We've updated our Privacy Policy to make it clearer how we use your personal data. We use cookies to provide you with a better experience. You can read our Cookie Policy here.


Mark a Course for the Future – Biomarkers in Cancer Immunotherapy

Mark a Course for the Future – Biomarkers in Cancer Immunotherapy  content piece image
Listen with
Register for free to listen to this article
Thank you. Listen to this article using the player above.

Want to listen to this article for FREE?

Complete the form below to unlock access to ALL audio articles.

Read time: 6 minutes

Immunotherapy has reshaped clinical oncology. Immune checkpoint inhibitors, which enhance the immune system’s ability to target and destroy tumors, have enabled clinicians to reimagine what's possible for cancer patients. However, immunotherapies are not a silver bullet for all. It's critical to differentiate between those that will benefit from immunotherapy and non-responders, and biomarkers are the crucial lynchpin in this process. Biomarkers help categorize patients into actionable response groups, which can improve clinical outcomes as well as reduce costs in drug development and in the healthcare system overall. While investment in discovering novel biomarkers has been high, few novel actionable biomarkers have made it from bench to bedside in this class of therapies.


Here we explore how biomarkers are driving the immuno-oncology landscape. We dive into cutting-edge biomarker research and the challenges facing the field. Finally, we imagine the future of biomarkers in immuno-oncology and discuss the steps needed to get there. 


What are biomarkers?


Before we dive into discussing how biomarkers are shaping immuno-oncology, let's define them. Broadly, biomarkers are biological signals, often specific molecules, used to inform patient care. Biomarkers play multiple roles, including differentiating patients based on their ability to respond to treatment, monitoring a patient’s response to therapy, or managing toxicity and side effects. In immunotherapy specifically, biomarkers are often used to measure the degree of tumor inflammation and/or the functional status of the immune system, either directly through immune-cell infiltration and activation or indirectly via proxies like genomic mutational load. These measurements can be used to assess the ability of one’s immune system to fight cancer, and thus predict a patient's potential response to immunotherapy.


Targeting of the PD-1/L1 pathway is the primary immunotherapy strategy used today, but not all patients respond to this immunotherapy modality, driving interest in the discovery of biomarkers that can inform which patients are more or less likely to respond to treatment. Currently, there are three biomarkers that inform the use of these therapies: PD-L1MSI-H, and TMB. The binding between PD-1 on an immune cell and PD-L1 on a tumor cell can inhibit the ability of immune cells to recognize and attack the tumor. Thus, high PD-L1 expression in a tumor can indicate that the inhibition of this PD-1/PD-L1 interaction could be therapeutically effective (e.g., by restoring the ability of the immune system to recognize and attack the tumor). Both microsatellite instability (MSI) and tumor mutation burden (TMB) are measures of the mutational load of tumors.  Patients with high mutational loads (e.g., MSI-high and TMB-high) can have better responses to immunotherapy, presumably because of increased tumor neoantigen production, but predictive utility is highly variable within and across tumor types.

While these biomarkers are currently used to guide immunotherapy treatment today, it's important to note that no single biomarker is the perfect predictive ground truth. For example, many patients who express high levels of PD-L1 will not respond to checkpoint inhibition, and some patients who do not express PD-L1 will respond to immunotherapy; same goes for patients who are MSI-high and TMB-high. In some cancer types, these biomarkers have no predictive value at all. In order to optimize patient care and maximize the success of clinical trials, the search for more sensitive and specific biomarkers continues.


Snapshot of biomarkers in immuno-oncology clinical trials. Credit: DeciBio.

Challenges in the biomarker space


The dynamic interplay between a patient's immune system and the tumor is incredibly complex, making it difficult to distill the dynamic biology of cancer to individual, discrete, static biomarkers with a singular “cutoff”, or threshold for distinguishing between responders and non-responders. While single biomarkers are relatively inexpensive and easy to implement in the clinic, they fail to sufficiently recapitulate the complexity of the tumor-immune interaction to accurately predict patient response. For example, TMB, a measure of the number of mutations in a tumor, is a proxy for neoantigen expression, which is a proxy for immunogenicity, which is a proxy for tumor inflammation, which can inform which patients may or may not respond to immunotherapy. While a convenient, single metric, one can see that TMB, as a biomarker, is multiple degrees removed from the actual tumor-immune interactions that drive tumor pathology, which may be why it has failed to demonstrate strong clinical utility to date. To improve the ability to stratify patients, a new biomarker paradigm, one that provides deeper, more precise insight into the tumor microenvironment and pathological processes, is needed. This new paradigm of biomarker analysis should reflect the interplay of multiple physiological / pathological processes and account for the heterogeneity that exists within a tumor, all from increasingly limited sample.


