common lymphoid progenitor and require expression of the IL-7 receptor or, in the case of NK cells, the 1L-15 receptor. In contrast to B and T cell lineages, ILC development also requires the transcriptional regulator called inhibitor of DNA-binding 2 (ID2). Two types of ILCs that have been thoroughly investigated for many years include Natural Killer (NK) cells and lymphoid tissue-inducer (LTi) cells. NK cells were first described in 1975, and have similar anti-viral functions as CD8+ cytotoxic T lymphocytes. LTis, which were first described in 1997, produce cytokines required for the formation of lymphoid tissues during embryonic development. In adults they are needed for regeneration of lymphoid organ structure after viral infections. More recently, additional members of the ILC family of cells distinguished by the cytokines they produce have been identified, and the grouping of ILCs into three subsets, all arising from a common ID2+ precursor, has been proposed (1) . These three subsets have distinct transcription factor and cytokine expression profiles which are respectively similar to TH1, TH2 and TH17 cells.

Innate Lymphoid Cell Subsets
Group 1 ILCs include IFN and TNF expressing cells, some of which require the T-bet transcription for development and function, all features shared by TH1 cells. The prototype of this subset is NK cells, mentioned above. There is some controversy about the T-bet dependency of most NK cells, and some investigators in this field do not agree that NK cells should be grouped with other ILCs. In addition, T-bet-dependent IFN secreting ILCs with minimal cytotoxic function are also put in Group 1, even though they may arise from plastic Group 3 ILCs, described below. The major functions of Group 1 ILCs are immunity against viruses and other intracellular pathogens, via perforin and granzyme mediated cytotoxicity and by enhancement of inflammatory responses. Group 2 ILCs (previously called natural helper cells or nuocytes) mainly produce one or more of the “type 2” cytokines IL-5, IL-9, IL-13 as well as an epidermal growth factor-related cytokine called amphiregulin, and they require the transcription factors GATA3 and ROR for their development and function. In mice, group 2 ILCs are required for protection against helminths and also for homeostasis and amphiregulin mediated repair of bronchial epithelium. Group 3 ILCs produce either or both IL17A and IL-22, and depend on the transcription factor RORt as well as the aryl hydrocarbon receptor (AHR) for development and function, features shared by TH17 cells. Within this Group 3 subset are LTi cells, mentioned earlier, which produce LT, LTm, IL-17A and IL-22. Two additional types of Group 3 ILCs have been identified, which exist mainly in intestinal mucosa. These include NCR+ ILC3 cells, which produce IL-22 and express the NK activating receptor NCR1, and NCR-ILC3 cells which produce IFN IL-17 and IL-22 but not NCR1. While LTi cells are uniquely involved in formation of lymphoid tissues, all three identified types of Group 3 ILCs cells contribute to immunity to certain extracellular bacteria. In addition, the IL-22 secreting ILC3 subtypes maintain mucosal epithelial barrier homeostasis, and in the gut ILC3 cells expressing class II MHC and IL-22 present antigen to and suppress CD4+ T cell responses to commensal bacteria (2).

Human ILCs have been described that have cytokine expression patterns similar to those of Group 1, 2 and 3 subsets. The different human ILCs are identifiable as LIN- cells (lineage-negative referring to the absence of markers of T and B cells) that express the same transcription factors seen in the mouse subsets, but have different sets of surface markers than the mouse cells (3). Group 1 human ILCs include the classic NK cells, as well as CD56+ NCR1,2,3+IL-7R- poorly cytotoxic ILC1s. Human Group 2 ILCs express ST2 (IL 1RL1 or IL-33 receptor), IL 17-RB, and CRTH2 (prostaglandin D2 receptor 2). Group 3 human ILCs include ILC3 cells that are CD56+ NCR1,2,3+IL‑7Rα+, and LTi cells that are IL‑7RαhiCD45intRORγt+.

