Researchers have developed a virus-based vaccine expressing a broad spectrum of undefined antigens from normal human prostate tissue that they claim completely eradicates established mouse prostate tumors in experimental animals. Reporting in Nature Medicine, the team in the U.S. and U.K. that developed the approach says the vaccine can be delivered systemically without the need for tumor targeting, appears to have no side effects, and doesn’t induce autoimmunity. Moreover, any tumor cells that do manage to evade the initial antitumor immune response are mopped up after a second-line virus-based immunotherapy comprising a different set of antigens. The Mayo Clinic’s Richard Vile, M.D., and colleagues, describe their findings in a paper titled “Broad antigenic coverage induced by vaccination with virus-based cDNA libraries cures established tumors.”
Effective cancer immuno- and virotherapies are hindered by a lack of identified tumor-associated antigens (TAAs), poor coverage of the antigenic repertoire expressed by tumors, and difficulties associated with targeting vectors to tumors in vivo, explains Dr. Vile, together with co-lead author Alan Melcher, at the U.K. Leeds Institute of Molecular Medicine and Cancer Research U.K. Clinical Center, and colleagues.
The team’s previous work had already shown that killing normal cells in situ using the adjuvant hsp70 generated T cell responses to antigens that mediated rejection of tumors of the same histological type. While this approach led to a broad repertoire of individually weak T cell responses against multiple TAAs that cumulatively resulted in strong selective pressure against immune escape, the need to inject the vectors into normal tissues often results in toxicity, they admit.
Nevertheless, building on the positive aspects of this approach, the researchers postulated that it may be possible to provide a broad antigenic repertoire for in vivo immune selection of relevant tumor-associated antigens by expressing a cDNA library of a normal tissue using a systemically delivered, immunogenic vector. This approach, they reasoned, might activate autoimmune and antitumor T cell responses, and dispense with the need for administration directly into normal tissues. They chose VSV as the vector for delivering and expressing the antigens in vivo.
The team cloned a cDNA library from normal human prostate, which they termed altered self antigen and epitope library (ASEL) into VSV, and systemically injected the antigen-expressing vector into normal experimental mice. After 60 days there was no evidence of autoimmune consequences. In contrast, intraprostatic ASEL injection resulted in the animals developing prostatitis, which was probably induced by the observed T helper type 17 (TH17) and IFN-γ responses against prostate antigens, the authors note.
They then went on to test the virus-expressed human cDNA library in mice with established prostate tumors. Intravenous injection generated a prostate-specific TH17 response and led to the CD4+ T cell-mediated cure of established tumors, without generating an IFN-γ response. The cure rate in treated animals increased in parallel with increasing numbers of injections administered. Three intravenous injections of ASEL conferred greater survival than either intratumoral ASEL. Nine intravenous injections of ASEL were sufficient to cure over 80% of mice with the prostate tumors, with no detectable autoimmune prostatitis.
Three intravenous ASEL injections typically induced initial tumor regression, but with subsequent aggressive tumor recurrence. The recurrent tumors demonstrated substantial histological differences from the parental tumors, and extensive interstitial lymphocyte infiltrates. In addition, recurrent tumors expressed genes associated with an epithelial-to-mesenchymal–type transition, and had lost, or at least reduced, expression of mouse homologs of the human RNAs encoding prostate-specific antigens in the ASEL.
To address the problem of tumor recurrence, the researchers evaluated combinations of vaccination with different libraries. They first tried a second vaccination comprising virally expressed immune-escape epitope libraries (IEELs) constructed using cDNA from the recurrent tumors of ASEL-treated mice. This had no benefit on overall survival. This failure, they suggested, may have been due to neutralization of the IEEL virus by neutralizing antibodies raised against VSV as a result of the initial ASEL vaccination. To try and negate this problem the researchers retried the sequential ASEL-IEEL treatment with the IEEL library virus preloaded onto CD8+ T cells—which they termed T(IEEL). More encouragingly, although four of seven mice treated with ASEL followed by T(IEEL) developed recurrences, these occurred more slowly than those treated with ASEL alone, and three remaining mice were recurrence- and tumor-free for over 100 days. Over two separate experiments, 7 of 14 mice (50%) treated with ASEL followed by T(IEEL) were cured long term, compared to no cures (0%) from treatment with ASEL alone. In fact, when ASEL was preloaded onto T cells, sequential therapy using T(ASEL) followed by T(IEEL) generated long-term cures with only a single recurrence.
By varying the order and type of vaccination carried out and evaluating resulting immune responses, the authors were able to determine that whereas with ASEL-mediated rejection of the original tumors was associated with CD4+ T cells and a TH17 response, sequential rejection of the recurrent tumors was associated with a TH1-like IFN-γ response mediated by CD8+ T cells.
In a final set of experiments the authors investigated how important it was that the vaccine was based on human (i.e., xenogeneic proteins), for it to be effective against the mouse tumors. They constructed a self-epitope expressed library (SEL) using cDNAs from normal mouse prostate and compared its effectiveness against mouse prostate tumors with that of the human cDNA-based ASEL library. This confirmed that the ASEL was significantly better at protecting against mouse tumors than SEL.
The ASEL treatment approach was also tumor-specific, as the prostate ASEL had no effect on mice bearing melanomas, and didn’t stimulate splenocyte reactivity against non-prostate targets, the researchers add. “This suggests that immune reactivity generated by ASEL is directed not simply against xenogeneic household proteins, but against lineage-specific antigens,” the authors conclude.
They suggest their approach could have widespread clinical utility for human cancer therapy. “Virus-expressed cDNA libraries from normal tissues of either human (true self) or xenogenic (altered self) origins can be readily constructed for off-the-shelf use and can be easily delivered systemically in vectors amenable to production at clinical grade, thereby representing a new, readily transplantable therapy with considerable potential as an antitumor agent,” the team claims. “Virus-expressed cDNA libraries represent a novel paradigm for cancer treatment addressing many of the key issues that have undermined the efficacy of immuno- and virotherapy to date.”