Reclassification of Splicing Gene Variants in Hereditary Cancer: Cases Report and Literature Review

Abstract

Alternative splicing (AS), a crucial cellular process, is a source of transcriptomic expansion and protein variability. Its contribution to cancer development and progression among a vast repertoire of human diseases, is highlighted lately and is under extensive investigation. In this review, the relative recent aspects of AS as a hallmark of cancer are described. In parallel, the importance of the identification of splicing-related variants through next-generation sequencing technologies is discussed. Cancer therapy and the management of patients and their families can highly benefit by the classification of these variants.

RNA splicing is a prevalent eukaryotic cellular process, which affects the expression of the vast majority of human genes at both qualitative and quantitative levels (1). Splicing is defined as the removal of the internal sequences (introns) from the premature mRNAs and the subsequent joining (henceforth ‘splicing’) of the remaining exonic regions, resulting in mature functional mRNAs that are translated to proteins. While the process of splicing tends to produce one or a couple of dominant (constitutive) transcripts from a given locus, alternative splicing (AS) refers to the process of generation of different transcripts from the same premature mRNA. AS is an essential source of protein variability, which, in turn, increases the complexity of the human transcriptome (2,3). AS explains the discordance between the limited number of the human protein coding genes identified by the Human Genome Project and the large number of proteins with different functions, since ~20,000 human protein coding genes may produce as many as ~145,000 transcripts (4). AS is quite prevalent in metazoans, with more than 95% of genes with multiple exons giving rise to alternative transcripts (5,6).

An ever-expanding number of examples demonstrate that AS contributes to a vast repertoire of human diseases. Dysregulation of the splicing process is associated with cardiovascular disease (7), autoimmune disease (8), neurological diseases (9), diabetes (10), human aging and longevity (11), and finally cancer, including human solid tumors and hematological malignancies.

A number of technical developments in the field of next-generation sequencing (NGS) over the last years has enabled the integration of multi-gene panel analysis in clinical practice. A major challenge in such analyses is the interpretation of Variants of Unknown Significance (VUS) that are identified in high rates (35-48%) and often become obstacles in management decisions (12-15). Clinical testing associated with hereditary cancer gene panels results in the identification of splicing variants (SVs) as a significant proportion of the VUS (15%-25%) (16). More frequently, canonical splice sites located at the exon/intron boundary sequences contain GT and AG motifs at the 5’ and 3’ ends of the intron, respectively (17).

AS is crucial in maintaining cellular homeostasis and regulating cell differentiation and development. However, aberrations in AS may cause changes in the normal expression pattern that may be associated with processes typically contributing to human cancer among other human diseases (18)⁠. Recent advances in transcriptomic sequencing and the access in large patient databases have revealed that aberrant AS incidents in human cancers are present in such an extent that can be characterized as a cancer hallmark (19,20). In addition, studies of the AS landscape in >8,000 tumors across 32 cancer types have revealed >15,000 cancer specific splice variants (21). This kind of deregulated splicing is involved in the biogenesis and progression of tumors, while in parallel may serve as a candidate cancer specific marker and therapeutic target (22,23)⁠.

In this work, we first describe the basic mechanism of AS and then we review some of the major aspects of AS in association with cancer, with particular focus on cancer metabolism, cell proliferation, metastatic potential, and immune escape. In this context, we present two interesting cases analyzed in our lab, where we investigated single-nucleotide variants affecting splicing using a hereditary cancer gene panel approach. We conclude by discussing the importance of splicing variants as a form of “genetic dark matter” in the diagnosis and specialized treatment of cancer.

