Since the dawn of the 20th century it was thought that the immune system could ultimately play a role in the treatment of cancers (Wirth and Kunhel, 2017). There has been a resurgence of interest in the last decades in the relationship between the immune system and the development of cancer, and as a consequence, the potential for immune therapy. These activities culminated in the first FDA approvals of a number of immune checkpoint inhibitory antibody drugs including ipilimumab (2011), nivolumab (2014) and pembrolizumab (2014) to treat cancer (Marshall and Djamgoz, 2018). The increased interest in and focus on immunotherapy reached a public crescendo with the 2018 Nobel Prize for Physiology or Medicine being awarded to James P. Allison and Tasuku Honjo for their discovery of cancer therapy by inhibition of negative immune regulation, the basic science behind immune checkpoint biology and its resulting potential in oncology. While it seems that currently no detailed immune therapy is the long-sought after, broadly-acting ‘magic bullet’ yet, there has been tremendous progress across various indications. Going forward the potential for further harnessing the power of the immune system through, for example, combination therapies engaging other critical molecular targets to treat cancer is significant (Chung-Han et al., 2018).
In this edition of a two-part serial blog we will attempt to first characterize neoantigens and then highlight some interesting facets of the interaction between neoantigens and immune biology. In the second part of the series of blogs (next month) we will summarize the current state of immunotherapies in cancer. We will also describe some interesting future directions of both single-agent and combination immunotherapy, as well as combinations of non-immune and immunotherapies in cancer treatment.
What are neoantigens and why do they matter?
Accumulation of mutations in cancer is the source of neoantigens and their variation
In order for the immune system to modulate the progression of cancer/tumors it is necessary for the tumor cells in question to be or to become immunologically distinct, but how does this happen?
It has been shown, that the progression of cells along the path to malignancy is linked to the accumulation of various mutations in their genomes. Though the full nature of the transition is still not completely understood, at some point along this progression axis precancerous cells acquire one or more additional mutations that help transform into a malignant phenotype. At some further point down the line it is likely for the tumor/cancer to acquire additional mutations triggering metastatic potential. The exact number of somatic mutations acquired, and the timing of their occurrence can vary widely among tumor types (Vogelstein et al., 2017). The totality of mutations in a particular cancer is commonly referred to as tumor mutational burden (TMB).
It is not a hyperbole to say that tumorigenesis is incredibly complex and involves the acquisition of capabilities (the so-called ‘Hallmarks of cancer’ reviewed in Hanahan and Weinberg ) by cells that give them selective advantage which – as described above – is linked to the accumulation of mutations. Inherent in this process is a whole spectrum of interactions and feedback loops between precancerous/cancer cells and normal cells. It is beyond the scope of this discussion here to delve into greater depth on this topic – we can highly recommend to consult the foundational review by Hanahan and Weinberg (2011).
Classification of neoantigens and related antigens
Once a cell has acquired a new mutation, and in order for it to be immunologically relevant the mutation must occur in a protein-coding gene. The mutations that lead to expression of detectable antigens are roughly divided into the following groups (Wirth and Kunhel, 2017);
- Tumor-specific antigens
The group of antigens thought to be most broadly useful in diagnosis and treatment of cancer are termed tumor-specific antigens (TSA) or sometimes tumor-specific mutant antigens (TMSA). These novel antigens arise from mutations in the DNA of cancer cells where the mutation is located within the coding sequence of an expressed protein and where the coding sequence change leads to the incorporation of a different amino acid at one or more positions in a protein. Since these neoantigens are not present in the proteins encoded by the normal genomes, they can in principle be uniquely targeted by the immune system as they are not subject to thymic selection and tolerance.
- Driver mutations versus passenger mutations
Some neoantigens are linked to protein changes where the mutant product confers a selective advantage to the growth of the cell. These are termed driver mutations and the genes they derive from as driver genes. These driver mutations and genes are relatively few in numbers versus the larger sum of mutations present in cancer cell. The remaining mutations are termed passenger mutations as they confer no selective advantage to the cells that contain them (Vogelstein et al., 2017).C
- Cancer testis antigens
Some of the earliest work to identify antigens present on cancer cells identified a series of molecules that are expressed in tumor context but also can be detected in testis tissue in normal context. These were termed cancer testis antigens (CTA). Since germ cells lack some required antigen presentation machinery, peptides derived from CTA can still be modeled as tumor-specific.
