Genetics of congenital solid tumors

Abstract

When we discuss the genetics of tumors, we cannot fail to remember that in the second decade of the twentieth century, more precisely in 1914, Theodore Boveri defined for the first time the chromosomal bases of cancer. In the last 30 years, progresses in genetics have only confirmed Boveri’s remarkable predictions made more than 80 years ago. Before the cloning of the retinoblastoma 1 (RB1) gene, the existence of a genetic component in most, if not all, solid childhood tumors were well known. The existence of familial tumor aggregations has been found much more frequently than researchers expected to find at random. Sometimes, the demonstration of this family predisposition was very difficult, because the survival of children diagnosed as having a certain tumor, up to an age at which reproduction and procreation is possible, was very rare. In recent years, advances in the diagnosis and treatment of these diseases have made it possible for these children to survive until the age when they were able to start their own families, including the ability to procreate. Four distinct groups of so-called cancer genes have been identified: oncogenes, which promote tumor cell proliferation; tumor suppressor genes, which inhibit this growth/proliferation; anti-mutational genes, with a role in deoxyribonucleic acid (DNA) stability; and micro-ribonucleic acid (miRNA) genes, with a role in the posttranscriptional process.

⧉ Introduction

Cancer is defined as a different group of diseases that have as common denominator the disruption of the mechanisms of proliferation, differentiation, and cell growth. It is among the most common causes of death, being placed, in developed countries, in second place after heart diseases. Different risk factors are involved in the appearance of cancer, such as: environmental factors, age, gender and last, but not least, genetic predisposition. From the end of the twentieth century, it was known that neoplasia is a complex condition caused by genetic and epigenetic changes in multiple genes [1,2,3].

Extremely rapid advances in genetics, especially after the design of the human genome, the emergence and development of molecular genetics technologies that can analyze the human genome, transcriptome, and proteome of normal and cancer cells and, last, but not least, the sequencing and mapping of the whole human genome has produced a real revolution in the diagnosis and treatment of cancer. Until 2017, a number of 612 genes involved in the pathogenesis of cancer were known.

Aim

The present paper aims to review the main genetic changes that occur in solid tumors of the child.

⧉ Risk factors for the appearance of cancer

Carcinogenesis defines the general process of cancer formation, this process being multistage, involving the accumulation over time of mutations and epigenetic alterations that will produce changes of the genes. In fact, in carcinogenesis there is an imbalance between proliferation and cell death. At adults, environmental factors, and lifestyle (smoking, overeating, sedentary lifestyle, etc.) play a major role in cancer, unlike children where they play an insignificant role.

Genetic factors are incriminated in both the cancerous pathology of the adult and the child. The main genes involved are oncogenes and tumor suppressor genes. In addition to these two main categories of genes, other genes are also discussed, such as: stability (anti-mutational) genes and micro-ribonucleic acid (miRNA) genes. Deoxyribonucleic acid (DNA) alteration is considered, in 70% of cases, the main factor involved in the etiology of cancer. Alterations in DNA consist of changes in its chemical structure, while mutations occur in changes in the DNA sequence, with different subsequent consequences. The genetic predisposition caused by germline mutations is responsible for 30% of cancers (causes hereditary or genetically predisposed cancers).

All these mechanisms (somatic mutations, germinal and epigenetic changes) cause instability of the genome decreasing its ability to remain stable, the occurrence of unrepaired DNA alterations resulting in new somatic mutations and epigenetic alterations.

Oncogenes

Proto-oncogenes are normal genes with a role in regulating cell division (causes cell growth, profiling and inhibits apoptosis). They encode proteins located in various cellular compartments that are expressed at different stages of the cell cycle. As a result of structural or functional changes with overregulation of function (“gain of function”), protooncogenes become oncogenes that cause marked cell proliferation, inhibit cell death, and cause malignant transformation of the carrier cell. The gene product of an oncogene is called an oncoprotein. The mutations that determine the activation of oncogenes are dominant, with “gain of function”, practically, for the activation in cancer, another major event is necessary (mutations in other genes) [4].

Oncoproteins are classified, according to their role and location in the cell into seven classes:

(i) Cell growth factors: SIS, HST, INT1, and INT2. They encode proteins discharged into the extracellular space. They contribute to malignancy by permanently activating membrane receptors that normally initiate proliferation or differentiation.

(ii) Protein kinases, with or without receptor function: ERBB1, KIT. Up to date, 30% of the described oncogenes encode proteins with kinase activity, mostly growth factor receptors. In the absence of specific ligands, these oncoproteins phosphorylate the tyrosine residues that provide, permanently and uncontrolled, intracellular signals, thus triggering unlimited proliferation. Examples: ERBB1 specifies the kinase domain of the epidermal growth factor receptor (EGFR); KIT is the receptor for macrophage growth factor (MGFR).