Alternative Splicing Is Fundamental in Regulating Protein Expression

The splicing machinery. AS refers to the process of the differential inclusion or exclusion of all or part of an exon or intron of the pre-mRNA into a suitable message for translation of the encoded protein, the final mRNA product (24). The five common modes of AS include exon skipping, mutually exclusive exon splicing, alternative 3’ and 5’ splice site usage, and intron retention (Figure 1A). These events are not mutually exclusive, and combinations thereof may also occur. AS patterns thus result in altered expression and/or different function of the encoded protein by either generating different open reading frames of the same mRNA or affecting the regulatory sequences in UTRs. In addition, AS may even act as an expression repressive mechanism by giving rise to transcripts with premature stop codons, leading to transcript degradation via non-sense mediated decay (NMD). In terms of molecular mechanism, AS relies on both cis- and trans-factors. The molecular machinery that effects splicing is the spliceosome. In addition, specific oligonucleotide elements flanking the splice sites, called splicing enhancers or silencers may enhance or repress the process respectively (also see below). Both the spliceosome and cis-acting elements are responsible for the shaping of both constitutive and AS (25).

Figure 1. Splicing machinery. A) The main alternative splicing events are divided into five types: exon skipping (also called cassette exon); intron retention; mutually exclusive exons (only some exons appear in mature mRNA); A5SS (the position of the 3’ end of the exon changes); A3SS (the position of the 5’ end of the exon changes). SS: splice site; A5SS: alternative SS. B) Spliceosome assembly. The core spliceosome is composed of 5 small nuclear RNPs (snRNPs; U1, U2, U3, U5, and U6 snRNPS). Cis-acting elements include sequences on the target pre-mRNA that guide the recruitment of both the spliceosome and the trans-acting factors required to complete intron excision and exon linkage. They are categorized as intron splicing enhancers (ISEs) and silencers (ISSs), as well as exon splicing enhancers (ESEs) and silencers (ESSs). The trans-acting factors SR and hnRNP proteins, are splicing regulators, recruited by the cis -acting elements to complete splicing.

Spliceosome is the enzymatic machinery recognizing splice sites at the borders of introns and exons and distinguishing them according to four canonical consensus motifs: (i) The 5’ splice-site (SS; characterized by a GU dinucleotide at the 5’ end of the intron), (ii) the 3’ SS (containing AG at the 3’ end of the intron), (iii) the branch point sequence (BPS; located upstream of the 3’ SS), and (iv) the polypyrimidine tract (located between the BPS and the 3’ SS). The core spliceosome is composed of 5 small nuclear ribonucleoproteins (RNPs) (snRNPs; U1, U2, U4, U5, and U6 snRNPs), complexes of small nuclear RNAs (snRNAs) and 300 associated proteins (26,27). The high level of fidelity required for splicing though, depends not only on the specific interactions between the spliceosome and the aforementioned consensus sequences, but also on the additional action of cis-acting elements and trans-acting factors. Cis-acting elements include sequences on the target pre-mRNA that guide the recruitment of both the spliceosome and the trans-acting factors required to complete intron excision and exon linkage. They are categorized as ISEs (intron splicing enhancers) and ISSs (silencers), as well as ESEs (exon splicing enhancers) and ESSs (silencers) (Figure 1B) (28-30).

Trans-acting factors in AS are mainly RNA binding proteins (RBPs), which bind adjacent to splice sites, promote the recruitment of the spliceosome and, through a complex interplay of enhancing or repressing the appropriate splice site usage of the substrate, determine the resulting pattern of exon joining (31,32). RBPs are divided in three major groups in humans, the SR (serine/arginine rich) proteins, that typically bind ESEs and ISEs and promote spliceosome assembly and splicing, the more structurally diverse canonical hnRNPs (heterogeneous nuclear ribonucleoproteins) that usually bind ESSs and ISSs to inhibit splicing and the hnRNP-like proteins expressed in a tissue or temporal specific mode (33-35).

In eukaryotic cells, the AS regulatory network is modulated by functional coupling between transcription and RNA processing. The transcription machinery can influence AS decisions by affecting the time in which cis-regulatory elements are transcribed or by assisting in the recruitment of trans-acting regulatory proteins. In the first case, changes in the elongation rate of RNA polymerase II are required, whereas in the second case chromatin structure affects the recruitment of splicing factors (36). Chromatin organization marks exon-intron structure through an increased nucleosome-occupancy level of exons with respect to introns (37,38).