- Tumor-associated antigens
Related to CTA are antigens that are more broadly expressed in normal cell/tissue context but overexpressed in cancer cells. As a group these are termed tumor-associated antigens (TAA). TAAs are thought to generally be less useful in immunotherapy context due to potential side-effects such as autoimmunity and conversely low-avidity T-cell populations due to operation of normal tolerance mechanisms.
Image credits: Kakimi et. al, 2017.
Various antigen creating processes and presentation
Regardless of how a neoantigen is created the route to an immune reaction is not guaranteed. All peptides that are destined to be recognized by the T-cell based immune system must be processed via one of two major routes and presented to appropriate T-cells. Sequences that fail to be presented in one of these two fashions will not generate a T-cell response. As a matter of fact, quite a large fraction of the total neoantigen pool fails to elicit a response (Schumacher and Scrieber, 2015)
One system processes antigens by proteasome digestion: The resulting peptides are then handled by the transporter system associated with antigen processing (TAP) via the endoplasmic reticulum. Here the resultant peptides are loaded on major histocompatibility (also known as human leukocyte antigen (HLA)) complex I (MHC I) molecules and displayed on the cell surface. Subsequently the peptide:MHC I complex is presented to CD8+ T-cells (Aurisicchio et al., 2018).
In the second system peptides are proteolytically processed following autophagy: The resultant peptides are then complexed with MHC class II molecules. These complexes are normally presented by antigen presenting cells (APC) but other cell types are also capable of performing this task. These peptide:MHC II complexes are normally presented to CD4+ T-cells (Nogueira et al., 2018).
Whichever route of presentation predominates in a given case, researchers have explored ways to predict beforehand which neoantigens might be most relevant and most useful in immunotherapy. Various strategies leveraging RNA-seq, whole exome sequencing, diagnostic mass spectroscopy as well as predictive algorithms for binding of peptides to MHC I and II have been applied. To date, most success has been achieved in predicting MHC I binding in the context of the CD8+ T-cell presentation route. Prediction algorithms for MHC II binding and the CD4+ route are currently not as effective (Karasaki et al, 2017; Nogueira et al., 2018). In the long run, better understanding and predictability of both routes will likely be required as there are reports of anti-tumor effects proceeding via both MHC I and MHC II dependent mechanisms (Conway et al., 2018).
Mechanisms that alter neoantigen effects
There are several other key concepts that influence how neoantigens are sensed by the immune system and/or how they relate to effectiveness of immune therapies.
The TMB concept touched upon earlier has been used as a clinical biomarker for some time. Simply stated more mutations means potentially greater numbers of neoantigens and thus more opportunity for an immune response. Certain types of tumors with a relatively high rate of somatic mutations are the ones that respond best to certain types of immune therapy (reviewed in Conway et al., 2018) especially those that promote a broader attack on the target cancer as opposed to those targeting a single antigen.
Another facet that affects neoantigen biology is the concept of intra-tumor heterogeneity (TH or ITH). TH means that not all cancer cells for a given patient are guaranteed to have identical genomes and identical collection of mutations. Thus, cancers with higher TH have the potential to escape any particular therapy via suppression (e.g. immunoediting) of some clonal lines and outgrowth of others that lack the responsiveness to the particular therapy. So, in principle, low TH may correlate with higher treatment success or the opposite in cases of high TH (Aurisicchio et al., 2018, Conway et al., 2018).
The tumor microenvironment
The last facet affecting neoantigen biology in the context of for this discussion is the property termed the tumor microenvironment (TME). While strictly speaking not a neoantigen property, the complex cellular environment of a tumor can lead to suppression of a significant immune response despite the presence of antigens. This can be in the form of a physical barrier in the tumor whereby tumor stroma cells or extra-cellular structures prevent access by T-cells. Alternatively, the tumor can recruit immunosuppressive regulatory cells into the tumor environment or otherwise create conditions that prevent immune cells from mounting a successful response (Wirth and Kunhel, 2017).
Image credits: Hanahan and Weinberg (2011).
Assuming an appropriate neoantigen(s) exists in the case of a particular type of cancer or tumor, immunotherapy seeks to take advantage of them to treat the disease in the clinic. Various methodologies to do this have been successfully developed and many more are currently being tested in clinical trials. More details of these strategies will be described in the upcoming second part of this series that focuses on how neoantigens can be leveraged in immunotherapy.