(iii) Receptors without tyrosine kinase activity: the c-mas proto-oncogene encodes the angiotensin receptor.

(iv) Membrane G proteins. The ras proto-oncogene family encodes proteins with a molecular weight of 21 kDa, with the property of binding and hydrolyzing adenosine triphosphate (ATP).

(v) Cytoplasmic serine/threonine kinases: the c-mos genes regulate the development of normal meiosis, and the c-raf genes stimulate mitogenesis.

(vi) Cytoplasmic localization regulators: c-crs genes that phosphorylate tyrosine.

(vii) Transcription factors: more than 20 transcription factors are currently known which consist of two functional domains that produce DNA binding (promoter) and transcriptional activation. We mention the c-erb A gene that encodes the erb A protein (thyroid hormone receptor) and the MYC gene family (c-MYC, n-MYC and l-MYC).

More than 100 protooncogenes have been identified but only a small fraction of them have been shown to play an important role in the development of cancer [5,6].

Tumor suppressor genes (antioncogenes)

Tumor suppressor genes (antioncogenes) are genes that in the normal state (without mutations) control the production of proteins that are part of the cell division regulatory system. Through their products, they inhibit cell proliferation and stimulate cell death (apoptosis). Gene mutations, with loss or reduction of function, produce uncontrolled cell proliferation and inefficient apoptosis. Genes have a recessive character at the cellular level; to change the behavior of the cell it is mandatory that both alleles be inactivated [loss of heterozygosity (LOH)]. Loss of tumor suppressor gene function appears to be more significant in tumor production than oncogene activation.

The retinoblastoma protein (pRB1) was the first tumor suppressor gene identified (1986). The p53 protein (also called the “guardian of the genome”) has multiple functions: DNA repair, apoptosis, transcription, regularization of the cell cycle. Mutation of the p53 gene, located on chromosome 17p13.1 is involved in over 50% of cancers (colon, breast, lung, but also leukemia, lymphomas, sarcomas and tumors of the nervous system), increasing the therapeutic interest in the study of this gene in all cancers. Mutations in the p53 gene may be hereditary, as in Li–Fraumeni syndrome (early onset, multiple tumors, and multiple family cancers), in which there is an increased risk of cancer (choroid plexus carcinoma, leukemia, adrenal, breast, ovary, etc.). Other examples of tumor suppressor genes are: B-cell lymphoma 2 (BCL-2) (leukemia/B-cell lymphoma), switch/sucrose non-fermentable (SWI/SNF) is a protein complex with a role in remodeling the chromatin packaging pathway, Von Hippel–Lindau protein (pVHL), adenomatous polyposis coli (APC), cluster of differentiation 95 (CD95) (involved in various neoplasms and diseases of the immune system).

Stability (anti-mutational) genes

Stability (anti-mutational) genes are responsible for the integrity of the genome and the correctness of information transfer. They correct the damage in the DNA due to physical, chemical or biological agents. If both genes are lost (recessive model), the DNA damage cannot be repaired, leading to a large number (of the order of thousands) of stable DNA damage transmissible to offspring.

Regulatory genes (miRNA genes)

Regulatory genes (miRNA genes) are genes that do not code proteins, but sequences of single-stranded non-coding RNA (ncRNA), miRNA, long non-coding RNA (lncRNA), which are involved in the transcriptional or posttranscriptional regulation of gene expression. The miRNA genes act either by blocking the translation of messenger RNA (mRNA) into proteins or by degrading mRNA, thus controlling the cell cycle, aging and cell apoptosis. The lncRNA genes are involved in the control of transcription and post-transcriptional phenomena of mRNA.

Landscaper genes

They have a role in modifying the cellular microenvironment, thus favoring cell proliferation and growth, as they do not have a direct role in malignant transformation. Their role is in controlling the stability of the membrane and the cell matrix, thus controlling the cell–cell, cell–extracellular matrix interactions, their mutations inducing uncontrolled cell proliferation [7].

Epigenetic changes in cancer

Epigenetic mechanisms do not cause changes in the nucleotide sequence of DNA, but only in the nucleoprotein complex that forms the nuclear chromatin, influencing the spatial or temporal expression of genes. The main epigenetic phenomena are DNA methylation, histone modification (acetylation, methylation) and nucleosome repositioning. These changes play a vital role in normal development, while ensuring the stability of the genome (maintaining the stability and function of centromeres and telomeres). In cancer, epigenetic disorders occur in all cells, thus disrupting the degree of chromatin compaction and gene expression. In this way, oncogenes can be activated, and tumor suppressor genes inactivated, causing mutations and, implicitly, cancer.

