T cells get a lot of attention in cancer immunotherapy, but dendritic cells are what drive T cell responses forward. Honing in on strategies that kickstart dendritic cells to induce strong antitumor T cell responses in otherwise cold tumors, Oba and Long et al. designed a rational, multi-step combination therapy called in situ immunomodulation (ISIM), which consists of sequential administration of Flt3L to recruit dendritic cells (DCs) to a tumor, followed by radiotherapy to trigger immunogenic cell death of cancer cells and induce DC maturation, and then TLR3/CD40 stimulation to stimulate antigen-loaded cDC1s to prime tumor-specific CD8⁺ T cells. The team recently tested this regimen in various mouse models, and the results were published in Nature Communications. 

To test the ability of ISIM to convert cold tumors, Oba and Long et al. used orthotopic, syngeneic mouse models of AT-3 breast cancer, 4T1 breast cancer, and B16 melanoma, all of which were poorly infiltrated by T cells, heavily infiltrated by TAMs, and refractory to anti-PD-L1 therapy. While individual components or incomplete combinations of the ISIM regimen were capable of delaying tumor growth and improving survival, the full ISIM regimen induced rapid tumor regression in all tumor models starting just one day after TLR3/CD40 stimulation, substantially extending survival in mice. In mice bearing contralateral tumors, treatment of the primary tumor induced an abscopal effect in the distant tumor.

Using the AT-3 model, the researchers showed that daily in situ administration of Flt3L effectively increased the mobilization of CD103⁺ DCs, enhancing their infiltration into tumors as well as their accumulation in lymph nodes. The addition of localized irradiation induced CD40 and CD86 expression on tumor-resident DCs, suggesting enhanced maturation of these cells. Finally, the inclusion of TLR3/CD40 agonists decreased CD103⁺ DCs in the tumor, but increased them in the tumor-draining lymph nodes, consistent with inducing migration. Using scRNAseq to evaluate immune infiltrates, the researchers identified a cluster of cDC1s expressing high levels of MHC-II-related genes, and another cluster, distinct from both cDC1s and cDC2s, that produced high levels of IL-12 and increased in frequency within tumors following ISIM. 

Looking at the T cell response following ISIM in AT-3 tumors, Oba and Long et al. identified increased CD8⁺ T cell infiltration and activation, but a decreased frequency of CD4⁺ T cells. Using the 4T1 model, they found that ISIM induced priming of tumor antigen-specific CD8⁺ T cells. While incomplete combinations of ISIM components increased central memory T cell subsets, only the full ISIM regimen generated effector memory T cell subsets and enhanced PD-1 expression on T cells – particularly on tumor antigen-specific T cells. Further, scRNA sequencing showed that ISIM induced a massive influx of stem-like progenitor-exhausted T cells expressing Tcf7 and Slamf6 into tumors. 

Picking apart the individual contributions of TLR3 agonism and CD40 agonism, the researchers found that use of the TLR3 agonist poly(I:C) increased central memory CD4⁺ and CD8⁺ T cells, while the agonist anti-CD40 mAb contributed to upregulation of T cell activation markers and T cell differentiation towards an effector memory phenotype. While both TLR3 and CD40 agonism contributed independently to immune remodeling and antitumor efficacy, their effects were enhanced in combination. 

Looking at the balance between T cells and immunosuppressive myeloid cells in the TME, the researchers found that the increase in CD8⁺ T cells induced by ISIM was associated with a decrease in CD11b⁺ myeloid cells (including macrophages consistent with an M2-like phenotype), reversing the CD8⁺/CD11b⁺ ratio. ISIM also increased PD-L1 expression on TAMs and tumor cells. Interestingly, both the reduction of TAMs and their increased expression of PD-L1 were found to be dependent on the presence of CD4⁺ T cells. 

Overall, the full immunological and therapeutic effects of ISIM were proven to be dependent on Batf3-dependent DCs, IL-12, IFNγ, CD8⁺ T cells, and CD4⁺ T cells. ISIM efficacy was also shown to be dependent on immune trafficking from secondary lymphoid organs, suggesting that the response is mediated by newly activated, rather than pre-existing T cell responses.

Given that ISIM appeared to influence PD-1 expression on T cells and PD-L1 expression on TAMs and tumor cells, Oba and Long et al. investigated whether ISIM might render tumors more responsive to PD-L1 blockade. Mice treated with anti-PD-L1 alone were not responsive to treatment; however, mice treated with ISIM plus anti-PD-L1 showed a significant delay in tumor growth and improved survival in all three tumor models. Interestingly, anti-PD-L1 therapy altered gene expression strongly in Tcf7⁺Slamf6⁺ T cells, but had little effect on terminally exhausted T cells. PD-L1 blockade also appeared to restore CD4⁺ T cell frequency, which had been reduced by ISIM, and skewed CD4⁺ T cells towards Th1-polarization. 

