What is a biomarker?
A biomarker is a measurable biological factor that can tell us about certain aspects of human health. In the context of cancer immunotherapy, biomarkers can provide insights into each patient’s individual cancer—its genetic makeup, its behavior, and its interactions with the immune system—which doctors can then use to determine the approach most likely to benefit a particular person.
What types of biomarkers are there?
Biomarkers come in diverse forms. There are several categories of biomarkers that doctors may use to answer important questions both before and after a patient has begun treatment:
- Diagnostic: What type of cancer is it?
- Prognostic: What is the expected outlook?
- Predictive: How likely is it that this treatment will work? Are side effects likely to occur?
Biomarkers During and After Treatment
- Short-Term Monitoring: Is the treatment working? Have any side effects arisen?
- Extended Monitoring: Is the cancer still stable or in remission?
Together, these different types of biomarkers can help guide doctors’ decision-making regarding the best course of action for a particular person in order to increase the likeliness that treatment will maximize survival and quality of life.
How are biomarkers measured?
Biomarkers come in many forms, which can be measured in various ways depending on the specific biomarker. Some can be measured through scans, whereas others may require a biopsy (surgical removal of tumor tissue). While tumor samples provide incomparable insight into tumor-immune dynamics, important biomarker information can also be acquired from more accessible places. Blood provides an especially rich source of biomarkers relating to the activity of both immune and cancer cells, and recent advances have demonstrated that saliva, breath, urine, and stool can also reveal important biomarker information.
Biomarkers in the clinic
Already, several biomarkers have been incorporated (to varying degrees) into clinical practice, while others’ value has only recently been recognized and is still being explored. Some of the most promising biomarkers today in the context of immunotherapy include:
Cancer cells (and other cells within tumors) that express the PD-L1 protein can shut down “killer” T cells (via the PD-1 receptor) and prevent them from eliminating tumors. A study by the CRI-SU2C Dream Team was one of the first to reveal an association between PD-L1 expression and responses to PD-1/PD-L1 checkpoint immunotherapy in patients with several types of cancer. As a result, the level of PD-L1 within tumors has been used as a biomarker to predict which patients might benefit from these immunotherapies. While patients whose tumors have high expression of PD-L1 are typically much more likely to respond, patients whose tumors don’t express any PD-L1 can still respond to these immunotherapies, too, due to the dynamic nature of PD-L1 expression. In other words, just because a patient’s tumor isn’t expressing PD-L1 at one point in time, doesn’t mean that it won’t “turn on” PD-L1 expression later to protect itself from killer T cells.
Pre-Existing Immune Responses
The presence of CD8+ “killer” T cells within and around tumors—a biomarker sometimes referred to as the Immunoscore—has been associated with improved outcomes in cancer patients, regardless of what treatment they receive. Tumors infiltrated by killer T cells often also express the PD-L1 protein to protect themselves from immune attack, making patients whose tumors have these biomarkers more likely to benefit from checkpoint immunotherapies targeting the PD-1/PD-L1 pathway. Additionally, a higher diversity of T cells (as determined by the T cell receptor repertoire) and the presence of cell-killing (cytotoxic) gene signatures have also been associated with positive patient responses. Combining these killer T cell-related biomarkers with the PD-1/PD-L1 biomarker provides even greater predictive power, as shown by the CRI-SU2C Dream Team. While detecting the presence of killer T cells within tumors has traditionally required a potentially invasive biopsy, recent work by a CRI-funded postdoctoral fellow used a technique that may allow doctors to obtain this information non-invasively.
Genetic mutations are a hallmark of malignant tumors and are responsible for the vast majority of cancer’s life-threatening characteristics, such as ceaseless growth and metastasis, or spreading within the body. These mutations can also lead to the production of mutated proteins that make tumors “stand out” and provide targets—known as neoantigens—that the immune system can use to attack cancer cells. The number of mutations that a tumor has accumulated is itself a biomarker, referred to as tumor mutational burden (TMB). The more mutations a tumor has, the more mutated proteins and neoantigens it’s likely to express for the immune system to target. Consequently, patients whose cancers have high TMB have been found to be much more likely to respond to checkpoint immunotherapy. High TMB is also associated with pre-existing immune responses and expression of PD-1/PD-L1. Additional work by the CRI-SU2C Dream Team showed that a tumor’s genetic status also plays a role in patient responses to anti-CTLA-4 checkpoint immunotherapy.
