Immunogenicity Assay Standardization Efforts Underway

By Emily Bass

This year saw the launch of three separate projects designed to ease comparison of results from vaccine trials by standardizing aspects of assays that measure cellular immune responses (Elispot and intracellular cytokine, or ICC) and neutralizing antibodies.

It’s a step that’s many feel is critical for the vaccine field to move forward. “As soon as possible, we need to get a handle on comparing the types of assays used so that [the different groups doing clinical trials] can get a sense of how vaccines compare with one another,” said John Shiver of Merck Research Labs at the IAVI-organized vaccine satellite meeting in Barcelona.


HIV Vaccine Trials Network (3 sites) Walter Reed Army Institute for Research
AIDS Clinical Trials Group
National Institute of Communicable Diseases (South Africa)
Vaccine Research Center
US Centers for Disease Control 
John Mascola, Vaccine Research Center, Bethesda
David Montefiori, Duke University, Durham Christianne Moog, Institute of Virology, Strasbourg, France
Christos Petropoulos, Virologic, San Francisco George Shaw, University of Alabama, Birmingham   

BD Biosciences
Merck & Co.
National Institute of Communicable   Diseases (South Africa)
US Vaccine Research Center
 Viral Isolates





 Related Co./Research Group
Harriet Robinson (Emory University) & Bernie Moss (NIH)
Monoclonal Antibodies: 2g12, IgG1B12, 2F5, 4E10 


Elispot assays are currently the most widely used measure of vaccine-induced cellular immune responses. The assay measures the number of T-cells activated by a specific antigen. Following antigen exposure, responding cells are detected by staining for secreted (extracellular) cytokines—signaling molecules that are markers of activation. These cells are tallied by computerized “spot-counting” machines. Patricia D’Souza (National Institute of Allergy and Infectious Diseases; NIAID), coordinator of the Elispot standardization exercise, says that she was spurred to action during analysis of data from HVTN 203, a Phase II canarypox vaccine trial. “Data using the same samples and the same assays from the Duke and Seattle labs were different,” she recalls. “If two dominant labs in the first world can’t get the same data, then what’s happening in the rest of the world?”

To address this problem, D’Souza and colleagues Jo Cox (NIAID), Guido Ferrari (Duke University, Durham) and Spyros Kalams (Partners AIDS Research Center, Boston) set up a study involving 11 labs (see table above) to identify and eventually reduce sources of variation in Elispot data. Each lab received a standardized panel of peptides and other reagents, and carried out the assays on 11 different cell samples according to their own lab protocol.

Along with their own counts, these labs returned their Elispot plates, which were sent to contractors for independent counting. Results will be discussed at a meeting in Washington, DC in late November, where predictable sources of variation (i.e., those between spot-counting machines) will be identified, along with less predictable sources of fluctuation. “It’s too early to tell whether we need to focus on the assay itself, on [training for] a particular technician, or suggesting optimum parameters for Elispot readers,” says D’Souza,” stressing that this exercise is a first step towards standardization.

Intracelluar Cytokine Assays

In a related effort, Pierre-Rafick Sekaly, Program Leader and Scientific Director at CANVAC (Montreal), together with Jill Gilmour (IAVI) is coordinating a multi-lab comparison of ICC assays. ICC is a newer assay than Elispot and also measures cytokine production in response to a specific antigen. In this case, cytokines are detected inside (rather than outside) cells using fluorescently-labelled, cytokine-binding antibodies. Fluorescing cells are then counted using flow cytometry. The technique can be used to stain multiple markers, and therefore to define more precisely the population of activated cells (e.g., CD4 or CD8).All labs participating in this exercise will assay the same samples using the same peptides, antibodies and data-analysis software from Becton-Dickenson, the company which invented the flow cytometer.

The exercise, scheduled to begin before the end of 2002, will compare data from fresh, fixed and frozen samples at different time intervals following stimulation. These results and their implications for assay standardization will be discussed at the May 2003 HVTN meeting.

Neutralization Assays

Coordinated by David Montefiori (Duke University, Durham), the neutralizing antibody assay comparison involves five laboratories and an extensive panel of reagents, including eight primary isolates of HIV that have all been used as the basis for vaccines now in development (see table).

Neutralizing antibody assays provide a measure of whether an antibody or serum sample blocks HIV entry into human T-cells. This is done by incubating virus with antibody for a set time, adding cells, incubating the mixture to allow for several rounds of viral replication and then measuring the amount of virus present, usually as a function of viral antigen expression. The traditional assay does this in peripheral blood mononuclear cells (PBMC) by collecting the cell culture supernatant and testing for p24 (viral core protein) secreted from the infected PBMCs.

The collaboration will test this technique against three newer assays. One, developed by John Mascola (Vaccine Research Center, Bethesda), is a straightforward variation of the traditional PBMC assay, using antiretroviral agents to stop viral replication after a single round. These cells are then enumerated by staining intracellular p24 and analyzed by using flow cytometry. In a paper comparing this assay with the traditional PBMC assay (J. Virol 76:4810;2002), Mascola found these two assays offered highly similar results.

The two other approaches to be analyzed make use of “reporter” genes that luminesce when expressed, allowing quantification by a luminometer. One version of this assay, developed by Chris Petropoulos and colleagues (Virologic, San Francisco), builds luciferase (a firefly enzyme) into an env-deleted HIV DNA construct. Target cells are co-transfected with this and an env-containing DNA construct, producing infectious virus particles that cannot produce new, functioning virus. Instead, replication stops after a single cycle, and infected cells, which luminesce, are counted. George Shaw (University of Alabama) uses a similar approach, in which the luciferase gene is incorporated into a target cell line.

Assays such as these may produce different results than traditional methods, says Montefiori, because engineered viral particles and/or cell lines change the type of cell-surface proteins contained in the coating of new virions—which, in turn, may influence the type and magnitude of neutralization detected. The standardization exercise will provide important information on how to compare the results of assays that use virus grown and assayed in PBMCs, with those done with cloned viruses made and assayed in cell lines.

But the luciferase assays offer several technical advantages, says Montefiori. They are faster and [at least for those using luciferase-expressing cells] easier to perform than PBMC assays. They offer a five-fold reduction in cost compared to traditional PBMC assays. Montefiori says that understanding how luciferase and PBMC assays compare will help increase capacity for neutralization assays around the world. “It’s hard to get PBMC assays up and running at resource-poor international sites. Luciferase in cell lines will be much more portable.” Results from the exercise will be presented in mid-2003.