Immuno-oncology (IO) focuses on harnessing the tremendous power of the human immune system to detect and destroy cancer cells. The specificity (with virtually infinite antigen recognition), the adaptability (due to genetic and epigenetic changes), and the long-term memory (that can prevent cancer recurrence) make the immune system, at least theoretically, the ideal and ultimate anti-cancer treatment.
After a first wave of hope and disappointments in the late ’80s/early ’90s, the development of cancer immunotherapy has now reached an important inflection point in the history of cancer treatment. In fact, our increased/improved understanding of cellular and molecular tumor immunology has enabled the identification of novel and more effective ways to manipulate the immune response to cancer, resulting in an explosion of new immunotherapeutic approaches now in development, as single agents and in combination with each other and/or with traditional cytotoxic or targeted therapies.
However, despite recent clinical success of checkpoint inhibitors (such as inhibitors of the PD-1/PD-L1 pathway and CTLA-4 antagonists) as single agents in some solid tumors, outside of melanoma and a subset of Hodgkin’s Disease (where tumor-specific biology is at play), the vast majority of patients with metastatic solid tumors do not respond to checkpoint inhibitors. Combinations of checkpoint inhibitors with non-redundant mechanisms of action, although potentially more effective (at least in melanoma), have shown to be also more toxic and prohibitively expensive (Larken).
Therefore, in order to achieve optimal development and clinical use of the different immunotherapeutic approaches, we require integrated biomarker support for both mechanistic understanding of the drugs and patient selection.
Due to the complexity of the interaction between immune system and tumor biology, it is unlikely that the traditional biomarker approach (successfully used for targeted therapies such as tyrosine kinases inhibitors) based on a single analyte measurement would be very informative in immuno-oncology. This is the case even when the analyte is the putative target of the drug, as recently confirmed by the experience with the PD-L1 IHC test(s), which have shown in multiple clinical settings to be suboptimal tools for guiding the administration of anti-PD-1/PD-L1 inhibitors (Ung and Kockx).
Three potential sources of inter-patient heterogeneity have the potential to influence the interaction between cancer and immune system:
- Host: e.g., Germline Single Nucleotide Polymorphisms (SNPs) in immuno-regulatory genes
- Tumor: e.g., somatic mutations and epigenetic changes in tumor cells
- Environment: e.g., gut microbiota
Each of these sources of heterogeneity is currently clinically measurable; therefore, a simultaneous, integrated analysis encompassing inborn genetic traits, somatic genetic and epigenetic alterations, and environmental contributions may be possible. This would allow us to explore and identify rate-limiting step(s) in the immune/cancer interaction that may need to be selectively targeted in a patient-specific fashion (i.e., Precision Immuno-oncology).
Analytes that can be measured in the tumor and/or host include RNA (in the form of gene expression profiling of the tumor and/or isolated immune cells), DNA [(tumor molecular alterations and diversity of host T-cell Receptor (TCR regions)], and proteins (including their post-translational modifications). To date, all of these analytes have been explored independently rather than in an integrated manner (different studies, using different platforms, by different investigators, using different therapeutic approaches).
It is our belief that the next generation of biomarkers in IO needs to measure and integrate the complexity of host, tumor, and environment. This requires multi-omics measurement (measuring DNA, RNA, and proteins simultaneously) with the same sample (where possible, thus maximizing the amount and type of information obtainable from each clinical sample) and the same units of measurement (to facilitate the data integration).
The ability to measure any combination of DNA, RNA, and protein from the same sample simultaneously on a single system is what we at NanoString define as 3D Biology™ and what we believe provides the level of granularity needed for the development of the next generation of highly-predictive biomarkers in IO. This level of granularity is now achievable on the NanoString platform.
Above: NanoString 3D Biology enables development of multi-analyte signatures: information content of 3D tests increases exponentially with each data dimension.
- Larken, J. “Combined Nivolumab and Ipilimumab or Monotherapy in Untreated Melanoma.” New England Journal of Medicine (2015): 373(1):23-34. Epub.
- Ung, Christopher and Mark Kockx. “Challenges & Perspectives of Immunotherapy Biomarkers & The HistoOncoImmune Methodology.” Expert Review of Precision Medicine and Drug Development (2016): 9-24. Epub.