Mutations Affecting Alternative Splicing

Trans acting factor mutations and alternative splicing. Trans acting mutations, i.e., mutations in genes encoding splicing regulatory proteins RBPs, have a larger impact on the cellular splicing network compared to mutations in cis-elements they are binding on, as RBPs affect a larger number of splicing events (39,40). The components of the spliceosome complex that are most affected by somatic mutations in cancer (especially in hematological malignancies) are the spliceosome core components SF3B1 and U2AF1 and SR proteins SRSF2 and ZRSR2 (41). Cancer-related mutations have been extensively observed in the small subunit of the U2AF complex (U2AF1) and a substrate binding subunit of the U2 snRNP (SF3B1) are core components of spliceosome (42). These mutations exhibit specific effects like exon skipping or use of alternate 3’splice sites rather than general splicing inhibition, explaining the dispensable nature of these mutated spliceosome components in the context of an optimized assembly (43,44).

ZRSR2 mutations and U2AF1 hotspot mutations at residues S34 and Q157 in the zinc finger domains affect the recognition of alternate 3’splice sites have been reported in myeloid malignancies (45-48). Approximately 50% of chronic myelomonocytic leukemia cases (CMML), as well as other hematologic dysplasias, are characterized by the mutation of proline residue P95 of the SR protein SRSF2, the best characterized example of a mutant RBP in cancer (49,50). Mutation at P95 located between the RS and RRM domains of SRSF2 protein to either arginine or histidine, shift the sequence specificity from the CCNG motif, which is preferred by the wild-type protein, to the GGNG RNA motif, leading to changes in the splicing patterns of hundreds of mRNAs (51).

Cis acting Factor Mutations and Alternative Splicing

Initially, the somatic mutations associated with the development and progression of tumors were constrained in missense, nonsense and frameshift mutations directly affecting the protein coding sequences. The explosion of available data, in the form of complete patient genome sequences, allowed the exploration of the impact of mutations on regulatory sequences, with hundreds of recurrent mutations identified in promoter and enhancer elements (52). Even though such mutations are more difficult to be ascribed to a functional role, their annotation and functional dissection have been the subject of a number of ongoing projects (53).

Outside the regulatory sequence, mutations in splicing cis-acting elements, involved in the regulation of splicing, are actually easier to identify as the corresponding sequences are often sequenced as part of Whole Exome Sequencing (WES), which is much more restricted compared to complete genome sequencing. It is also easier to assess the functional footprint of such mutations, as long as they overlap with known splicing regulatory elements. In terms of cancer-associated genes, a recent global exome analysis revealed mutations in more than 1,000 genes in 8,000 tumor samples producing new alternative splice sites or affecting ESEs (exon splicing enhancers) and ESSs (silencers) recruiting RBPs (54). Regarding the effect of these mutations, intron retention is recognized as a frequent consequence of the somatic mutations in cis-acting elements since they reduce the spliceosome’s efficiency in assembling on an intron and promoting its removal (55). In most of the cases the retention of one or more introns, introduces novel stop codons resulting in loss of function through protein product truncation or non-sense mediated decay of the mRNA (56).

Alternative Splicing and the Hallmarks of Cancer Cells

The molecular basis of cancer has been extensively studied under the prism of chromosomal abnormalities and genetic mutations affecting the protein coding sequences of genes. However, it is now well established that aberrant splicing can contribute to the development and progression of cancer. Alternate isoforms of immune and signaling proteins and enzymes with key role in tumor proliferation and apoptosis, tumor metabolism and angiogenesis, metastasis and immune escape of the cancer cells reveal the role of AS as a hallmark of cancer.