⧉ Hemangiomas and vascular tumors

Hemangiomas are the most common tumors in childhood, occurring in 3–10% of newborns and infants. Morphologically, they are dysplasia characterized by an abnormal organization of vascular tissue (proliferation of vascular endothelium), causing dense agglomeration of blood vessels (capillaries, veins, arteries, lymph or any combination of them). In the first stage, the hemangiomas undergo a rapid initial proliferation, followed by a stabilization stage, and in most cases involution and regression [8]. There are currently several hypotheses regarding the pathogenesis of hemangiomas. According to Cheung et al., there are two theories that lead to the appearance of hemangiomas: the first refers to the fact that hemangiomas arise from the recrudescence of “sleeping” embryonic angioblasts; the second stipulates that hemangiomas are in fact tumors of neoangiogenesis. After the onset, there are two forms: infantile hemangiomas and congenital hemangiomas. Infantile hemangiomas are the most common, have an onset after birth, proliferate in the first year of life, and then regress. According to Haggstadd et al. by the age of 5, about 50% of hemangiomas involve and by the age of 9, about 90% [9]. Congenital hemangiomas begin at birth. Subsequently, some involve fast or slow, while others show no regression [10,11].

Histologically, infant hemangiomas are glucose transporter 1 (GLUT-1) positive, an erythrocyte-type glucose transporter. Vascular endothelial growth factor (VEGF) appears to have angioproliferative effects. Zang et al. demonstrated that there are increased plasma amounts of VEGF-A in patients with infantile hemangiomas during the proliferative phase compared to the infantile involution phase and also compared to the control groups [12]. Congenital hemangiomas, on the other hand, are GLUT-1 negative. Polymorphic mutations of VEGF have been found in some studies, particularly in infantile hemangiomas [13].

There are differences between hemangiomas themselves and vascular malformations. Hemangiomas are neoplastic lesions that go with epithelial hypertrophy, while in vascular malformations the epithelial turnover is normal [14,15].

The literature describes many genetic syndromes that associate hemangiomas in the clinical phenotype, such as: Von Hippel–Lindau (VHL) syndrome and Maffucci syndrome [16]. VHL syndrome is an autosomal dominant disease with a predisposition to both benign and malignant tumors (retinal, cerebellar, hemangioblastoma, renal, pheochromocytoma, pancreas). Genetically, there are mutations in the VHL tumor suppressor gene, located on chromosome 3. There are two types of VHL syndrome: type 1 that associates hemangioblastoma, renal cell carcinoma, but not pheochromocytoma. It is associated with large and truncated deletions of the gene leading to loss of protein product or hypofunction. Pheochromocytoma occurs in type 2 and the mutations are nonsense mutations of the VHL gene that lead to the formation of a functional, but insufficient protein [17,18,19]. Maffucci syndrome combines hemangiomatosis and enchondromatosis. Hemangiomas are usually cutaneous or subcutaneous, rarely causing bleeding [20]. Although this syndrome occurs sporadically, certain mutations, associated with the occurrence of enchondromatosis, have been identified [21,22]. One study found that 77% of patients with Maffucci syndrome had mutations in isocitrate dehydrogenase (IDH) in their tumors, especially IDH1 (98%) and IDH2 (2%) [23]. These mutations are associated with hypermethylation and downregulated expression of several genes. Notably, even tumors positive for these mutations demonstrated “intraneoplastic mosaicism”, a condition in which there exists a mixture of cells that did and did not express the mutant protein.

⧉ Teratomas

Teratomas are embryonic tumors that are histologically composed of cells from two or all three embryonal structures (ectoderm, mesoderm, and endoderm), with an increased incidence in newborns and children [24]. In 1863, Virchow first described this tumor form, and after 25 years the second case was reported by Horsleng [25]. The most common locations of teratomas are at the sacrococcygeal level (the incidence being one in 27 000 newborns), followed by those in the ovaries/testicles and brain [26].

In most cases, they occur sporadically, family cases being rarely described in the literature. In cases with familial aggregation, transmission was autosomal dominant, with a 50% risk of recurrence in offspring.

There is a direct relationship between aneuploidy and the histological grade of the tumor (immature teratoma). Grade 1 or two immature teratomas are usually diploid in about 90% of cases, aneuploidy being much more common in grade 3, chromosomal abnormalities being present in over 66% of cases [27]. Trisomy for chromosome 3 is the most common trisomy reported in immature ovarian teratomas (in adults). The testicular teratoma was associated with abnormalities of the short arm of chromosome 12 (12p) [28].

Various chromosomal abnormalities were also reported: at chromosome 7 (deletion 7q, associated with partial trisomy 2p); 22; 8p; partial duplication of the long arm of chromosome 1; deletions and mutations in the hSNF5/INI1 gene. Increased activity of the mutant p53 protein is associated with an increased risk of malignancy. The variety of cytogenetic patterns described supports the hypothesis that different genetic pathways may lead to the phenotype characteristic of this type of tumor [29].

