Blog
Nov 8, 2022 / Oncology

Multiomic single cell profiling reveals T-cell chromatin state primes cancer immunotherapy response

Jeanene Swanson

Making cancer immunotherapy more effective is a primary goal for researchers in the immuno-oncology space—and that takes not only understanding what makes a therapy effective, but also the factors underlying therapeutic resistance. In this publication summary, we feature a publication that makes use of Chromium Single Cell Multiome ATAC + Gene Expression, which combines single cell gene expression and chromatin accessibility mapping, to delve into patient response to PD-1 checkpoint blockade immunotherapy at the single cell level (1). What the scientists discovered has implications for both improving treatment response and using single cell analysis to identify, based on pre-existing physiological cell states, which patients may best respond to a specific therapy.

In the past decade, immunotherapy for treating cancer has taken off, becoming standard for an increasing number of tumor types, from melanoma to lung cancer, bladder cancer, lymphoma, and others (2). Instead of attacking the tumor itself, immunotherapy drugs help a person’s own immune system target cancer cells by activating tumor cell–killing T cells. Immune checkpoint inhibitors act to block the checkpoint proteins that reduce T-cell overactivity, thereby allowing T cells to be free to seek and destroy cancer cells. One type of checkpoint therapy is an antibody that targets the checkpoint protein, PD-1, by binding to it and inhibiting its activity.

While checkpoint inhibitors have proven effective for some cancers, they don’t work for everyone or in all types of cancer. Glioblastoma, pancreatic, and prostate cancers are particularly resistant to checkpoint blockade therapy (2). Scientists believe that decreased anti-tumor T-cell activity and a suppressive immune cell tumor microenvironment play a large role in anti-tumor immunity.

Type I interferons (IFN-Is)—cytokines released by several types of immune cells—are key components of anti-tumor immunity and positive immunotherapy response. However, chronic IFN-I signaling can lead to expression of inhibitory factors, including PD-1, which can result in a tumor escaping therapy (3). In this publication summary, we feature work led by Giselle M. Boukhaled, PhD, at the Princess Margaret Cancer Center in Toronto, who discovered that patient responsiveness to IFN-Is before treatment with PD-1 checkpoint therapy was predictive of long-term survival. Surprisingly, they found that people who were less responsive to IFN-Is survived longer whereas those who were more responsive experienced treatment failure and the progression of their cancer.

Creating response scores by measuring proteins

IFN-Is trigger the expression of hundreds of genes. However, instead of just measuring whether these genes are turned on or off by IFN-Is, the team wanted to see if single cell protein expression patterns of IFN-I reactivity would offer insight into the cells that respond to therapy and how their responses affect the outcome. To that end, in the first part of their study, Dr. Boukhaled and her team developed a single cell mass cytometry approach to measure IFN-I-stimulated protein (ISP) expression across the human immune system.

After developing a cytometry by time-of-flight (CyTOF) panel for detecting 13 ISPs in healthy donor peripheral blood mononuclear cells (PBMCs) in vitro with IFN-β, they followed by probing PBMCs from three cohorts of patients treated with anti-PD1 immunotherapy (pembrolizumab or nivolumab). The discovery cohort consisted of clinical trial patients with both head and neck squamous cell carcinoma or melanoma, while the two validation cohorts comprised patients that had either cutaneous melanoma or non-small cell lung cancer.

Based on the average signal intensity of the 13 ISPs, they were able to compute an IFN-I score for each peripheral blood immune cell subtype, reflecting an interferon-induced cell state; a higher IFN-I response capacity (IRC) would mean that more ISPs are upregulated and to a greater extent. Intriguingly, they found that patients who progressed after treatment had higher IRC before therapy in clusters of CD4+ and CD8+ T cells. Similar to the discovery cohort, a low pre-therapy IRC in CD4+ and CD8+ effector T-cell (Teff) populations in both validation cohorts was associated with longer overall survival.

Multiomic single cell analyses unveil a scarred chromatin cell state affecting responsiveness

To better understand the differences in IFN-I responsiveness, the team performed single cell RNA sequencing (scRNA-seq) and single cell assay for transposase-accessible chromatin (scATAC-seq) using Single Cell Multiome ATAC + Gene Expression on two healthy control and eight pre-therapy PBMC samples from the melanoma validation cohort. In focusing on the CD4+ Teff cell population as in the CyTOF analysis, they revealed that patients’ cells had substantial differences in gene expression and chromatin accessibility compared to healthy donor CD4+ Teff cells. Looking into these differences by way of gene set enrichment analysis, they found that CD4+ Teff cells with high IRC were enriched in pathways associated with restricted response to PD-1 checkpoint immunotherapy.

Employing SCENIC (4), which pairs transcription factors with their regulons, to understand the transcriptional networks behind the IFN-I responsive states, they discovered that epigenetic modification pathways were some of the most upregulated networks in the high IRC CD4+ Teff cells. These pathway genes code for many enzymes involved in transcriptional repression, activation, cell-fate specification, and chromatin stability, suggesting that fixed epigenetic changes are what cause increased IFN-I responsiveness. Further analysis showed that the high and low groups actually lacked differential interferon-stimulated gene expression, meaning they did not respond differently to IFN-Is before therapy. The authors suggested that, “Responsiveness to IFN-Is is an epigenetic scar of previous encounters that differentially poises them to be hyper- or hypo-responders.”

Pre-determined cell states help predict checkpoint therapy efficacy

By relying on single cell tools to probe both gene expression and chromatin changes, Dr. Boukhaled and team determined that IFN-I resistance among certain subsets of immune cells before treatment can predict long-term survival, and that high responsiveness is actually associated with a worse response to PD-1 checkpoint immunotherapy. Further work is needed to understand how to better apply this knowledge to direct treatment regimens of anti-PD1 immunotherapy in cancer patients.

To learn more about the power of multiomic profiling for deeper single cell characterization, visit our website.

References:

  1. Boukhaled GM, et al. Pre-encoded responsiveness to type I interferon in the peripheral immune system defines outcome of PD1 blockade therapy. Nat Immunol 23: 1273–1283 (2022).
  2. https://www.mdanderson.org/cancerwise/why-doesnt-immunotherapy-work-for-everyone.h00-159385101.html
  3. Boukhaled GM, et al. Opposing roles of type I interferons in cancer immunity. Annu Rev Pathol 16: 167–198 (2021).
  4. Aibar S, et al. SCENIC: single-cell regulatory network inference and clustering. Nat Methods 14: 1083–1086 (2017).