Activation and Stability of ILC Subsets
The signals that simulate cytokine production by ILCs are only partially understood, but a common theme across the subsets is that cytokines from other cell sources are the major signals that induce ILC cytokine expression (1). For Group 1 ILCs, it has been known for many years that NK cells can be activated to secrete IFN by other cytokines, including IL-12, IL-15, IL-18, and IL-2, independent of signals from the NK activating receptors that bind to membrane bound ligand on other cells. IL-5 and IL-13 secretion by ILC2 cells is stimulated by IL-25, IL-33 and thymic stromal lymphopoietin (TSLP) produced by epithelial cells. The homeostatic role of ILC3 cells in the intestine is mediated by continual steady state IL-22 secretion, i.e in the absence of infection, and this is dependent on IL-1 from mononuclear phagocytes exposed to commensal bacteria, while IL-22 secretion by ILC3 cells in response to infection depends on mononuclear IL-23.

Some plasticity has been documented in ILCs, and cytokines may drive conversion of one phenotype to another. In particular, IL-12 and IL-18 may act on IL-22 secreting ILC3 cells and convert them to IFN secreting ILC1 cells, and this switch may be accompanied by loss of RORT and gain of T-bet expression. This is a similar story to that described for TH17 cells changing into TH1 cell in some mouse models of autoimmunity.

The role of ILC subsets in disease
Pathogenic roles of ILCs have been uncovered in several mouse models of inflammatory diseases. Many of these roles are mediated by cytokines that helper T cell subsets can produce, but the importance of the ILCs as a source of the cytokines in vivo is supported by studies in Rag knockout mice, which lack T (and B) cells. ILC2 cells contribute to airway disease, including virus induced airway hyperactivity and allergy and allergic asthma, both dependent on IL-13, and papain-induced airway inflammation, a purely experimental model dependent on IL-9.   Rag-/- mice treated with anti-CD40 or infected with Helicobacter hepaticus develop colitis, and IL-17 and /or IFN producing ILCs are required for the intestinal inflammation to develop. It is likely that the pace of work on human disease associations of different ILC subsets will increase rapidly. Recent examples include reports of increased blood ILC1 cells in severe asthmatic patients (4), and high frequency of intestinal IFN producing ILC1 cells in patients with Crohn’s disease (5). A major barrier to assessing the contribution of ILCs to disease, both in experimental models and in humans, is a paucity of tools to selectively purify and delete these subsets from otherwise immunologically intact individuals.


  1. Spits, H., D. Artis, M. Colonna, A. Diefenbach, J. P. Di Santo, G. Eberl, S. Koyasu, R. M. Locksley, A. N. McKenzie, R. E. Mebius, F. Powrie, and E. Vivier. 2013. Innate Lymphoid Cells--A Proposal for Uniform Nomenclature. Nat Rev Immunol 13: 145-149.
  2. Hepworth, M. R., L. A. Monticelli, T. C. Fung, C. G. Ziegler, S. Grunberg, R. Sinha, A. R. Mantegazza, H. L. Ma, A. Crawford, J. M. Angelosanto, E. J. Wherry, P. A. Koni, F. D. Bushman, C. O. Elson, G. Eberl, D. Artis, and G. F. Sonnenberg. 2013. Innate Lymphoid Cells Regulate CD4 T Cell Responses to Intestinal Commensal Bacteria. Nature.
  3. Walker, J. A., J. L. Barlow, and A. N. McKenzie. 2013. Innate lymphoid Cells--How Did We Miss Them? Nat Rev Immunol 13: 75-87.
  4. Barnig, C., M. Cernadas, S. Dutile, X. Liu, M. A. Perrella, S. Kazani, M. E. Wechsler, E. Israel, and B. D. Levy. 2013. Lipoxin A4 Regulates Natural Killer Cell and Type 2 Innate Lymphoid Cell Activation in Asthma. Science of Translational Medicine 5: 174ra126.
  5. Bernink, J. H., C. P. Peters, M. Munneke, A. A. te Velde, S. L. Meijer, K. Weijer, H. S. Hreggvidsdottir, S. E. Heinsbroek, N. Legrand, C. J. Buskens, W. A. Bemelman, J. M. Mjosberg, and H. Spits. 2013. Human Type 1 Innate Lymphoid Cells Accumulate in Inflamed Mucosal Tissues. Nat Immunol 14: 221-229.