Alternative Splicing and Tumor Metabolism

The metabolic pattern of cancer cells is altered compared to normal cells with changes occurring in glucose, glutamine, fructose, and fatty acid metabolism. Genes expressing enzymes participating in these metabolic pathways are often affected by AS.

Tumor cells in contrast to normal cells prefer to metabolize glucose into lactic acid even under oxygen sufficient conditions in order to obtain energy more rapidly (also known as the Warburg effect) (57)⁠. Glycolysis is a 10 enzymatic reactions series with pyruvate kinase, the enzyme catalyzing the last step, leading to production of pyruvate (58). PKM1 and PKM2 are two AS isoforms of the pyruvate kinase gene PKM that differ in one alternative exon (9 and 10 for PKM1 and PKM2, respectively) (59-62)⁠. While the PKM1 isoform promotes oxidative phosphorylation in normal cells, PKM2 promotes aerobic glycolysis in cancer cells (63). Certain splicing regulators e.g., SRSF3, hnRNPA1, hnRNPA2 determine the AS of PKM affecting the dynamic ratio of PKM1/PKM2 isoforms (64). Enhancing the expression of PKM1 and reducing that of PKM2, limits the production of lactic acid and inhibits tumor progression, as observed in cases of pancreatic, hepatocellular, and colorectal cancers, suggesting them as therapeutic target candidates (65-67)⁠.

Tumor cells are characterized by enhanced reliance on the metabolism of glutamine into glutamate, a process catalyzed by the enzyme glutaminase (68,69). Glutaminase exists in two alternatively spliced variants, glutaminase C -dominant in adenoma, glioma, colorectal cancer, and breast tumors- and kidney-type glutaminase that has a slightly different C-termini and distinct 5’UTRs (70,71). Tumor cells exhibit a higher-level expression of glutaminase C isoform, giving a priority to glutaminase C expression compared to the kidney glutaminase isoform. Specific inhibition of the glutaminase C isoform can inhibit tumor progression (72-74), whereas knockdown of the polyadenylation factor CFIm25 targeting the 5’UTR of glutaminase C, affects AS and inhibits glutaminase C isoform expression in cells using it (75).

Fructose is used as a better carbon source for tumor cells, since its intermediate and final metabolic products can provide substrates for synthesis of phospholipids, triglycerides, ribose and NAPDH (76). Ketohexokinase (KHK) is the first enzyme participating in fructose metabolism and exists in two alternatively spliced variants, KHK-A and KHK-C, with tissue specific but mutually exclusive expression (77).

Expression of KHK-C is important in the progression of several diseases like diabetes, liver diseases and hypertension (78-80). Recently, it was demonstrated that hepatocarcinoma cells (HCC) reduce the fructose metabolism rate, which is mediated by heterogeneous nuclear ribonucleoprotein (hnRNP) H1 and H2-dependent AS of the KHK gene, resulting in a switch from high activity fructokinase (KHK)-C to low activity KHK-A isoform expression (81). KHK-A acts as a protein kinase to phosphorylate and activate PRPS1, leading to enhanced nucleic acid synthesis necessary for tumorigenesis (81).

Alternative Splicing and Cell Apoptosis and Proliferation

Normal cells manage to survive by maintaining a balance between proliferation and apoptosis. Tumor suppressor genes and oncogenes work as switches on certain checkpoints of the normal cell cycle, controlling the processes of division and death and preventing them from becoming carcinogenic. AS affects the expression of these genes and splicing factors, with aberrant spliced variants maintaining the abnormal proliferative and apoptotic behavior.

