Biomarkers – Essentials of the Immunotherapy Toolbox

Biomarkers are the indicators which reveal molecular signatures of diseased state. We discuss their relevance in modern immune-oncology, performance and development over the years.

By Ajay Vitthal Patil, PhD

In order to avoid stressful situations becoming irreversible catastrophe, psychologist Gary Klein suggested to adopt a method called, ‘the premortem’[1]. It means, imagining such situation before it happens, to identify parts which could potentially fail, and put a system in place to avoid it. It is possible to identify these failing parts only if we have reliable indicators. These indicators, decide the reversibility or at worst, the severity of any situation. In any chronic disease like cancer, biomarkers are those indicators which hint at the scheduled pathology.

Articles of interpretation – Biomarkers

Clinicians have been using biomarkers long since the dawn of medical science. But with advances in technology they could pick more specific and sensitive biomarkers for each disease. For instance, in early days, oncologists staged cancer looking at CT scans, lymph nodes or bone marrow sections; today, molecular signatures of various gene panels provide better insight about the systemic health of the cancer patient. In the last decade, cancer therapy has gone through some fundamental changes; immunotherapy showed great results for certain cancers. And, rigorous research is being carried out to adopt these results for more severe solid tumors. However, correct classification of patient response to these therapies requires precise biomarkers. Henceforth, it will be absolutely important to have clear understanding of biomarker profile of the patient to make the best treatment decision among available therapies.

Related Article: Fighting Lung Cancer with Precision Warfare

Challenges in developing robust biomarkers for immunotherapy

Inherent heterogeneity of tumor microenvironment among patients poses a major challenge in identifying consensus biomarker for clinical use. ‘Expression analysis of important surface proteins on tumor cells’ (e.g. PD-L1 expression) and ‘somatic mutation load analysis’ (e.g. tumor mutation burden – TMB score) are two prime areas of biomarker analysis for immunotherapy currently. Biomarkers identified from these concepts complemented well during the development of early immunotherapy treatments. But, comparative studies analyzing their performance, across different trials revealed their limitations for the clinical adoption. Mainly these biomarkers are used for binary classification of patients into eligible and non-eligible populations,but they are not well equipped for the selection of therapy among various immunotherapy options [2]. However, inthelast few years, transcriptome analysis (RNA sequencing), multiplex immunofluorescence assays, and in-vivo pathology technologies, have greatly accelerated the pace of biomarker research [3]. Also, companies like Merck & Co./Nanostring and Foundation Medicine Inc. Roche (FMI-Roche) are now progressing to develop panel-biomarker testing services for precision therapy as well as clear business model to address this need.

Current biomarkers for immunotherapy

Tumor PD-L1 expression

The good

1. Provides platform for initial checkpoint drugs and widely used for anti-PD-L1/PD1 immunotherapies
2. Direct and comprehensive application in certain cancer (e.g. melanoma, NSCLC)
3. Relatively optimized clinical setup for testing the expression

The bad

1. Its dynamic expression on both tumor and immune cells makes it difficultto identify tumor specificity.
2. Lack of standardization across PD-L1–detection tests – Different companies use different staining methods, antibodies and scoring criteria, challenges during comparative analysis.
3. Heterogeneity in terms of response – In some cases higher PD-L1 expression doesnot translate into better response due to varying factors.

Alternatives: CD8⁺T cells, CD45⁺memory cells, Granzyme B – activated T cells, interferon gamma expression, clonality of T cell response within tumor (T cell receptor sequencing) [4,5]

Tumor Mutation Burden

The good

1. Strong correlation with patient response in certain cancers.
2. Advanced sequencing technology will help to rapidly adapt with new findings.
3. TMB has been positively correlated with neoantigen expression.
4. TMB can be used for broader range of immunotherapies (e.g. checkpoint blockade, vaccines, combination therapies)

The bad

1. Not all the mutations are equally immunogenic.
2. Lack of standardization across – Number of genes covered in current testing services (e.g. gene panel sequencing) may lose out on some mutations.

Alternatives: Whole exome sequencing – covering all the mutations which could be potentially lost in current composition of gene panels.

Microsatellite instability (MSI)

The good

1. Considered as one of the root cause which induces increased TMB, increased neoantigen expression and consequently tumor infiltration of T cells
2. Gives information about mismatch repair (MMR) failure and useful for diverse therapies
3. Outperforms other biomarkers in certain cancers (e.g. colorectal cancer)

The bad

3. Not all the mutations are equally immunogenic.
4. Lack of standardization across – Number of genes covered in current testing services (e.g. gene panel sequencing) may lose out on some mutations.

Alternatives: Immunescore assay – recently reported for better performance than MSI high in colorectal cancer

New biomarkers in development

Immunescore

Immunoscore is a score developed to measure CD3-positive and CD8-positive cell densities in the tumour centre and invasive margin. Immunoscore shown better performance in predicting time to tumor relapse and overall survival than the established tumor-node-metastasis staging system for colon cancer. It has a larger relative prognostic value than lymphovascular invasion, tumour differentiation, and MSI status in colorectal cancer [3,6].

T cell receptor clonality

Tumor infiltrating lymphocytes target specific antigens of tumor microenvironment through T cell receptors. Actual regions in contact with these antigenic epitopes are called ‘complementarity determining regions (CDRs)’. Each arm of TCR has three CDRs and CDR3 amino acid sequence primarily decides the specificity of the TCR and in turn the T cell. Modern sequencing technology allows us to identify these tumor specific CDR3 sequences and hence the overall clonality of the T cell population in the tumor environment. TCR clonality is probably the most reliable biomarker in pipeline as it has the potential to reveal both the therapeutic potential and the possible toxicity [6].

Ideal future

Prognostic and predictive biomarkers will decide the pace and efficacy of evolving immunotherapy drugs. And, with our growing understanding of underlying mechanisms of these therapies, it will be easier to identify, design and utilize most efficient biomarkers for every cancer. Interestingly recent findings from ‘molecular pathogenic epidemiology’ (MPE) research have suggested to consider some other factors like microbiome. Intratumour microbiota reportedly affects responsiveness to therapy in more than one cancers. Effects of fiber foods, omega-3 polyunsaturated fatty acids on tumour–immune interactions and microbiome have been reported. Such interdisciplinary research will provide tools for personalized analysis of cancer epidemiology in future [7,8]. Till then, clinical utility and cost-effectiveness of proven biomarkers will take the front seat of biomarker business.

References

1. Klein, HBR, 2007.
2. Morrison et al. Journal for ImmunoTherapy of Cancer, 2018.
3. Ogino & Giannakis., Lancet, 2018
4. Padmasee Sharma talk on OncoLive, 2016
5. Spencer et al., ASCO Annual Meeting, 2016
6. Voong et al., Ann Transl Med, 2017
7. Mehta et al., JAMA Oncol, 2017
8. Song et al., JAMA Oncol 2016

Author

GeneOnline