Figure. Subsets of Innate Lymphoid Cells ( ILCs). The common lymphoid precursor ( CLP) gives rise to an ILC precursor that expresses the inhibitor of DNA-binding 2 (ID2) transcription factor, which in turn gives to all the different types of ILCs. IL-7 is required for all ILC differentiation except NK cells, which require IL-15. The transcription factors required for each pathway of differnataton are shown in the boxes overlying the solid blue arrows. The cytokines produced by each ILC type are shown in the pink boxes. Three groups of ILCs, indicated at the left, are distinguished by the types of cytokine they produce, analagous to Th1, Th2 and Th17 cells.The cytokines that activate the ILCs to produce their signture cytokines are shown over the dotted arrows.


Back to top.

Selected Recent Clinical Trial Results

Edited by Carla J. Greenbaum, MD, Benaroya Research Institute

Phase 2A Trials of Dupilumab in Persistent Asthma with Elevated Eosinophil Levels

Clinical Trial:
An overview of Wenzel S. et al., Dupilumab in Persistent Asthma with Elevated Eosinophil Levels. New England Journal of Medicine. Published online May 21, 2013  PMID: 23688323

Disease: Asthma

Drug: Dupilumab is a fully human monoclonal antibody to the IL-4 receptor α subunit that inhibits both IL-4 and IL-13 signaling. In this trial, 300mg of dupilumab (or matching placebo) were administered as SQ weekly injections for 12 weeks.

Study Design:

  • Phase 2a randomized, placebo-controlled trial assessing whether dupilumab is safe and effective in patients with persistent, moderate-to-severe asthma and elevated eosinophil levels.
  • Patients were randomly assigned to receive once weekly subcutaneous (SQ) injections of dupilumab (n=52) or matching placebo (n=52) for 12 weeks. Patients also initially received standard therapy of fluticasone, an inhaled glucocorticoid, and salmeterol, a long-acting beta agonist. The salmeterol was discontinued at week 4, and the fluticasone was tapered and discontinued from weeks 6 through 9. Thus, in the early part of the study the effects of dupilumab were evaluated in addition to two or one drug standard-of-care therapy and later as monotherapy.
  • Patients (age 18-65) were those not well-controlled on standard therapy with moderate-to-severe asthma (with lung function measured as FEV1 ≥ 50%) and elevated blood eosinophil count (≥300 cells per µL) or elevated sputum eosinophil level (≥3%).
  • The primary efficacy endpoint was occurrence of an asthma exacerbation. A number of Th2-associated biomarkers were also measured including the eosinophil-attracting eotaxins (CCL26) and thymus and activation-regulated chemokine (TARC, or CCL17). These are C-C chemokines produced by structural cells (such as fibroblasts and epithelial cells) that are known to be upregulated by Th2 cytokines.


  • A total of 491 patients were screened in order to identify 104 eligible patients for randomization. Two-hundred sixty (260) of these screen failures were due to low eosinophil count.
  • The primary endpoint was met: 23 patients on dupilumab had an exacerbation, while only 3 patients on dupilumab did (p<0.001), an 87% reduction in exacerbations with dupilumab. The clinical relevance of the definition of exacerbations in this study is uncertain. Moreover, in the time to exacerbation analysis, the survival curves only start to favor the dupilumab group after withdrawal of standard therapy.
  • Patients on dupilumab had significantly higher absolute FEV1 and asthma symptom questionnaire scores starting at week 2 and week 3, respectively. These differences were both statistically and clinically significant (e.g. absolute FEV1 improvement of 200ml). TARC, eotaxin-3, and IgE levels were significantly lower by week 4, and were sustained or lower through the duration of the study. While the majority of patients had little or change in peripheral blood eosinophil levels in either group, 4 patients in the dupilumab group had increased blood eosinophilia suggesting a reduction in the migration of eosinophils from the blood to the lung compartment.
  • There were so significant safety issues or serious adverse events related to study drug. More patients on dupilumab experienced injection-site reactions, nasopharyngitis, nausea, and headache. Among the adverse events (AE) leading to discontinuation of study drug was one related-AE of angioedema in the dupilumab group.