Some tumors lose the ability to repair their DNA, which leads to extremely mutated tumors characterized by high microsatellite stability (MSI-hi) / deficient mismatch repair (dMMR). Patients with these MSI-hi/dMMR tumors are especially likely to benefit from PD-1/PD-L1 checkpoint immunotherapy, in part because they are often infiltrated by killer T cells and also express PD-L1. In May 2017, an anti-PD-1 immunotherapy became the only treatment of any type to be approved for patients with MSI-hi/dMMR solid tumors, regardless of what type of cancer it is.
Individual mutations also have great potential as biomarkers in cancer immunotherapy. In two CRI-funded studies, some tumor mutations were unfortunately shown to lead to primary as well as acquired resistance to immunotherapy. However, other mutations can serve as the basis for personalized immunotherapies, such as vaccines that can enable tumor-targeting immune responses specific to a patient’s cancer. These personalized vaccines are now being evaluated in clinical trials in several cancer types.
In addition to the neoantigens that arise from mutations, tumor gene expression often becomes dysregulated in a way that causes them to produce normal proteins in abnormal ways. These too can be used to target cancer cells through immunotherapy. One example is HER2, a growth-related protein found on healthy cells that is often expressed at abnormally high levels on cancer cells. Already, several targeted immunotherapies have been approved to treat patients with different types of HER2+ cancers. Another example is NY-ESO-1, a protein that is normally only expressed in fetal cells and adult testicular cells. Sometimes cancer cells can inappropriately turn back on NY-ESO-1 production. In ovarian cancer, NY-ESO-1 expression is associated with more aggressive disease. Fortunately, a CRI-funded study showed that a vaccine targeting NY-ESO-1 was associated with improved survival in patients with this type of aggressive ovarian cancer. Lastly, tumors infected by viruses such as HPV (human papilloma virus) and EBV (Epstein-Barr virus) can cause cancer cells to express abnormal viral proteins that can provide targets for the immune system and immunotherapy.
The microbiome is the collection of bacteria, viruses, and other microorganisms that reside in our bodies (mainly in the gut) as well as on our skin. Scientists have recently recognized biomarkers associated with the tens of billions of bacteria found in the gut. These bacteria interact with the immune system and, as a result, can have a significant impact on the effectiveness of PD-1/PD-L1 checkpoint immunotherapy, a connection first revealed in mice in work involving a CRI-funded postdoctoral fellow. More recently, this connection was shown to exist in humans, too, as it was found that patients with more diverse gut bacteria were more likely to respond to immunotherapy. Certain types of bacteria have also been associated with improved responses after immunotherapy. Alternatively, one study found that patients who were treated with antibiotics, which can lower bacterial diversity among other consequences, had worse overall survival after being treated with immunotherapy.
Biomarkers Moving Forward
Ultimately, there is unlikely to be a single, perfect biomarker that will apply in all cases because the importance of various biomarkers is likely to depend on what type of cancer a patient has as well as what type of immunotherapy he or she receives. Some biomarkers may only be relevant in certain tumor types or with certain immunotherapies, and in many instances their importance will have to be determined in a case-by-case basis. In the future, panels of multiple biomarkers will likely be developed to provide doctors with the most comprehensive and actionable insight into their patients’ cancers and immune systems, which will enable patients to be treated in the ways most likely to help them.
In addition to the above biomarkers, the incorporation of other biomarkers will also help to significantly improve their potential use in the clinic. Discovering and validating new biomarkers remains an extremely active area of investigation, especially with respect to: other types of immune cells, the complexity of tumor-immune interactions, and the steps involved in the process of the immune system launching an adaptive response against tumors. To that end, advances in biomarker development will likely be tied to technological advances that enable more in-depth investigations into the myriad factors that govern the relationship between cancer and the immune system. New tools could also enable biomarker information to be obtained less invasively than traditional methods. Clinical trials will also play a crucial role in this advancement, as they provide a setting for doctors to seek answers to the most important questions regarding how to make immunotherapy more effective.
Kunle Odunsi, M.D., Ph.D., of Roswell Park Comprehensive Cancer Center discusses how biomarkers are already being used to help patients today as well as how we’re working to harness the full potential of biomarkers moving forward
Haidong Dong, M.D., Ph.D., of the Mayo Clinic presents the role biomarkers play in making treatment decisions and delves into the latest research to identify new biomarkers in a CRI webinar in August 2017.