Mcl-1, an important member of the Bcl-2 gene family, is traditionally regarded as an anti-apoptotic factor. Mcl-1 pre-mRNA undergoes AS, producing anti-apoptotic Mcl-1L and pro-apoptotic Mcl-1S isoforms (82,83). Recently it was found that in gastric cancer (GC) the antiapoptotic Mcl-1L and pro-apoptotic Mcl-1S proteins were up-regulated and down-regulated, respectively. Moreover, increased Mcl-1L and decreased Mcl-1S levels contributed to gastric tumor proliferation and poor prognosis. Additionally, it is confirmed that a shift of Mcl-1 splicing from Mcl-1L to Mcl-1S could markedly enhance apoptosis and inhibit proliferation in GC (84). An innovative direction towards tumor treatment involving targeting apoptotic proteins through the manipulation of AS was suggested.

Alternative Splicing and Angiogenesis

Angiogenesis, the formation of new blood cells, is critical for tumor progression since it allows its continuous expansion. Vascular endothelial growth factor VEGF, a well-known angiogenesis inducer, can be regulated by AS, both directly and indirectly. The AS of the last 3 exons (6, 7, and 8) of VEGF-A results in different isoforms that control angiogenesis antagonistically, either by promoting or by inhibiting it (85). Down-regulation of the tumor suppressor factor Wilms’s tumor suppressor 1 (WT1), inhibits serine-arginine protein kinase 1 (SRPK1) expression, indirectly suppresses Serine and arginine rich splicing factor 1 (SRSF1) and results in the splicing of VEGF into isoform VEGF-120 with an anti-angiogenic effect (86). However, upregulation of the expression of VEGF-A by an alternatively spliced variant of the glioma associated oncogene homolog-1 (GL1) gene enhances the angiogenesis of glioblastomas (87). VEGF-A can be used as an anti-angiogenic therapeutic target (88).

Alternative Splicing in Cancer Cell Invasion and Metastasis

Determining whether a tumor is invasive and metastatic is important since this behavior might be an obstacle in its treatment. Tumor cells interact with their microenvironment promoting cancer metastasis through abnormal activation of epithelial-mesenchymal transition (EMT) (89). During EMT, tightly packed epithelial cells become loosely connected and transit to mesenchymal cells that can migrate and contribute to cancer cells metastasis. A major gene affected in this process is CD44, a cell-surface glycoprotein involved in cell–cell interactions, cell adhesion and migration (otherwise known as HCAM). CD44 may be alternatively spliced to produce two isoforms, CD44v and CD44s, which are gradually lost and gained, respectively, in epithelial cells undergoing EMT (90-92). Breast cancer tumor metastases have been associated with elevated levels of CD44s, reflecting the effect of AS in different stages of tumor progression (93).

Alternative Splicing and Immune Escape

The immune system can show two faces in tumor initiation and progression (94). On one hand the immune system plays an important role in recognizing neoantigens, surface antigens expressed by cancer cells to change their microenvironment and attack them through activation of an adaptive immune response organized to eliminate them (95). On the other hand, the infiltration of innate immune cells together with the growth factors and the cytokines they produce, enhances tumorigenesis (96,97). Tumor associated altered splicing events give rise to abundant neoantigens, increasing their variety and also affecting the efficiency of cancer immunotherapy (21,96). AS events have been described in immune signal transduction in macrophages limiting inflammatory response (98,99), and in the differentiation and activation of T and B cells (100).

AS increases the diversity of immunoglobulin (Ig)E transcripts in B cells. IgE exists in various subtypes, depending on different stimuli (101), including secreted and membrane-bound (102,103). HuR protein is a regulator of AS events in B cells, since its loss leads to B cell death (104,105). CD45 is used as marker to discriminate between naive (CD45RA+ or CD45RB+) and memory (CD45RO+) T cells (60,100). Alternatively spliced transcripts of CD45 by the splicing factor hnRNPLL, at different stages of lymphocyte development, indicate that CD45 and its isoforms are critical for the development and function of T lymphocytes (101,102). Evidently, identifying the role of different splice isoforms in specific cancers and the regulatory role of splicing factors is of great significance for revealing the tumor evasion mechanisms of the immune response.

