Read the Full Video Transcript

Andrea Miyahira: Hi, everyone. Thanks for joining us today. I'm Andrea Miyahira here at the Prostate Cancer Foundation. Today I'm joined by Dr. Daniela Brina, a postdoctoral researcher at the Institute of Oncology Research. She'll be discussing her group's recent paper, "The Akt mTOR and MNK/eIF4E Pathways Rewire the Prostate Cancer Translatome to Secrete HGF, SPP1, and BGN and Recruit Suppressive Myeloid Cells," published in Nature Cancer. Dr. Brina, thanks so much for joining us today.

Daniela Brina: Thanks a lot to you for the invitation. Our project starts from the observation that myeloid-derived suppressor cells are a relevant subset of immune cells infiltrating prostate cancer, in particular, PTEN-driven prostate cancer. Previous studies in the lab demonstrated that the elimination of myeloid-derived suppressor cells with a CXCR2 antagonist can ameliorate the progression of the disease. But still, a proportion of the myeloid cells remain in the tumor. The relevance of myeloid-derived suppressor cells is underlined by the fact that the neutrophil-to-lymphocyte ratio in prostate cancer, as well as in other types of cancer, is related to worse survival.

Our aim was to identify unknown recruiters of myeloid-derived suppressor cells. To do this, we investigated the extracellular interaction between prostate cancer and myeloid-derived suppressor cells. We wanted to pursue this aim by using a technique called polysome profiling because it allowed us not only to identify the gene expression profile of prostate cancer but also to identify which types of mRNA are particularly specifically translated by the cells. So we are able to isolate the ribosome-bound mRNA. We performed this technique in five different genetic backgrounds of prostate cancer that are characterized by the most frequent alterations found in human prostate cancer: PTEN deletion, p53 deletion, and the TMPRSS2:ERG fusion.

What emerged from this study is that we found 83 ligands that are regulating the tumor that are matched with the 60 receptors overexpressed in tumors and expressed by myeloid-derived suppressor cells. We finally identified 13 ligand-receptor couples that are listed here, and among them, we identify CXCL5, validating our finding. Among the most expressed ligand-receptor pairs that we identified, we focus on biglycan, osteopontin, and HGF that have a common feature. They are translationally upregulated in prostate cancer cells, meaning that the translation machinery has a greater affinity for this mRNA compared to others, meaning that this mRNA probably is needed for the fitness of the cancer cells.

We validated the upregulation of biglycan, osteopontin, and HGF in PTEN, p53 double-null tumors and the relevance of these three secreted factors in the context of the PTEN, p53 null mouse model. So while these secreted factors are not able to affect proliferation in vitro, they slow the tumor growth in vivo, and importantly, they reduce the myeloid-derived suppressor cells' infiltration into the tumor, associated with an increase of CD3+, CD8+ positive T cells.

When we look at the alteration, the molecular alteration in a prostate cancer cell, we found that the level of the regulator of the translation machinery PDCD4 is almost completely lost together with a strong increase in phosphorylation of eIF4e. This means that PDCD4 lost the capability to sequester eIF4e, which is a member of the eIF4F complex that is needed for the initiation of translation. MNK phosphorylates more eIF4e, and this phosphorylation renders eIF4e more active to translate a specific subset of mRNA.

In our context, we identified that, indeed, Phospho-eIF4e has greater affinity for biglycan, osteopontin, and hepatocyte growth factor compared to other upregulated secreted factors in prostate cancer. At the level of translation, if we block together the phosphorylation of eIF4e with the rescue of PDCD4, we obtain a strong decrease in the translation of the secreted factor.

So, we moved in vivo to identify if this translation inhibition overall affects the disease infiltration and tumor growth. To mimic the same effect that we obtain in vitro, we investigated which drug is able to rescue PDCD4 levels. We found that ipatasertib, which is currently tested in clinical trials for prostate cancer, is able to dose-dependently increase PDCD4 levels, rendering it an interesting drug to be tested in vivo. We combined ipatasertib treatment with eFT508 and we obtained an amelioration at the histopathological level of prostate adenocarcinoma with a less aggressive phenotype together with the reprogramming of the tumor-infiltrating immune cells. Indeed, the myeloid-derived suppressor cells are less infiltrating the tumor with the corresponding increase in the T cells that infiltrate the prostate glands, suggesting that this dual treatment can be effective to favor an antitumor immune response.

The second observation that we did looking at the different genetic backgrounds that could be relevant in human prostate cancer is that in the presence of the TMPRSS2:ERG fusion, there is an upregulation of Cxcl5 together with Cxcl3 and Cxcl2 that are all ligands of a CXCR2 receptor that is able to be blocked by the CXCR2 antagonist, suggesting that a combination with a translation inhibitor and the CXCR2 antagonist in this context can have a beneficial role.

We combine the translation inhibitor eFT508 with the CXCR2 antagonist AZD5069 and we obtained a less aggressive phenotype of the TMPRSS2:ERG, PTEN null tumors together with less infiltration by myeloid-derived suppressor cells and increased CD8 T cells in the prostate glands.