While researchers have the ability to measure biomarkers in these ways today (more on that later), multiple barriers exist to implementing more sophisticated biomarker approaches in the clinic. Not surprisingly, increasing the plex, sensitivity and resolution of biomarker analysis increases the cost and complexity of testing and data analysis compared to single biomarker analysis. Many clinical labs, especially those in resource constrained regions, are not equipped or capable of implementing such complex and costly assays. Additionally, extracting more data from a tissue sample often requires more sample, which would put additional burden on patients. Lastly, conducting more complicated molecular analyses takes more time than single-biomarker analysis, which can delay patient treatment decisions. To be adopted into routine care, novel biomarkers must achieve a fine balance between clinical utility, cost-effectiveness and scalability / accessibility. Until such a balance is achieved, the key decision-makers in cancer care (e.g., payors, regulators and clinicians) may be apprehensive to “buy into” a new biomarker paradigm.


What is on the horizon? 


Despite the challenges associated with bringing more sophisticated biomarkers to the clinic, innovation charges ahead, with new tools and technologies enabling increasingly comprehensive molecular analysis. Three areas emerging as focal points of immuno-oncology biomarker research are spatial omics, single-cell omics, and multiomic biomarker analysis. Here we provide examples of each of these innovations and explore how shifting established paradigms could improve the biomarker space.


Let’s begin with spatial omics. Tumor-immune interactions occur in three-dimensional space, which the current biomarker paradigm fails to reflect. Today, most histological biomarkers are measured as the aggregate level of expression across a tissue sample; however, research suggests that where biomarkers are expressed within a tumor as well as in relation to other biomarkers, can be just as important as whether or how much of the biomarker is expressed. Spatial omics refers the incorporation of spatial or positional context into biomarker analysis. In the case of immunotherapy biomarker analysis, spatial context can provide deeper insight into how immune cells interact with a tumor, which can inform whether immunotherapy might be effective.  Excitingly, recent advancements in molecular analysis enable the analysis of proteins, DNA and RNA – alone or in combination – in a highly resolved spatial context, providing unprecedented ability to understand the tumor microenvironment, which may translate into novel clinical biomarkers.


Another, related example is single-cell analysis. Typically, for genomic biomarker analysis, nucleic acid material is extracted from tissue samples in bulk, without regard for which cells the nucleic acid markers came from. Cancer is both clonal and heterogenous in nature, thus, each cell behaves differently, and contributes in unique ways to the pathology of a tumor. Ignoring the differences in biomarker expression between individual cells can provide a misleading picture of the tumor. Today, there are numerous technologies which allow for the analysis of a wide variety of biomarkers (e.g., protein, DNA, RNA, epigenetic markers, etc.) from single cells, in some cases with spatial context as well. Single-cell spatial omics represents perhaps the richest source of molecular data that can be generated from cancer samples today. These single-cell data can be used to characterize the tumor microenvironment more comprehensively, allowing for the possibility of more accurate patient stratification.


Lastly, multiomics, the concept of integrating the analysis of multiple different types of analytes (e.g., DNA, proteins, RNA, epigenetic markers, etc.), is demonstrating potential to yield improved biomarkers. There are numerous interacting molecular and cellular pathways at play in cancer, and the drivers of pathogenesis can occur throughout the central dogma of biology (e.g., at the DNA, RNA or protein level). Thus, analyzing multiple markers and multiple types of markers can yield more insight into the myriad of molecular pathways involved in cancer. Various studies are underway assessing “composite” biomarker signatures (i.e., those combining two independent biomarkers into a single signature) and preliminary data suggests that the composite biomarker (in this case TMB + a gene expression signature) is better at stratifying patients than either marker individually. Outside of immunotherapies, multiomic approaches are also showing promise in early cancer detection and therapy selection for other cancer treatment modalities


A marker, but not an end


A cancer diagnosis is always devastating, but actionable biomarkers can be a silver lining for some patients, ensuring they receive an optimal treatment plan. In the world of precision medicine, improving the ability to identify which patients will respond to therapy is as important as developing new therapies altogether. With increasingly powerful tools at our disposal, and the determination and resolve of the oncology community, including patients, clinicians, payors, regulators and researchers, to continue to innovate, we are better equipped than ever in the battle against cancer.


About the authors:

Andrew Aijian, PhD, is a partner at DeciBio Consulting. Andrew specializes in the development, commercialization and utilization of research tools, diagnostics and digital technologies across the entire precision medicine spectrum, from early discovery through the patient journey. Andrew works to reduce the barriers to innovation between precision medicine stakeholders. 

Colin Enderlein, MS, is a senior project leader at DeciBio Consulting. Colin has been active in precision medicine innovation and CDx commercialization at multiple phases of the R&D value chain in both the public and private sectors. He specializes in research related to oncology therapies and their associated biomarkers, spanning both the pharmaceutical and genomic tools markets.