by ecancer reporter Clare Sansom

Cancer vaccines, which are composed of one or more molecules designed to elicit tumour-specific responses from patients’ immune systems, have recently been shown in clinical trials to have some definite therapeutic benefit.

However, it has become clear that in many vaccinated patients the expected antigen-specific T-cell response is not associated with an objective anti-tumour response.

Scientists have already identified several mechanisms that may allow tumour cells to escape from attack by the vaccine-primed immune system, some of which involve the adjuvant in which the vaccine is delivered to the patient.

Many peptide-based cancer vaccines that have been tested in clinical trials are formulated in incomplete Freund’s adjuvant (IFA), in which the antigen is emulsified in mineral oil with mannide monooleate as a surfactant.

A group of researchers led by Willem Overwijk at the University of Texas MD Anderson Cancer Center, Houston, Texas, USA have now tested the immune and anti-tumour responses of an IFA- based peptide vaccine in a mouse model of melanoma.

Transgenic melanoma-bearing mice were vaccinated with an emulsion of a peptide derived from the pigment cell-specific glycoprotein antigen gp100 in IFA.

This vaccination initially induced a significant expansion of CD8 T cells in response to the antigen, but the T cell levels dropped to near zero after three weeks and did not rebound after subsequent vaccination (i.e. there was no memory effect).

Further experiments suggested that this very short-lived response was associated with the IFA adjuvant rather than with the particular peptide vaccine or T-cell subtype involved, and that it was not dose dependent.

Overwijk and his co-workers then used luciferase to visualise the accumulation of T cells in mice vaccinated with different formulations of gp100.

When the antigen was delivered in the adjuvant IFA, the T cells accumulated at the antigen-rich vaccination site ; in contrast, T cells induced by vaccination with a virus encoding this peptide antigen (VSV.gp100) accumulated in tumours.

The “hypo-responsiveness” of tumour cells induced by vaccination with gp100/IFA was found to be associated with apoptosis of T cells at the vaccination site.

The researchers found that T-cell apoptosis was driven by a prolonged interaction between the T cells and peptide antigen that persisted at the vaccination site, and mediated by interferon-gamma (IFN-γ) and the Fas ligand (FasL or CD95L).

Limiting the exposure of the T cells to the antigen-rich vaccine depots accumulated at the vaccine site was found to rescue them from apoptosis.

These results led Overwijk and his co-workers to suggest that limiting the persistence of antigen in depots at the site of vaccination might prevent T-cell apoptosis and induce a stronger and longer-lived immune response to the vaccine.

They tested this in the same mouse model, first simply by replacing the poorly biodegradable IFA adjuvant with saline. 

This gp100/saline vaccine did not cause antigen to accumulate at the site of injection, but also failed to induce a T-cell response.

The researchers then added a combination of immunostimulatory molecules known as covax to both the gp100/IFA and gp100/saline preparations, and found that this increased the persistence of the T cell response but did not entirely prevent antigen accumulation at the vaccine site with the IFA-based vaccine.

T cells induced by the gp100/saline/covax preparation were found to accumulate in the mouse tumours, causing suppression of melanoma growth and a reduction in the size of established tumours.

Interestingly, the IFA adjuvant was found to work better with longer (20-residue) peptide antigens, which can only be presented to T cells by relatively rare dendritic cells; the long peptide antigens did not accumulate or sequester T cells at the vaccine site, and the antitumour response was stronger.

The gp100/IFA/covax vaccine preparation, but not the gp100/saline/covax preparation, induced a strong up-regulation of genes in T cells that are involved in “dysfunctional” processes such as apoptosis, T-cell exhaustion and tolerance.

Taken together, these results suggest that the use of a long-lived vaccine preparation involving a poorly biodegradable adjuvant such as IFA will induce T cells that are rapidly sequestered before undergoing apoptosis.

In contrast, however, a less persistent vaccine preparation will induce longer-lived, functional T cells that can localise in tumour tissue and generate an anti-tumour response.

It remains to be seen how these results in mice will translate to human medicine where there is a much larger difference between the body size and the size of any conceivable persistent antigen depot.

Nevertheless, Overwijk and his colleagues suggest that the use of rapidly biodegradable vaccine antigens might increase the efficacy of many peptide-based cancer vaccines.

Reference

Hailemichael, Y., Dai, Z., Jaffarzad, N. and 16 others (2013). Persistent antigen at vaccination sites induces tumor-specific CD8 T cell sequestration, dysfunction and deletion. Nature Medicine, published online ahead of print 3 March 2013. doi: 10.1038/nm.3105