Clinical Utility of Functional RNA Analysis for the Reclassification of Splicing Gene Variants in Hereditary Cancer

Splicing variants (SVs) identified with the implementation of NGS technologies and characterized as VUS lead to incorrect intron removal and alterations in the open reading frame in some cases (106). The detection of such splicing variants provides strong evidence of pathogenicity according to the guidelines of the American College of Medical Genetics and Genomics (ACMG) that is predicted also by in silico analysis (107). Most of them though are classified as VUS due to the lack of functional studies at the RNA or protein level (108). In order for the patient to follow precise therapeutic management according to the international guidelines and for the first-degree relatives to be tested for the finding, immediate reclassification of SVs classified as VUS is needed.

Below, we present the investigation of SVs in hereditary cancer genes and the utility of RNA analysis in clinical diagnostics as shown in two individuals referred to Genekor Medical S.A (Athens, Greece) for genetic testing using a hereditary cancer panel of 43 genes (Roche NimbleGen, Pleasanton, CA, USA). The clinical significance of variants was further examined using the standards and guidelines for the interpretation of sequence variants recommended by the ACMG Laboratory Quality Assurance Committee and the Association for Molecular Pathology (AMP) (107)⁠. The impact of the variants on splicing was computationally examined using the VarSeak software (108)⁠.

Two splicing variants were identified in two individuals that were initially classified as VUS using the ACMG guidelines. RNA analysis was performed in a new blood sample and the variants were reclassified as likely pathogenic.

Case 1

RNA analysis of a variant in BRCA2 (c.632-3_632-2delCA) revealed the deletion of exon 8, resulting in a premature stop codon. The splicing variant c.632-3_632-2delCA in the BRCA2 gene was identified in heterozygosity in a 66-old year-old woman who was diagnosed with pancreatic cancer (Figure 2). BRCA2 (c.632-3_632-2delCA) was predicted to affect a nucleotide within the consensus splice site of the intron. However, an adjacent adenine (A) nucleotide at c.632-4 preserves the canonical AG splice acceptor site sequence. This prediction was confirmed by RNA analysis, which showed that the c.632-3_632-2delCA variant affects splicing by creating an alternative splice site and thus generating a premature translation stop signal 10 amino acid residues downstream, predicted to result in a truncated protein (Figure 2C).

Figure 2. Analysis of BRCA2 c.632-3_632-2delCA. A) Schematic representation of exons. Arrow shows the position of the variant identified in the present study. B) Pedigree of the proband with the c.632-3_632-2delCA variant in BRCA2. C) RNA analysis of the variant revealed the deletion of exon 8, resulting in a premature stop codon (the second chromatogram) compared to the analysis of a wild-type sample (top chromatogram).

This variant has not been reported in the literature in individuals with BRCA2-related disease while it is present in the mutation database ClinVar (Variation ID: 96837).

The RNA analysis confirmed the disrupting impact of the splice site variant c.632-3_632-2delCA in the BRCA2 mRNA, leading to the definite classification of this variant as likely pathogenic. The truncated protein products caused by AS may be associated with an increased risk of pancreatic cancer.

Case 2

RNA analysis of a variant in the ATM gene (c.4109+1G>A) revealed the insertion of part of intron 26 immediately after exon 26, resulting in a premature stop codon. The splicing variant c.4109+1G>A in the ATM gene was identified in heterozygosity in a 43-old year-old woman who was diagnosed with breast cancer (Figure 3). ATM (c.4109+1G>A) was predicted to produce a loss of function of an authentic splice site. This prediction was confirmed by RNA analysis, which showed that the c.4109+1G>A variant affects splicing by creating an alternative splice site and thus generating a premature translation stop signal 10 amino acid residues downstream, predicted to result in a truncated protein (Figure 3C).

Figure 3. Analysis of ATM c.4109+1G>A. A) Schematic representation of exons. Arrow shows the position of the variant identified in the present study. B) Pedigree of the proband with the c.4109+1G>A variant in ATM. C) RNA analysis of the variant revealed the insertion part of intron 26 immediately after exon 25, resulting in a premature stop codon (the second chromatogram) compared to the analysis of a wild-type sample (top chromatogram).

