Using mouse models, direct injection of CpG/anti-OX40 into tumors creates a systemic immune response.
A recent study, using mouse models, showed that direct injection of a vaccine into tumors creates a tumor-specific systemic immune response.
By way of background, let's look at T cell-driven cancer treatment. T cells within the body have been increasingly used for cancer treatment in recent drug developments. Training the T cell to increase its response has been the focus on treatment through: enhancing a T cell response in 1 region to spread throughout the body and attack cancer cells; introducing antibodies to attack antigens on cancer cells; and removing the “stop points” in T cells to allow them to attack cancerous cells.
This experiment used a combination of unmethylated CG—enriched oligodeoxynucleotide (CpG) and anti-OX40 antibody (anti-OX40) in the treatment of cancerous mouse models.
OX40 is expressed on activated CD4 T cells that are involved in the tumor necrosis factor receptor apoptotic pathway in tumor-attacking effector T-cells. An anti-OX40 antibody would work, in a sense, as an agonist to stimulate activation of this process.
Toll-like receptor 9 ligands have been shown to promote expression of OX40. Using this logic, combining anti-OX40 and TLR9 should hypothetically work together in an additive or synergistic fashion. The first part of this experiment was to determine if CpG induces OX40 expression, and CpG was shown to successfully induce OX40 expression in CD4-T cells.
The goal of the experiment was to use the tumor antigen in mice as the source in designing a successful tumor reduction in-vivo by T-cells mediated by the immune-activating vaccine.
What did they do?
Researchers injected A20 B cell lymphoma, B16-F10 melanoma, CT26 colon carcinoma, or 4T1-Luc breast carcinoma at 2 different sites in the body of a mouse. One tumor was administered through injection with CpG only or the vaccine, and the other tumor was left alone (distant tumor). The tumor size was measured in both sites.
As transplanted tumors are not exactly representative of genetic or naturally occurring tumors, researchers also tested the vaccine on breast cancer. Using an oncogene driven by a mouse mammillary virus that promotes tumor growth, female mice developed breast cancer and were treated with the vaccine.
Along with the genetically prone breast cancer mice, researchers published the main article focusing on the injected B cell lymphoma line, as it had the most favorable results. The other injected tumors were published as supplemental data, which still produced favorable exploratory results.
B Cell Lymphoma
When mice were treated individually, both CpG and anti-OX40 had a delay of growth in the non-injected (distant) tumor, but only CpG had a complete regression in the local injection site (anti-OX40 caused a delay of progression). Together, the combination of CpG and anti-OX40 caused complete regression of both the injected and non-injected (distant) tumors. This complete regression was consistent for long-term survival, as the only mice that survived until the end of the study were the ones that received the combination treatment. There were 3 mice in the vaccine group that had a recurrence of a tumor in a distant site. However, all 3 were shown to be sensitive to the vaccine when injected directly into the relapsed tumor. Because of its mechanism of action of inducing expression of OX40, CpG worked best when it was directly administered into the tumor, while anti-OX40 had similar effects when given systemically at higher doses.
Genetically Breast Cancer-Prone
In this cancer in mice, metastases lead to tumor growth in the lungs as it progressed. Both the directly injected and non-injected (distant) tumors regressed in mice treated with the vaccine and had significant reduction of lung metastases. A key finding in this study was that after treatment, the mice had developed anti-tumor CD8 T cells to produce IFN-γ when exposed to tumor cells from within the body. This suggests the possibility that the mice developed an immune response against independently developing tumors, like an immune response to a foreign body.
To test if the mice could fight off a new tumor post-treatment without a new vaccine given, researchers injected new tumors into the treated mice. B-cell lymphoma tumor-cured mice were injected with either another B cell lymphoma or colon carcinoma. The B-cell lymphoma tumor had a similar effect as it did when treated, while the colon carcinoma progressed.
In order to confirm that the local injected tumor controls which type of tumor the immune system will fight throughout the body, B cell lymphoma tumors were injected into 2 sites, and colon carcinoma was injected into one site. The vaccine was injected into 1 lymphoma tumor and both lymphomas (distant and injected) regressed as the colon carcinoma progressed. When switching tumors to treat, the same effect was seen.
What does this mean?
Vaccination with the CpG and anti-OX40 combination directly into the tumor develops a tumor-specific response that is effective throughout the body for that tumor line.
Does this mean that it is only effective for 1 type of tumor treatment and then ineffective? Researchers mixed both tumors described above into 1 site and injected each tumor individually somewhere else with a 1:1:1 ratio of B-cell lymphoma, colon carcinoma, and combination tumor. When the vaccine was administered into the combination tumor, both distant tumors showed regression. This demonstrates the adaptability of the vaccine to immunize against a variety of tumors without prior knowledge on specifics of the tumor antigens.
Where is this treatment now?
Those who are interested in where these treatments are in development may be surprised to find that the individual components are in human clinical trials. Researchers used CpG named SD-101 (NCT02927964, NCT02266147, NCT01745354, NCT02254772, and NCT02521870) and anti-OX40 named MEDI6469 (NCT02559024, NCT01644968, NCT02221960, NCT02318394, NCT02274155, NCT01862900, NCT01303705, and NCT02205333) in this experiment.
The advantage of not being antigen-specific allows this vaccine to be a possible treatment for cancers for which unique antigen identifiers have not yet been discovered. One obvious challenge to a vaccine that requires direct injection into the tumor would be the accessibility of the tumor. Another concern would be patients who are immunosuppressed and their ability to develop a sufficient immune response or if it will have an effect on transplants. Common autoimmune responses and clot risks associated with antibody administration are always a concern, but local injections at low doses may be beneficial in curbing these adverse effects.
Because of the ability of the vaccine to induce “memory” of the tumor, could this be a vaccine that deters surgical measures for those more at risk of cancer (ie, BRCA1-positive mutation mastectomy)? Will future genetic testing of mutations be less important than they are when deciding initial treatment for certain populations? With continuing clinical trials on both agents in human testing, these questions and more may be addressed in the relatively near future.
Sagiv-barfi I, Czerwinski DK, Levy S, et al. Eradication of spontaneous malignancy by local immunotherapy. Sci Transl Med. 2018;10(426).