Summarizing, in PTEN null-driven prostate cancer, we obtain that ipatasertib can rescue PDCD4 levels, reducing the level of eIF4A that's available for the eIF4F complex formation. Together with the eFT508 treatment that inhibits Phospho-eIF4e, we can decrease the translation of the secreted factors that we identify, reducing the myeloid-derived suppressor cells infiltrating the tumor. The combination with the translation inhibitor can also be performed with the classical immunotherapy based on the CXCR2 antagonist that reduces the myeloid-derived suppressor cells in genetic contexts like PTEN, TMPRSS2:ERG fusion, and counteracts immune evasion.

Interestingly, biglycan, osteopontin, and HGF are increased in human prostate adenocarcinoma and even more in castration-resistant prostate cancer, so they are relevant for the pathogenesis of the disease. We identified that a patient who has high levels of the ligand together with high levels of Phospho-eIF4e has worse survival compared to patients with low levels of the ligand and low phosphorylation of eIF4e.

Another interesting observation is that PDCD4 is a poorly investigated oncosuppressor in prostate cancer. But still, its mRNA level is associated with worse disease-free survival in both the TCGA and the Taylor database. In our prostate tissue microarray at the level of protein, we determined that patients who have low PDCD4 levels and high expression of the ligand have worse survival probability compared to patients with high PDCD4 and low expression of the ligand, suggesting that PDCD4 is a relevant oncosuppressor in the context of prostate cancer.

Lastly, we find that biglycan, osteopontin, and HGF are positively correlated with CD33, which is a marker of myeloid-derived suppressor cells in the tumor.

We can conclude that we identified a signature of translationally-regulated secreted factors that are mediating the MDSC recruitment into the tumor, that the combined treatment with eTF508 and ipatasertib can dampen MDSC recruitment and tumor growth in PTEN null and p53 null prostate cancer, and that the conventional immunotherapy with the CXCR2 antagonist can be combined with the translation inhibitor in the context of the TMPRSS2:ERG prostate cancer. Lastly, biglycan, osteopontin, and HGF are overexpressed in adenocarcinoma and castration-resistant prostate cancer, underlining the relevance for the progression of the disease. With this, I thank you for your attention.

Andrea Miyahira: Thank you so much, Dr. Brina, for sharing. This is such an interesting study. So, are there higher numbers of infiltrating MDSCs in tumors with PTEN deficiencies?

Daniela Brina: The PTEN loss is correlated with high infiltration of myeloid-derived suppressor cells. This was demonstrated by a different group than our group, by the Pandolfi lab and Hong Wu lab. So the PTEN deletion correlates with the infiltration of myeloid-derived suppressor cells with a decrease of CD8 T cell infiltration.

And interestingly, there is a paper in Science Translational Medicine in which the reactivation of PTEN also leads to a decrease of myeloid cells and the reactivation of the antitumor immune response. So PTEN null deletion and downstream, the activation of the Akt/mTOR pathway is demonstrated to reprogram the tumor microenvironment and it leads to favor the increase of myeloid-derived suppressor cells. So the target at different levels of this pathway can indeed have a role in the reprogramming of the tumor immune response.

This observation is important since we know that PTEN, p53, and also ERG are the most frequent alterations seen in human prostate cancer. So if we know which specific target and the consequences, we can understand and find an effective therapeutic strategy.

Andrea Miyahira: Thank you. Do you have any insights as to why eIF4e, which is a general translation initiation factor, specifically increases HGF, SPP1, and BGN? And is this phenomenon specific to tumors with PTEN loss?

Daniela Brina: Okay, so we start from our RNA immunoprecipitation experiment in which we find that Phospho-eIF4e has an increased affinity for this mRNA. The exact mechanism by which Phospho-eIF4e selects specific mRNA is not yet understood. There are papers that suggest that Phospho-eIF4e boosts the translation of specific mRNA based on a specific sequence in the 5' material of this mRNA. The increase of phosphorylation of eIF4e is suggested to promote the recycling of the mRNA that allows translating more proteins, starting from the mRNA. So at the end, the result is the production of a high number of proteins, but the mechanism is not understood.

What we understand from our data is that when we look at the translationally regulated mRNA, we found that there is a common motif that we are currently investigating. So in the next step, based also on the manipulation of this specific sequence, we can understand if it's really involved in the Phospho-eIF4e mechanism of regulation. So hopefully, we will understand in the next months.

Andrea Miyahira: Okay, thank you. And then, does your team have translational plans for these research findings and are there treatments or targeting combinations that you think may be promising?

Daniela Brina: Yes, I think this combination that we propose could potentially be very interesting, so we are in contact with our collaborators in the hospital to try to understand which is the best setting in which to try to understand if this combination can be effective. And of course, I think it's fundamental to stratify patients for the specific alterations that we found, meaning PTEN loss, p53 loss, and also the TMPRSS2:ERG fusion, to identify the specific patients who can benefit more from this treatment and try to obtain also more data related to the correlation between PTEN MDSC infiltration, the neutrophil-to-lymphocyte alteration, to identify the best cohort of patients.

Andrea Miyahira: Thank you again. Thanks so much for coming on and sharing this with us today.

Daniela Brina: Thanks a lot, and to you.