Why the trial is of interest to the broader FOCIS community:
With increased understanding of mechanisms broadly underlying inflammatory and immune mediated disease, targeted therapies such this one inhibiting IL-4 and IL-13 therapy have been developed. Previous trials of the monoclonal antibodies lebrikizumab and tralokinumab did not demonstrate improvements in asthma symptoms and quality of life; this may be attributed to the fact that these therapies only target IL-13 or because of patient selection.

Thus equally important to identifying targets for disease pathogenesis is identifying the right subjects out of the heterogenous population with clinical evidence of disease. In this study, investigators built on the observed relationship between hypereosinophilia and therapeutic outcome in trials with other agents targeting these pathways. The investigators enrolled only those individuals with a hypereosinophilia as well as those with poorly controlled disease on standard therapy. While this selection targeted only ~20% of otherwise eligible subjects, it enhanced the probability that the drug would prove effective for the endpoints tested. Though the clinical importance of some of these endpoints is debatable, the marked effect on the primary endpoint as well as the clear demonstration of changes in targeted biomarkers make this an important proof of concept study. The concept being proved is not only that a correct target was selected; but that selectively enrolling subjects may limit the non-responders and enable a therapy to continue towards clinical development that otherwise might have been discounted. The down side of this approach from a clinical or public health point of view is that a therapy that is effective in only a selected group of subjects may still be demanded by clinicians and patients with the same disease without the same characteristics, thus resulting in increased use (and therefore costs) of a therapy without real benefit. Ultimately, results of this Phase 2 study showing clinical efficacy with withdrawal of background therapy will need to be replicated in long-term, Phase 3 studies before it is known whether it will be employed in treating those with asthma someday. These issues in no way detract from the importance of this trial demonstrating that targeting therapeutics to the right population can give you the results you expect.

Reviewed by Steven Lamola, MD Benaroya Research Institute

Phase 2 Trial of Anti-IP-10 Antibody for Ulcerative Colitis

Clinical Trial:
An overview of Mayer L. et al., Anti-IP-10 antibody (BMS-936557) for ulcerative colitis: a phase II randomised study. Gut 2013;0:1-9. PMID: 23461895

Disease: Ulcerative Colitis

Drug: BMS-936557 (formerly MDX-1100) is a fully human monoclonal antibody that targets the chemokine interferon-γ-inducible protein-10 (IP-10), also referred to as CXCL10.

Study Design:

  • 8-week, Phase II, double-blind, multicenter, randomized study to evaluate safety and efficacy of BMS-936557 in patients with moderately-to-severe active ulcerative colitis (UC).
  • Patients were randomly assigned to receive 10 mg/kg to 1000 mg of BMS-936557 (n=55) or matching placebo (n=54) intravenously every other week for 8 weeks. Eligible patients were ≥18 years old with an active UC flare within 2 weeks of enrollment while on stable treatment with standard, non-biologic therapy.
  • The primary efficacy endpoint was the rate of clinical response 2 weeks after the last study treatment. Clinical response was defined on the basis of standardized scores which include patient reported symptoms such as stool frequency and rectal bleeding.


  • The primary and secondary efficacy endpoints were not met in this study. The proportion of patients who achieved clinical response was not significantly different in the BMS-936557-treated group (52.7%) versus placebo-treated group (35.2%; p=0.083). However, patients who failed to adequately record symptoms were considered non-responders. That means that all such patients (n=7) were analyzed as if the drug was not helpful. This “imputation model” is an accepted statistical approach to deal with missing data points. If instead of imputing these subjects’ results as being negative the trial was analyzed (post-hoc) using the data that actually was available, the clinical response rates were significantly higher in the active treatment group.
  • The investigators also explored (post-hoc) the relationship of the drug levels [(trough concentrations (Cmin)] with clinical response rate. Sixteen patients with the highest Cmin values were observed to have a significantly higher clinical response rate (87.5%) versus the placebo group (p<0.001). Also, a significantly greater proportion of patients with BMS-936557 Cmin ≥ 100 µg/ml were determined to be in histological remission compared with placebo (73% vs. 41%, p=0.004), exhibiting a marked reduction in inflammatory infiltrates and decrease in erosion, ulceration and crypt destruction.
  • There were no related significant adverse events or deaths during this study. The BMS-936557-treated group had a numerically higher frequency of infections compared with placebo. This was not the case in a previous Phase 2 study of BMS-936557 in patients with rheumatoid arthritis. The safety profile was the same for patients in the Cmin ≥ 100 µg/ml subgroup.