This variant has not been reported in the literature but is present in the mutation database ClinVar. In this case as well, RNA analysis confirmed the disrupting impact of the splice site variant (c.4109+1G>A) in the ATM mRNA, which allows us to classify this variant as likely pathogenic. The truncated protein products caused by AS may thus be associated with an increased risk of breast cancer.

Discussion

AS is a crucial process in gene expression and expands the genome and transcriptome diversity in eukaryotic cells. It is thus not surprising that aberrations in AS may contribute to a number of pathological conditions, including cancer. All hallmarks in the process of the transformation of a normal to a cancer cell, including cell proliferation, escape from apoptosis, induction of angiogenesis, invasion and metastasis, metabolism and immune escape are now also studied in the context of AS. A large number of genetic variants affecting both the molecular machinery of the splicing process as well as the cis-acting splicing elements on protein coding genes have been associated with various cancer types. A major challenge though, is to identify the underlying mechanisms responsible for changes in the splicing pattern, which constitutes the first step in their use as potential therapeutic targets.

A key step in this process is the experimental validation of variants of clinical significance through RNA analysis, as presented in the two cases in this work. This allows, first, the unambiguous association of the identified variants as being splicing-related and then records their qualitative (and sometime also quantitative) impact on the splicing pattern. Once such a connection is made, it enables the clinical geneticist to document the variant as a risk factor. Besides its value in the field of diagnosis, this also means expansion of the list of possible targets against which precision medicine therapeutic interventions may be designed.

The main axes of therapeutic interventions against aberrant splicing events are small molecule inhibitors, antisense oligonucleotides (ASOs), CRISPR-Cas9 editing technology and immunotherapy. Small molecule inhibitors can be used as effective therapeutic molecules to increase tumor cell death by targeting individual spliceosome proteins e.g., SF3B1 commonly altered in certain cancers (109-111). ASOs are therapeutic vehicles modulating mRNA splicing by exon inclusion or exclusion that have already succeeded to improve clinical outcomes in patients with spinal muscular atrophy (SMA) (112,113) and Duchenne muscular dystrophy (DMD) (111). There is in vitro evidence of cancer inhibition using ASOs (111), while several pilot clinical trials are in progress to explore ASO use in solid tumors, leukemias, and lymphomas. In vivo application of these therapies though is not without barriers, due to their degradation or toxicity (112,113).

The implementation of NGS technologies in routine clinical practice must be considered a basic prerequisite for further advancements in this field. Gene panels may provide the initial data, while more targeted RNA sequencing of identified candidate loci may reveal all the splicing variants at the levels of both solid tumors and hereditary cancer syndromes. Besides the importance of their identification for cancer therapy, their classification remains an important task for the management of the patient and the application of the international guidelines on patient management and family counseling. Community efforts in constantly updating and expanding the corresponding databases are also crucial to uncover the hidden layers of transcriptome complexity, which may include AS.

The above efforts have resulted nowadays in large-scale datasets that need to be analyzed and interpreted. Efficient algorithms are constantly being developed for processing, building, and matching the genomes or determining the gene expression differences under normal or disease conditions (114). Machine learning (ML) is constantly being developed in order to aid bioscientists to predict the outcome of biomodels under investigation, uncovering the fundamental mechanisms in biology (114).

Conflicts of Interest

The Authors have no conflicts of interest to declare in relation to this study.

Authors’ Contributions

Conceptualization, D.B. and K.A.; Methodology, D.B. and K.A.; Software, G.N.T.; Writing – Original Draft Preparation, D.B.; Writing – Review & Editing, D.B. and K.A.; Supervision, E.P. and G.N.; Project Administration, G.N.

Acknowledgements

The Authors thank the doctors and patients who participated in this study for their contribution.