What if we could repurpose the flu vaccine into a cancer immunotherapy?
I was browsing Reddit this morning and came across this article and my first thought was THIS IS COOL. Thus, this blog entry was born. Below is a recap of the results from this manuscript:
Newman JH, Chesson CB, Herzog NL, Bommareddy PK, Aspromonte SM, Pepe R, et al. Intratumoral injection of the seasonal flu shot converts immunologically cold tumors to hot and serves as an immunotherapy for cancer. Proc Natl Acad Sci. 2019; 201904022. doi:10.1073/pnas.1904022116
To begin, there are conventionally two types of tumor immune environments: hot and cold. Hot tumors are those that have a lot of immune cell infiltrate (lung and melanoma are two examples of hot tumors). Cold tumors are those that have low immune cell infiltrate (prostate and pancreatic cancer are two examples of cold tumors). Immune cells that are present in cold tumors have suppressive regulatory phenotypes. Hot tumors respond well to immunotherapy because the immune cells are already within the environment to activate antitumor responses. In this manuscript, Newman et al investigate whether intratumoral injection of the FDA approved flu vaccine can ‘convert’ cold tumors into hot tumors to make them more permissive to immunotherapies.
In one of their first experiments, C57BL/6 mice were intravenously injected with B16-F10 (mouse melanoma cells). Intravenous injection of these cells results in localization within the lung and subsequent development of melanoma nodules within the lung. There were a total of four groups within this experiment (simplified shown below):
(1) B16-F10 (no treatment)
(2) B16-F10 + intranasal flu vaccine
(3) B16-F10 + anti-PD-1
(4) B16-F10 + intranasal flu vaccine + anti-PD1
They found that group 4 had a significantly lower number of lung nodules than the other three groups, and that administration of the intranasal flu vaccine alone also reduced the number of lung nodules compared to controls.
Next, using the Surveillance, Epidemiology, and End Results (SEER)-medicare linked database, they extracted data from 30,000 different lung cancer patients to assess if infection with the influenza virus (1 or more hospitalizations) had any impact on mortality. They found that patients that had at least one flu infection had lower mortality rates than those that did not have a flu infection.
Next, the researchers asked the same question as above to determine if administration of an intratumoral flu vaccine within melanoma tumors would reduce tumor growth. They found that melanoma growth was not reduced, as what was observed with the lung tumor model. They hypothesized that the lack of efficacy was because the skin lacks the molecular components required for flu virus infection, and thus the virus could be cleared faster than in the lungs. This reduces the amount of time it takes to clear the infection, and subsequently reduces the development of anti-tumors responses. Additionally, because the virus is cleared, inflammatory pathways (TLR engagement) are not as strongly activated.
To overcome this barrier, they instead used a heat inactivated flu vaccine (hiFLU). The heat inactivated vaccine exhibited strong TLR7 activation (response to ssRNA). Mice that received intratumoral administration of hiFLU within melanoma tumors exhibited reduced tumor growth, longer survival, and an increase in tumor specific CD8+ cells.
An additional experiment also showed that the antitumor effects of the hiFLU were systemic. Mice were inoculated in two spots (right and left flank) with melanoma tumors. One flank was treated with hiFLU, while the other was not. They found that there was a reduction in size for both tumors, despite the injection only occurring in one.
Next, they utilized a AIR-PDX (autologous immune-reconstituted patient derived xenograft) mouse model to test the FDA approved 2017-2018 flu vaccine in vivo. AIR-PDX mice contain tumor and peripheral blood mononuclear cells from the same patient (no HLA mismatch). Each mouse in the experiment had tumor and immune cells from a human patient.The experiment consisted of the following groups:
(1) no treatment
(4) FluVx + anti-PD-L1
Again, they found similar results from the first experiment. Mice that received both FluVx and anti-PD-L1 had reduced tumor growth and fewer distant metastases.
Some further experiments go into detail about the cellular phenotypes of the cells at play in the anti-tumor response. Overall, this seems like an exciting field of research, and potentially an easier therapeutic to bring to the clinics (compared to a novel drug).
I’ll put my two cents in here about pig models… The fact that the flu vaccine is already FDA approved, and not an experimental drug, shows promise and ease of testing moving forward. As with any new therapeutic regimen being tested, it would be beneficial to be tested within an alternative animal model. The Oncopig is an emerging cancer model in pigs. These pigs have an inducible p53 and KRAS mutations. Both p53 and KRAS are floxed and the localized injection of AdCRE (adenovirus carrying CRE) results in a mutation in each, and ultimately a tumor forms at the site of AdCRE injection. Different tumor types are still being investigated. However, this porcine model could be interesting to use to test the flu vaccine as an immunotherapy.
Thanks for reading and Happy New Year!
And of course, check out these references:
1. Newman JH, Chesson CB, Herzog NL, Bommareddy PK, Aspromonte SM, Pepe R, et al. Intratumoral injection of the seasonal flu shot converts immunologically cold tumors to hot and serves as an immunotherapy for cancer. Proc Natl Acad Sci. 2019; 201904022. doi:10.1073/pnas.1904022116
2. Schook LB, Collares T V, Hu W, Liang Y, Rodrigues FM, Rund LA, et al. A Genetic Porcine Model of Cancer. PLoS One. 2015;10: e0128864. Available: https://doi.org/10.1371/journal.pone.0128864