Neoantigens are mutations in the genome of a tumor that result in mutant proteins that are not found in normal cells, but that are instead unique to that tumor. The isolation, identification, and selection of neoantigens has drawn extensive attention from the field of cancer immunotherapy. Recent data from laboratory research and clinical applications have indicated that neoantigens, which arise as a result of tumor-specific mutations, may serve as biomarkers in T cell-based immunotherapies. Indeed, while clinical application of patient-specific cancer immunotherapy integrating neoantigen-induced immunoresponses is still in its infancy, the identification and integration of neoantigen-derived data in both preclinical research as well as guided reference for clinical treatment has been widely applied in recent times .
Here, we present a summary of the challenges as well as opportunities that may be encountered during the application of neoantigens by the pharmaceutical industry, physicians, and investors to gain a better insight into the decision-making process.
What Are Neoantigens?
The application of endogenous T cell-mediated tumoricidal potential was initially demonstrated in metastatic melanoma and subsequently verified in a series of different human malignancies. In brief, endogenous T cells are able to recognize two classes of peptide epitopes on the surface of malignant cells displayed as part of the Major Histocompatibility Complexes (MHCs). We will focus on the second class, which originates from sequences that are absent from the human genome and, in most cases, created by tumor-specific DNA alterations whose novel translation products are known as neoantigens. Although neoantigens can also derive from viral Open Reading Frames (ORFs) in virus-associated tumors, these varieties are usually not patient-specific and will not be included in this discussion.
Due to the randomness of mutations in each unique tumor genome, a significant fraction of these novel translation products are not shared among patients. They are therefore considered patient-specific.
Identification of Neoantigens by DNA Sequencing and Computational Modeling
Since human malignancies bear a high mutation load as part of their nature (usually 1–10 mutations per megabase of exome i.e. the part of the genome formed by exons), deep exome sequencing and personalized tumor exome profiling become highly effective in predicting potential candidate ORFs of neoantigens. High-throughput genetic analysis has been integrated into a routine for identifying the candidate pool of possible neoantigens. Indeed, cancer exome sequencing has a meager false-negative rate when used to confirm the neoantigens that had been previously identified by other methods. However, most mutations found within the cancer exomes do not contribute to T cell-recognizable neoantigens, for two main reasons. First, not all mutant protein can be successfully processed and presented as a mutant peptide by MHC molecules. Second, even with successful presentation, the mutant-MHC complex may not be recognizable by T cells. Hence, robust and efficient filtering methods for data derived from cancer exome sequencing results are essential.
Recently, significant efforts have been devoted to the development and optimization of in silico algorithms integrating factors such as the likelihood of proteasomal processing, transportability into the endoplasmic reticulum, the affinity for the relevant MHC classes, and the gene expression level for epitope abundances. However, as mentioned before, precautions should be taken when evaluating the efficiency and output of such in silico models, since the validation of algorithm efficiency usually takes peptides that have been identified as neoantigens using alternative means as input, and the sequencing in silico processing approach has an inherent low false-negative rate.
Validation of Neoantigen Predictions by Mass Spectrometry
In consideration of the above factors, there is an urgency to develop methods that could experimentally validate whether the in silico predicted neoantigen epitopes are successfully presented by MHC molecules. Mass spectrometry, a method that has been widely used for detecting the presence of specific peptide fragments in mixed samples, provides an opportunity to validate the successful presentation of predicted neoantigens by MHCs. Indeed, two studies at preclinical stages have demonstrated the feasibility of using mass spectrometry to identify the MHC presentability of a handful of predicted neoantigens.
Moreover, the output of validation data could be reapplied to further optimize neoantigen prediction algorithms. Recently, the advancement of mass spectrometry technologies has allowed unprecedented sensitivity in detecting trace amounts of neoantigen presence in biopsy-level samples. However, very few resources have been invested in exploiting this vacuum, most likely due to the significant discontinuity between clinical immunology and industrial/diagnostic mass spectrometry, especially within the field of personalized medicine.
A Missing Market Opportunity
Therefore, in silico modeling of neoantigen prediction alone may not suffice for efficient and personalized cancer immunotherapy. In the meantime, a promising and scientifically demonstrated market remains neglected only due to the lack of integration. Specifically, very few marketing efforts have focused on the integration of clinic diagnostic-certified mass spectrometry analysis services, as well as the development of the associated clinical sampling/transportation/result analysis pipeline, into the current clinical practice of cancer immunotherapy.
In summary, accurate prediction as well as identification of tumor-derived neoantigens would contribute to the most challenging yet promising part of potential neoantigen-derived cancer immunotherapies. The individual-specific nature of neoantigen profiling makes it an ideal candidate to add into the application pool of personalized medicine. In addition, as noted above, the challenge stems out of a logistic vacuum rather than difficulties or unknowns in technology and/or engineering, and thus clinical applications may be on the horizon. For example, Omixon, a leading provider of next generation sequencing services for human leukocyte antigen (HLA) genotyping, has reported that neoantigen-derived cancer immunotherapies could soon become the most dominant driver of the HLA market. Clinical applications of neoantigens could significantly improve availability of patient-specific treatments, thus enabling further development of personalized cancer therapy.