Why the trial is of interest to the broader FOCIS community:
Chemokines and chemokine receptors are central to the inflammatory process and therefore attractive therapeutic targets. This trial illustrates a potential advantage of targeting chemokines as opposed to their receptors in specific autoimmune diseases. Binding of the chemokine IP-10 to its receptor, CXCR3, induces differentiation and recruitment of Th1 and Th17 cells responsible for inflammation and tissue destruction. In addition, IP-10 has been shown to operate independently of its receptor, as a direct inhibitor of epithelial and endothelial cell proliferation, 1 and as a contributor to bone destruction (by upregulating RANKL)2 and apoptosis of pancreatic β-cells.3

Why then did this trial have a negative outcome? This trial illustrates how significantly study design can impact efficacy outcomes. Failure to meet the primary endpoint might be attributed to insufficient compliance on the part of participants to complete the electronic diary of symptoms. Thus, this trial not only had a negative outcome, it is a failed trial in that we are still unclear as to whether or not this therapy is worth pursuing. Though use of imputation attributing missing data as negative is an appropriate pre-specified analysis plan; this trial also illustrates that the choice of analytic approach may often dramatically affect the “results” of a study. 

In addition, the relationship between drugs levels and outcome suggests that a higher dose might have resulted in a different outcome. Because this was a post-hoc analysis, even this premise would need additional testing. There is still reason to believe that, at the right dose, an anti-IP-10 antibody may be effective in UC because of both inhibition of Th1 cell differentiation and trafficking, and promotion of crypt epithelia cell proliferation and regeneration. Based on both this study in UC and the previously published Phase 2 study in rheumatoid arthritis – which showed modest but significant response –,4 additional dose-ranging studies of BMS-936557may be informative.


  1. Campanella et al., PLoS ONE 2010: 5(9): e12700.
  2. Kwak et al., Arthritis Rheum 2008:58:1332-42.
  3. Schulthess FT et al., Cell Metab 2009:9:125-39.
  4. Yellin M et al., Arthritis Rheum 2012:64(6): 1730-1739.

Reviewed by Steven Lamola, MD Benaroya Research Institute

Interleukin-1 Antagonism in Type 1 Diabetes of Recent Onset

Clinical Trial:
An overview of Moran A. et al., Interleukin-1 Antagonism in Type 1 Diabetes of Recent Onset: Two Multicenter, Randomized, Double-blind, Placebo-controlled Trials. Lancet. Published online April 5, 2013
PMID: 23562090

Disease: Type 1 Diabetes

Drugs: Canakinumab is a human monoclonal anti-IL-1 antibody, given in this trial as a monthly subcutaneous (SQ) injection of 2mg/kg (maximum 300 mg). Anakinra is a recombinant human IL-1 receptor antagonist, given as a 100mg SQ dose every morning.

Study Design:

  • Results of two randomized, placebo-controlled Phase 2a trials assessing whether interleukin-1 blockade improves β-cell function in recent-onset type 1 diabetes (T1D) have been published in the same Lancet paper.
  • In one study, 69 patients were randomly assigned to receive 12 monthly injections of either canakinumab (n=47) or placebo (n=22). Eligible patients were 6-45 years old at onset of T1D diagnosed within 100 days of enrollment, were positive for at least one diabetes-associated autoantibody, and had a stimulated peak C-peptide during mixed meal tolerance test (MMTT) ≥ 0.2 nmol/L.
  • In the other study (AIDA Study) 69 patients were randomly assigned to receive daily injections of anakinra (n=35) or placebo (n=34) for 9 months. Eligible patients were 18-35 years old, had their first reported symptoms of T1D within 12 weeks of enrollment, were positive for GAD-65 autoantibodies, and had a stimulated peak C-peptide during MMTT ≥ 0.2 nmol/L.
  • All subjects in both trials received standard intensive diabetes treatment.
  • The primary endpoint of both studies was comparison of the area under the curve (AUC) stimulated C-peptide response over a 2-hour MMTT, at the end of the treatment period.


