When we consider the future of cancer treatment, chemotherapy pales in comparison to immunotherapy. Unlike chemotherapy, which stops or blocks rapidly growing cells and can lead to adverse side effects, immunotherapy modifies a patient’s own cells to enhance or restore the cancer-fighting ability of the immune system. Immunotherapy has led to a marked improvement in clinical outcomes, but unfortunately only a subset of patients respond positively to this treatment. It is well known that different tumors have diverse genetic and epigenetic alterations that contribute to the range of phenotypes, and ultimately the success of treatment. However, the field lacks a full understanding of how tumor heterogeneity confers sensitivity or resistance to immunotherapy. The degree of T cell infiltration into the tumor microenvironment (TME) has been proven to be a critical factor in clinical outcomes. T cells induce apoptosis in cells that display foreign antigens, such as virus-infected or tumor cells. The TME is comprised of extracellular matrix and local stromal and immune cells, a combination that is likely regulated by signaling pathways intrinsic to tumor cells.
The ability to investigate the regulation of distinct immune landscapes in tumors has been constrained by a lack of appropriate experimental systems that faithfully recapitulate the heterogeneity of the TME. Enter the laboratory of Dr. Ben Stanger. His team, including Cancer Biology graduate student Jinyang Li, examined primary tumors from a mouse model of pancreatic ductal adenocarcinoma (PDA) and saw wide-ranging variation in the number of intratumoral T cells, similar to what was seen in human PDA tumors. Li hypothesized that whatever was dictating this variation might be retained in these tumor cells if transplanted into healthy wild type mouse.
In order to test their hypothesis, the team generated a new experimental system to investigate tumor heterogeneity and interrogate the mechanisms behind the correlation of PDA patient survival and abundance of T cell infiltration. They created a library of congenic tumor cells clones by isolating late-stage primary pancreatic tumors from a fully backcrossed cohort of mice. Upon implanting these tumor cell clones into wild type mice, it was clear that the library replicated the TME heterogeneity seen in humans and mice with PDA. The implantation experiments also corroborated findings that high T cell infiltration is consistent with tumor regression and improved survival upon administration of modified T cells in a combination immunotherapy regimen. Interestingly, the group observed a greater correlation between the abundance of PD-1+CD8+ cytotoxic T cells in tumors and a positive response to therapy compared to total CD8+ T cell infiltration alone. PD-1 is a cell surface molecule that regulates the adaptive immune response. Its biological significance to the immune system makes it a potential therapeutic target. However, in this context, PD-1 appears to be a biomarker for predicting immunotherapy efficacy.
To determine how different PDA clones establish T cell levels in TMEs, Li injected an equal mixture of T cell high and T cell low clones into a single site on a mouse. He observed that the composition of the TME resembled that of a T cell low tumor, with increased myeloid cell infiltration as compared to a T cell high clone. Li postulated that T cell low clones exert a dominant effect by which myeloid cells suppress T cell infiltration to the TME, and thus the anti-tumor T cell response. RNA-seq analysis of sorted tumor cells revealed differential expression of chemokine Cxcl1 between T cell high and low tumors. Cxcl1 encodes a ligand for CXCR2, which is required for myeloid cell infiltration into TMEs. Additionally, Li observed that the Cxcl1 promoter region is more accessible in T cell low versus high tumor cells. He also found that reducing MYC protein, a known promoter of immunosuppressive TMEs, in T cell low clones decreases the expression of Cxcl1 in these cells. The reduction of Cxcl1 expression in T cell low clones resulted in cell survival and proliferation comparable to that of T cell high clones. Together, these results suggest that chromatin accessibility and MYC activity dictate expression of Cxcl1, and therefore T cell levels in TMEs of different PDA clones.
Li wanted to know if CXCL1 is the major driver of immune cell heterogeneity in the TME. He engineered T cell high clones to overexpress Cxcl1, and transplanted them into wild type mice. This resulted in increased myeloid infiltration and decreased T cell infiltration. Conversely, injecting T cell low clones into mice with a deletion of Cxcr2 resulted in a decrease in intratumoral myeloid cells and an increase of PD-1+CD8+ T cells. Therefore, tumor cell-derived CXCL1 is necessary and sufficient for recruiting myeloid cells into TMEs, consequently suppressing T cell infiltration and the anti-tumor response.
PDA tumors display unique T cell (red) infiltration that dictates immunotherapy sensitivity. The dominant tumor (green) phenotype is a non-T-cell-inflammed microenvironment, which mobilizes myeloid cells (yellow) rather than cross-presenting dendritic cells (blue) that promote recruitment of T cells. T cell low tumors are distinguished by differing epigenetic marks and transcriptomic status as compared to T cell-inflamed TMEs. An increase in tumor cell-derived chemokine CXCL1 is responsible for therapy-resistant in low T cells tumors.Ablation of CXCL1 can promote T cell infiltration and sensitivity to a combination immunotherapy regimen.
PDA tumors display unique T cell (red) infiltration that dictates immunotherapy sensitivity. The dominant tumor (green) phenotype is a non-T-cell-inflammed microenvironment, which mobilizes myeloid cells (yellow) rather than cross-presenting dendritic cells (blue) that promote recruitment of T cells. T cell low tumors are distinguished by differing epigenetic marks and transcriptomic status as compared to T cell-inflamed TMEs. An increase in tumor cell-derived chemokine CXCL1 is responsible for therapy-resistant in low T cells tumors. Ablation of CXCL1 can promote T cell infiltration and sensitivity to a combination immunotherapy regimen.
Li and his team were able to elucidate the major genomic, epigenetic, and transcriptomic features driving the formation of low T cell TMEs and resistance to immunotherapy. The biomarkers identified in this study expand our ability to predict immunotherapy efficacy in the clinic. Additionally, Li showed that interfering with expression of Cxcl1 and Cxcr2 could convert a low T cell TME into a high T cell TME capable of generating a robust and protective anti-tumor response. Of course, it’s unlikely that a singular molecular mechanism is solely responsible for the TME heterogeneity in PDA, but the identification of even this one pathway pinpoints possible therapeutic targets and offers clues to other contributing mechanisms. Importantly, this work would not have been possible without the library of congenic tumor cell clones that reproduced the spectrum of TMEs observed in patients. This platform will certainly be useful as we make greater strides toward optimized patient-specific immunotherapy.
Li, J, Byrne, KT, Yan, F, Yamazoe, T, Chen, Z, Immunity, B.-T., 2018. Tumor cell-intrinsic factors underlie heterogeneity of immune cell infiltration and response to immunotherapy. Immunity.