Although follicular lymphoma (FL) is the most common form of non-Hodgkin lymphoma, studying it has been a challenge. Isolating the malignant B-cell population from a surgical biopsy is a difficulty that has limited research, for example. But investigators from Stanford University and the University of California, San Francisco may have found a way to solve this problem with single-cell RNA sequencing (scRNA-seq). They used the technique to examine thousands of cells from primary FL tumors and successfully compared normal immune cell subpopulations and malignant B cells based on gene expression (Blood 2018; doi:10.1182/blood-2018-08-862292). Doing so gave them an unprecedented view of the tumor microenvironment—exactly what the field has needed to gain a better understanding of tumor-specific features such as immune checkpoint co-expression patterns.
“With this technology, we can really start detecting and understanding the underlying biology between these normal immune cells that are infiltrating the tumor and start looking at their potential interactions with the tumor cells,” said study author Hanlee P. Ji, PhD, Associate Professor of Medicine (Oncology) at Stanford University. “That's really the basis of all immunotherapy, so this is providing us with a very high-resolution picture of these interactions occurring within the tumor microenvironment that are absolutely critical for us to understand how immunotherapies work.”
Ji and his colleagues, including renowned FL researcher Ronald Levy, MD, isolated 34,188 cells derived from six primary FL tumors and identified individual normal B cells, malignant B cells, and the various malignant B-cell subclones that existed within each tumor. Given the inherent patient-to-patient variability of FL, the team chose to compare the gene expression profiles of normal and malignant B cells within each patient. Doing so avoided the drawbacks of physical isolation, as with fluorescence-activated cell sorting, and accounted for each patient's genetic background.
“We saw a vast diversity of different cell types,” Ji noted. “Even among the tumor B cells we saw variability indicative of different clonal populations. With this method, we can begin to see how tumor cells vary even within a given patient and even given an individual tumor site.”
The gene sequencing technique revealed that the malignant B cells had restricted immunoglobulin light chain expression (either Ig Kappa or Ig Lambda). While they noted the expected upregulation of the BCL2 gene, they also discovered a down-regulation of the FCER2, CD52, and MHC class II genes.
“There have been lots of studies published looking at transcriptional features of tumors, but they have all used methods that ground away the individual cellular features,” Ji explained. “But this type of study and our approach allowed us to see those interactions per individual cells and individual cell populations intact.”
The study also revealed novel candidate genes of T-cell regulation, which the researchers say are consistently co-expressed with known immune checkpoint molecules. They found certain genes co-expressed with the CEBPA and B2M in Tregs immune checkpoint molecules, and CD59 is co-expressed with both TNFRSF18 and TNFRSF4 in CD4+ memory cells. Because several immune checkpoint molecules, including CTLA-4 and PD-1, are current targets of immunotherapy, such findings go a long way to a better understanding of the gene networks involved in immune regulation, Ji said.
“The immune receptor aspect and their expression and the type of expression patterns we see from the T cells and the cancer cells is going to be a critical part of better understanding why immunotherapies work and, perhaps more importantly, why they don't work,” he added.
FL affects 1 in 5 lymphoma patients in the U.S. and is a relatively indolent condition, although it progresses into a highly aggressive disease at a rate of 2 percent per year. While many patients respond well to immunochemotherapy, the disease remains incurable and often follows a relapsing-remitting pattern (Epigenetics 2017; doi:10.1080/15592294.2017.1282587).
Early research is beginning to unravel some of the mystery surrounding the FL tumor microenvironment, leading investigators to believe certain gene mutations are responsible for aggressive subtypes of the condition. In addition, increased cytotoxic T cells and fewer myeloid cells may play a role in extended survival. “Thus, a more complete understanding of the diversity of the tumor cellular population and the immune microenvironment in early tumor evolution may reveal opportunities for intervention,” the authors wrote in the study.
Although RNA sequencing has been around for years, the technology wasn't always this advanced. In the past, researchers looked at 100 or 200 cells from multiple patient samples. But now Ji and his team sequenced thousands per individual tumor and “that provided us an advantage, having much higher resolution to define some of the heterogeneous cellular features that you would expect to see from a primary tumor biopsy, “ Ji said. And higher numbers provide greater sensitivity for defining the cellular features, he added.
Because this study was one of the earlier demonstrations of using single-cell sequencing to analyze primary tumor tissues from biopsies, Ji and his team needed a way to validate their findings—and using scRNA-seq on FL tumors specifically was the key. The team could use Levy's well-established FL flow cytometry assays, which are the gold standard for FL research, according to Ji.
“What we discovered from our single-cell sequencing we could practically validate efficiently using the well-vetted tumors with flow cytometry,” he explained.
It doesn't hurt that the process itself is relatively straightforward. Researchers can get primary tissues from patient biopsies and process them effectively to produce high-quality data.
Bolstered by such exciting findings, the team is already moving on from this pilot study. They plan to incorporate the technique into Levy's clinical trial testing the feasibility of an “in situ vaccination” that uses immune-enhancing agents designed to trigger a T-cell immune response locally that then attacks cancer throughout the body (Sci Transl Med 2018; doi:10.1126/scitranslmed.aan4488).
“We are doing the same type of analysis on patients before and after they get treated, Ji explained. “The big advantage of that is that we can see the types of changes that occur after these patients have received this vaccination strategy.” And at the resolution of individual cell populations, he added.
“You can understand how the tumor cells are or are not interacting with the immune cells and other components of the cellular microenvironement pre- and post-treatment and make those comparisons.”
Such technology is the next evolutionary step in personalized medicine, Ji believes, as it could lead to vast discoveries about the biological activity of any given immunotherapy. It may even allow clinicians to refine immunotherapy for each patient.
As much information as researchers have at their fingertips today, there is still a lot of unknown biology, he admitted. However, “this type of method provides an opportunity to understand that biology much more discreetly and capture more of the biological complexity. We can begin to see the fundamental alterations within biological pathways with these interventions, particularly when you are studying patients' tumors and the consequences of the therapy afterwards.”
Ji believes these new gene sequencing methods may prove revolutionary for patients in need of immunotherapy—welcome news for thousands of patients with FL facing the challenging treatment road ahead.
Rebecca Hepp is a contributing writer.