  • The primary endpoint was not met in either study: stimulated C-peptide concentrations did not differ between the active treatment and placebo arms. The difference in AUC C-peptide in the canakinumab study at 12 months was 0.01 nmol/L (95% CI -0.11 to 0.14; p=0.86); in the anakinra study at 9 months it was 0.02 nmol/L (95% CI -0.09 to 0.15; p=0.71).
  • Secondary endpoints also did not demonstrate significant difference between active and placebo groups: e.g. in neither trial was HgA1c or insulin dose over time different between treatment arms.
  • There were no significant safety signals in either study, with the exception of a higher rate of moderately severe injection site reactions in the anakinra group versus placebo. Despite the possible anti-inflammatory effects of these drugs, neither active treatment group experienced more frequent or severe infections. Neutropenia also did not differ significantly between active and placebo groups.

Why the trial is of interest to the broader FOCIS community:
The failure of these trials to demonstrate efficacy of IL-1 blockade in T1D in no way refutes the decades of preclinical research implicating this master cytokine as an important player in the pathogenesis of T1D. There is substantial evidence that IL-1 (or specifically IL-1β) has both direct toxic effects on β-cells and promotes expansion and differentiation of pathologic T-cells, including Th1 and Th17 cells. Instead, these trials illustrate that it is not enough to pick the right target. Equally critical is intervening on that target at the stage of disease development at which research suggests it plays its major role. As pointed out in this paper, IL-1β gene expression in peripheral blood monocytes is elevated at diagnosis of T1D, but levels are back to normal within 1 month of diagnosis. It is also known that diabetes autoantibody-positive first-degree relatives of individuals with T1D – who are at higher risk of developing T1D – have higher levels of IL-1β expression by monocytes and dendritic cells. It is therefore reasonable to expect that it may be too late to neutralize IL-1 after development of T1D. It should also be pointed out that blockade of IL-1β signaling has only slowed progression to diabetes in NOD mouse, not prevented or reversed disease. Therefore, in addition to intervening at the right time, use of a therapy like IL-1β blockade may only be effective in rational, synergic combination with other agents.

Reviewed by Steven Lamola, MD Benaroya Research Institute

Trial of Sirolimus Conversion to Reduce Risk of SCC Recurrence in Renal Transplant Recipients

Clinical Trial:
An overview of Hoogendijk-van den Akker JM et al. Two-Year Randomized Controlled Prospective Trial Converting Treatment of Stable Renal Transplant Recipients With Cutaneous Invasive Squamous Cell Carcinomas to Sirolimus. Journal of Clinical Oncology 2013; 31(10):1317-23. PMID: 23358973

Disease: Squamous cell carcinoma (SCC) in setting of renal transplantation

Drug: Sirolimus is an inhibitor of mammalian target of rapamycin (mTOR), a key serine-threonine kinase involved in regulation of cell growth and proliferation.

Study Design:

  • Randomized trial to evaluate recurrence of cutaneous invasive squamous cell carcinoma (SCC) in renal transplant recipients two years after assignment to either changing their immunosuppression regimen to include sirolimus (n = 74) or maintaining current immunosuppression (n = 81). Patients and transplant physicians were aware of trial assignment, however, evaluating dermatologists and pathologists were not (i.e. they were masked).
  • Participants were adults who were more than 12 months post-kidney transplant and had at least one biopsy-confirmed cutaneous invasive SCC. Subjects were all receiving maintenance calcineurin inhibitor, azathioprine, mycophenolate, and/or steroids for at least 12 weeks before randomization. Exclusion criteria included metastatic cutaneous SCC and internal malignancies. Randomization was stratified by transplantation center, number of SCCs, and recipient age.
  • Evaluation of invasive SCC recurrence was conducted by a dermatologist every three months (suspect lesions were biopsied for interpretation by a dermatopathologist). Laboratory data were obtained every three months, and adverse events evaluated through the two years of the study.


  • At the primary analysis timepoint of 2 years, subjects who were converted to sirolimus experienced a nonsignificant reduction in risk of SCC recurrence (HR of 0.76, p=0.255) compared with subjects who remained on conventional immunosuppression. However, there was a significant 50% risk reduction in development of new lesions after one year (p=0.021). Similarly, the subgroup of subjects who entered the trial with only one previous SCC were significantly less likely to develop a new lesion after one year (p=0.044).
  • 42% of subjects discontinued sirolimus because of adverse events.