While the combination of ISIM and anti-PD-L1 was more effective than either treatment alone, it was still not sufficient to clear tumors, and all treated mice eventually relapsed and succumbed to their disease. Repeating individual components or partial combinations of the ISIM regimen could not overcome this resistance, however, treatment with the full ISIM regimen mediated a second wave of tumor regression and further prolonged survival. The researchers observed that while CD103⁺ cells returned to baseline levels after the initial ISIM treatment, a second round of ISIM treatment recruited new CD103⁺ DCs. Consistent with this, new TCR clonotypes arose with subsequent ISIM cycles. 

When Oba and Long et al. combined serial ISIM with PD-L1 blockade in the AT-3 model, they were able to induce complete responses in all treated mice after four cycles. Surviving mice rejected rechallenge in a tumor-specific manner, suggesting the development of immunological memory. Curative potential for serial ISIM plus PD-L1 blockade was also observed in the B16 model.

Previous human studies have shown that Flt3L, radiotherapy, and TLR3 agonist Poly(I:C) in combination are well tolerated and potentially effective. Similarly, systemic agonist anti-CD40 mAbs have been shown to be safe in humans, and are currently being tested in combinations with Flt3L or PD-1 axis blockade. This preclinical study by Oba and Long et al. provides a mechanistic basis that the comprehensive ISIM combination has the potential to be an effective immunotherapy that targets DCs to induce potent antitumor T cell responses and tumor regression.

by Lauren Hitchings

Meet the researcher

This week, first co-author Takaaki Oba and lead author Fumito Ito answered our questions.

What prompted you to tackle this research question?
TO: Although immune checkpoint inhibitors have revolutionized the treatment for various solid malignancies, their efficacy is generally lower against 'immune-cold’ tumors such as breast cancer and pancreatic cancer. I am a surgeon for breast cancer in Japan, and came to Roswell Park, looking for a novel immunotherapeutic approach for such 'cold’ tumors. We became interested in conventional type 1 dendritic cells (cDC1), which excel at cross-presentation of tumor antigens, and strived to develop an effective combinatorial regimen using cDC1.   

FI: Poor T cell infiltration in the tumor correlates with worse prognosis and poor response to cancer treatment such as anti-PD-1/PD-L1 therapy. One major limitation of T cell exclusion from the tumor is the inefficient priming of tumor-specific CD8⁺ T cells. We were looking for the strategy to improve this process so that we could initiate the cycle of cancer immunity. We have been interested in an ‘in situ’ approach where we could generate polyclonal tumor-specific T cells against heterogeneous solid tumors without the need for identification of patient- and tumor-specific antigens. Radiation therapy has been known to trigger immunogenic cell death, and enhance antigen processing and cross-presentation. However, this process is insufficient in the suppressive tumor microenvironment. To overcome this limitation, I came up with the combinatorial in situ immunomodulation (ISIM) strategy to increase the immunogenicity of radiation therapy by maximizing the engagement of conventional type 1 dendritic cells (cDC1).

What was the most surprising finding of this study for you?
TO: When we analyzed tumor-draining lymph nodes (TdLN) in mice bearing GFP-expressing tumors using Imaging Cytometry, we were surprised to see more CD8⁺ T cell-cDC1 doublets containing GFP in mice treated with in situ immunomodulation (ISIM). One of the doublets had GFP on the surface of cDC1, as if they presented antigen to the CD8⁺ T cells, which was cool.  We were also surprised to see the remodeling of the TCR repertoire in the tumor by repeating ISIM, and that we were able to overcome not only primary resistance, but also acquired resistance of anti-PD-L1 therapy.    

FI: I expected that we would be able to overcome primary resistance to anti-PD-L1 therapy by the priming, expansion, and infiltration of tumor-specific CD8⁺ T cells. However, I thought it would be difficult to overcome acquired resistance and eradicate tumors because there are many mechanisms underlying acquired resistance to anti-PD-L1 therapy. Therefore, we were surprised to see the additional regression of the tumor when we repeated ISIM. We were glad to see serial ISIM induced the continuous remodeling of TCR repertoires in the tumor, which was the most likely driving force that contributed to overcoming acquired resistance to anti-PD-L1 therapy in our studies. 

What was the coolest thing you’ve learned (about) recently outside of work? 
TO: I worked at Roswell Park Comprehensive Cancer Center for two years and now got back to Japan. I have spent a really precious time with my family in the US. What I am surprised at is how our kids grow up in the US. Our kids can speak English more fluently than I do. They also learned a lot of things in the US, which are priceless experiences for us.  

FI: I like visiting historic landmarks. I had a chance to visit many national parks in the summer of 2018, and found Glacier National Park had many great hiking trails. This time of the year reminds me of the beautiful color change in Kyoto. I am also interested in contemporary art. When I attend national/international meetings, one of my favorite things to do is to visit contemporary museums of art.