Companies developing immune therapies with a particular emphasis on neoantigens
The scientific space around
neoantigens, immune therapy, and oncology is quite crowded, but a number of
companies are front runners in developing promising approaches that are based
on neoantigens – following below are just a few examples:
Genentech – which already has a significant immuno-oncology presence – announced in January 2019 a partnership with Adaptive Biotechnologies to develop a new type of cell therapy to target a patient’s neoantigens. Adaptive Biotechnologies, a privately held company founded in 2009 and headquartered in Seattle (WA), has a set of immunosequencing tools to allow selection of the T-cells receptors that bind specific neoantigens and in collaboration with Genentech will use this to focus on neoantigens shared between cancers in multiple individuals. Resulting specific T-cell therapies would then be expected to be very effective and useful in small, defined, groups of patients.
Gritstone Oncology, headquartered in Emeryville (CA) was founded in 2015 and trades with Nasdaq symbol GRTS. The company is focused on developing tumor-specific immunotherapies. EDGE™, a system for the identification of neoantigens and prediction of those that will prove useful in immunotherapy context, is one of their core technologies. To date, the company has brought forward two leading candidates: SLATE-001 which is intended to target neoantigens shared among a group of patients, while the second one (GRANITE-001) is intended to be a personalized immunotherapy regimen and has recently (Dec 2018) been granted fast track status in treatment of colorectal cancer. The company raised $216 million in total disclosed funding.
Moderna Therapeuticswas founded in 2010, is headquartered in Cambridge (MA) and trades with the Nasdaq symbol MRNA. The company is focused on the use of mRNA molecules as therapeutic agents in diverse areas including cancer, infectious disease and genetic disease. Within their oncology pipeline, Moderna has therapeutics to target changes to the TME and several products to act as neoantigen-driven cancer vaccines. They currently have clinical trials underway to evaluate their personalized cancer vaccines, as well as a study aimed at shared neoantigens in KRAS protein in various cancers. Moderna has partnered with Merck to create mRNA-based personalized cancer vaccines with Keytruda.
Neon Therapeutics was founded in 2013 and is headquartered in Cambridge (MA). One of Neon founders includes 2018 Nobel Prize winner James P Allison as well as several other founders with notable long publication histories in immunology. Neon has two products: NEO-PV-01, a personalized neoantigen vaccine, and NEO-PTC-01, a neoantigen T-cell therapy. Both are currently in clinical trial. In addition, in a 2017 Nature paper (Ott et al.) the authors, including some company co-founders, demonstrated an initially promising response to personalized neoantigen vaccine for melanoma. The company’s RECON™ bioinformatics engine has been designed to select the most relevant therapeutically active neoantigens. The company raised $161 million in total disclosed funding.
Aurisicchio et al., The perfect personalized cancer therapy: cancer vaccines against neoantigens. (2018) J Exp Clin Cancer Res, 37:86
Chung-Han et al., Update on tumor neoantigens and their utility: why it is good to be different. (2018) Trends Immuno, 39(7):536-548.
Conway et al., Genomics of response to immune checkpoint therapies for cancer: implications for precision medicine. (2018) Genome Med, 10(1):93.
Hanahan and Weinberg. Hallmarks of Cancer: the next generation. (2011) Cell, 144(5):646-74.
Kakimi et al., Advances in personalized cancer immunotherapy. (2017), Breast Cancer, 24 (1):16–24.
Karasaki et al., Prediction and prioritization of neoantigens: integration of RNA sequencing data with whole-exome sequencing. (2017) Cancer Sci, 108(2):170-177.
Marshall and Djamgoz. Immuno-oncology: emerging targets and combination therapies. (2018) Front Oncol, 8:315.
Nogueira et al., Improving cancer immunotherapies through empirical neoantigen selection. (2018) Trends Cancer, 4(2):97-100.
Ott et al., An immunogenic personal neoantigen vaccine for patients with melanoma. (2017) Nature, 547(7662):217-221.
Schumacher and Schrieber. Neoantigens in cancer immunotherapy. (2015) Science, 348(6230):69-74.
Vogelstein et al., Cancer genome landscapes. (2013) Science, 339(6127):1546-58.
Wirth and Kuhnel. Neoantigen targeting – dawn of a new era in cancer immunotherapy? (2017) Front Immunol, 19;8:1848.