Why the trial is of interest to the broader FOCIS community:
The risk of immunosuppressive drugs causing lymphoma – specifically lymphomas positive for Epstein-Barr virus (EBV) – has been recognized for some time. Less appreciated is the fact that cutaneous squamous cell carcinomas (SCCs) are the most common post-transplantation cancers. The risk of SCC in an organ transplant recipient is reportedly 65 to 250 times that of the general population, an effect usually attributed to inhibition of T cell-mediated immune surveillance. However, other mechanisms have also been proposed. There is experimental evidence as well as previous clinical studies that suggest use of mTOR inhibition would reduce the risk. For example, mouse studies suggest that calcineurin inhibitors (cyclosporine and tacrolimus) directly promote cancer development via a TGF-β mechanism and that mTOR inhibitors may block tumor progression in various cancer models. Understanding the mechanism for this markedly increased risk and what strategies may reduce this risk is important for immunotherapeutic approaches in other inflammatory and autoimmune diseases.

Why then, was the primary measure in this trial (reduction of new SCC two years after randomization to sirolimus) negative – i.e. why didn’t the change in immunosuppression regimen significantly reduce new cancers? The answer is that we don’t know because almost half the subjects assigned to sirolimus stopped taking the drug. This is an example therefore of a failed clinical trial – not because the primary outcome was negative, but because the results leave us without an answer to the question posed.

The demonstration that the exploratory endpoints at 1 year suggested a benefit of sirolimus only indicates that a new study is needed. As noted by the authors, additional testing of this concept will require strategies to reduce the adverse effects of sirolimus.

Reviewed by the European Society of Organ Transplantation (ESOT) and the Centre for Evidence in Transplantation (CET)

Click here to send your comments and suggestions for future Translational Immunology Updates.

Translational Immunology Update is the official publication of the Federation of Clinical Immunology Societies. Scope: Translational Immunology Update aims to provide a knowledge-sharing forum for basic, clinical and translational scientists around all aspects of human immunology including normal and cross-disease physiology, communication between laboratory-based and clinically-based scientists, the lack of negative results that are published and the difficulty for readers to keep up with important literature.

Federation: All members of FOCIS receive Translational Immunology Update. For more information about FOCIS, address correspondence to: Federation of Clinical Immunology Societies, 11950 W. Lake Park Drive, Suite 320, Milwaukee, WI 53224, Phone: (414) 359-1670, Fax: (414) 359-1671, Email: This email address is being protected from spambots. You need JavaScript enabled to view it.
Advertising: Inquiries concerning advertising should be addressed to Sarah Martis, FOCIS Executive Director.
Permissions: Copyright © 2013 Federation of Clinical Immunology Societies
Apart from any fair dealing for the purposes of research or private study, or criticism or review, as permitted under the Copyright, Designs and Patent Act 1988, this publication may be reproduced, stored or transmitted, in any form or by any means, only with the prior permission in writing of the publishers, or in the case of reprographic reproduction, in accordance with the terms and licenses issued by the Copyright Licensing Agency.
Translational Immunology Update is published bimonthly by the Federation of Clinical Immunology Societies. You may opt-out of receiving the publication at any time by clicking the unsubscribe link in the email.

Advertising: Inquiries concerning advertising should be addressed to This email address is being protected from spambots. You need JavaScript enabled to view it. , FOCIS Executive Director.
Permissions: Copyright © 2013 Federation of Clinical Immunology Societies
Apart from any fair dealing for the purposes of research or private study, or criticism or review, as permitted under the Copyright, Designs and Patent Act 1988, this publication may be reproduced, stored or transmitted, in any form or by any means, only with the prior permission in writing of the publishers, or in the case of reprographic reproduction, in accordance with the terms and licenses issued by the Copyright Licensing Agency.
Translational Immunology Update is published bimonthly by the Federation of Clinical Immunology Societies. You may opt-out of receiving the publication at any time by clicking the unsubscribe link in the email.