Rectal Cancer Splicing Analysis Explores FAM135A as an mRNA Vaccine Antigen Candidate and Immune Subtypes
Background: Finding antigen candidates as a starting point for vaccine development
An analysis of public rectal cancer genomic data proposed FAM135A as a candidate antigen for messenger RNA (mRNA) vaccines. The researchers narrowed the candidates by examining abnormal splicing alongside gene mutations and explored subtypes based on tumors’ immune characteristics. The findings require caution: this was a computational study using retrospective data. It did not produce a vaccine or test its effects in patients.
Finding targets that allow the immune system to distinguish tumors is a central challenge in cancer vaccine development. Even after a candidate is identified, separate studies must establish whether it induces an immune response and functions in the tumor’s immune environment. This study focused on identifying antigen candidates and classifying the immune environment.
The researchers examined alternative splicing (AS), in which exon combinations change during RNA processing. This is distinct from editing the DNA itself. The study analyzed genes associated with abnormal splicing alongside genes with frameshift mutations. It did not establish that splicing caused those DNA mutations.
Key findings: A focus on FAM135A among 217 candidates
The researchers compared tumor and normal samples from The Cancer Genome Atlas rectal adenocarcinoma dataset (TCGA-READ). Among AS types, exon skipping was the most common, while mutually exclusive exons were the rarest. They counted 4,480 AS events that increased and 3,328 that decreased in tumor samples. These figures refer to splicing events, not genes.
The intersection of differentially expressed genes, genes associated with abnormal splicing, and genes with frameshift mutations yielded 217 potential antigen candidate genes. The researchers then examined associations with survival and immune-cell infiltration, among other factors, and proposed FAM135A as a major candidate for follow-up validation. FAM135A expression was positively correlated with estimated infiltration of B cells, macrophages, and myeloid dendritic cells. These associations alone do not demonstrate actual presentation of the candidate antigen or T-cell activation.
Analysis of immune-related genes distinguished two subtypes, C1 and C2. The groups differed in survival and immune characteristics, and a survival difference between the subtypes was also reported in the independent public dataset GSE87211. This supports an association between the classification and prognosis, which should be distinguished from validation of vaccine response.
Implications and outlook: Hypotheses to test rather than demonstrated treatment effects
The researchers also estimated the potential for drug response by comparing gene-expression patterns in rectal cancer with treatment data from 47 melanoma patients who received immune checkpoint inhibitors. In the analysis, C2 appeared more likely than C1 to respond to drugs that inhibit the checkpoint proteins CTLA-4 and PD-1. However, this prediction was based on applying data from another cancer type. It should not be interpreted as an observed response rate in rectal cancer patients treated with those drugs or a vaccine.
Some interpretations are inconsistent within the paper. The descriptions of immune activity in C1 and C2, and the direction of the association between FAM135A expression and prognosis, are inconsistent between the results and discussion. This makes it difficult to conclude that a particular subtype is better suited to vaccination or that FAM135A expression levels can be used to determine prognosis.
The main limitation is that the immunogenicity and antitumor efficacy of peptides derived from FAM135A were not experimentally validated. Selection bias in retrospective data should also be considered. The study offers an antigen to test and a patient-classification hypothesis to assess in prospective research; demonstrating vaccine safety and efficacy remains future work.
The study methods and results are available in the provided full paper.
BACKGROUND: mRNA vaccines have emerged as a promising platform for cancer immunotherapy, particularly following the success of COVID-19 vaccines. However, the development of cancer vaccines presents challenges such as difficulties in antigen prediction and poor immunogenicity, especially in identifying and delivering highly immunogenic tumor-specific antigens. The variability and low immunogenicity of tumor antigens further complicates this process. METHODS: This study utilized public data and bioinformatics analysis to identify potential tumor antigens in cancer and characterize different immune subtypes. This approach aims to guide the development of cancer mRNA vaccines with enhanced immune response. RESULTS: In the cancer genome atlas rectal adenocarcinoma(TCGA-READ), Exon skipping was the most common alternative splicing event in TCGA-READ, whereas mutually exclusive exons were the least common. We identified 4480 upregulated and 3328 downregulated AS events, with missense mutations being the most frequent. A total of 217 potential antigen genes were identified by intersecting upregulated AS anomalies and frameshift mutations. CONCLUSIONS:
Vaccine research teams can use these findings as a starting point for prioritizing validation of candidate peptides derived from FAM135A. For example, they could first determine whether candidate peptides are actually presented by rectal cancer cells, then design studies involving T-cell activation experiments and assessments of immune responses and antitumor effects in tumor models. This would test whether a computationally promising candidate can function as an actual vaccine target.
The C1 and C2 classification remains a patient-stratification hypothesis to evaluate in relation to treatment response in future prospective rectal cancer studies. Current evidence is insufficient to select patients for vaccination or guide immune checkpoint inhibitor prescriptions. Reduced treatment burden, fewer side effects, and the possibility of a cure were not demonstrated as patient benefits in this study.