Other studies have shown the common clonal origin of teratomas and malignancies from testicular germ cells [30]. The defining element of teratomas, which also explains the particular scientific interest given, is represented by the special variety of histopathological changes.

Central nervous system (CNS) teratomas

CNS teratomas are different from those with other locations; they have an increased incidence among children. They are found in about 0.5–1.5% of all brain tumors in children, with an estimated survival rate of 7.8%. In CNS teratomas, chromosomal abnormalities are found with an average of 12.8 (8 with gain of function, 4.8 deletions), those at chromosomes 4 and 5 being the most frequently reported. Mutations in the genes have also been reported: cyclin D1 (CCND1) located on chromosome 12 (12p13), RB1 on chromosome 13 (13q14) and PR/SET domain 14 (PRDM14) located on chromosome 8 (8q13), genes that play an important role in the cyclin pathway, cyclin-dependent kinase (CDK)/RB-E2F, cyclins having the role of controlling the progression of cells through the cell cycle by activating CDK enzymes [31]. Interestingly, many pineal region germ cell tumors (GCTs) were found to have multiple copies of the X chromosomes, which may explain the high incidence of GCT in patients with Klinefelter’s syndrome.

⧉ Brain tumors

Brain tumors rank second as common childhood tumors. There are mutations that lead to the production of astrocytic infiltrative tumors: molecular studies have identified certain genetic changes that underlie the histopathological differences between these tumors.

Astrocytomas

Astrocytomas are CNS tumors derived from astrocytes. They are classified into two categories: the first category includes astrocytomas with narrow infiltration areas (pilocytic astrocytoma, giant cell subependymal, and infiltration xanthoastrocytoma); the second category includes astrocytomas with diffuse infiltration areas (low-grade infiltrative astrocytoma, anaplastic astrocytoma, and glioblastoma). About two-thirds of low-grade infiltrative astrocytomas (grade 2) had mutations detected in the p53 tumor suppressor gene, with cases of astrocytomas being reported in families with members with Li–Fraumeni syndrome, neurofibromatosis type 1 (NF1), Noonan syndrome.

Pilocytic astrocytomas (PA) are the most common astrocytic tumors of childhood and differ clinically and histopathologically from astrocytomas affecting adults. PA have been reported in 15% of patients with NF1 [32]. The protein encoded by the NF gene, neurofibromin, stimulates the intrinsic hydrolysis of guanosine triphosphate (GTP) and acts as a tumor suppressor [33]. Also, Noonan syndrome, which is characterized by germline mutations in the genes of the mitogen-activated protein kinase (MAPK) [protein tyrosine phosphatase non-receptor type 11 (PTPN11), SOS Ras/Rac guanine nucleotide exchange factor 1 (SOS1), Kirsten rat sarcoma viral oncogene homolog (KRAS), neuroblastoma RAS viral oncogene homolog (NRAS), Raf-1 proto-oncogene, serine/threonine kinase (RAF1), B-Raf proto-oncogene, serine/threonine kinase (BRAF), SHOC2 leucine rich repeat scaffold protein (SHOC2) and Cbl proto-oncogene (CBL)], associate PA, but extremely rarely [34].

There have been numerous studies that have shown changes in the BRAF oncogene, which encodes a protein that belongs to the RAS/MAPK signaling pathway that controls important cellular functions (cell growth and division, cell differentiation and migration, apoptosis). At the BRAF gene, a 2 Mb duplication occurs at the 7q34 level by fusing the BRAF oncogene with the KIAA1549 gene, resulting in a transforming fusion gene. The N-terminal end of the KIAA1549 protein replaces the N-terminal regulatory region of BRAF, while maintaining the BRAF kinase domain, which becomes constitutively activated. This fusion occurs in over 70% of patients with astrocytomas [35,36]. Although the identification of gene fusion between KIAA1549 and BRAF by whole genome sequencing or RNA sequencing is used as a positive diagnostic marker for brain pathology, there are some studies showing that this fusion is missing in glial astrocytomas including adult oligodendrogliomas where 1p and 19q deletion occurs and the IDH1 gene mutation [37].

Mutations of the BRAF gene without gene fusion have also been mentioned, such as the well-known V600E mutation in which valine is replaced by glutamic acid, as well as a series of small insertions that activate BRAF signaling [35]. Recent studies have shown mutations in the fibroblast growth factors receptor 1 (FGFR1) gene family and in the neurotrophic receptor tyrosine kinase (NTRK) gene, where gene fusions occur. At the FGFR1 level, the mutations were variable.

Glioblastomas belong to astrocytomas with diffuse infiltration areas and represent the most frequent and aggressive tumors in adults with an annual incidence of 5.26 per 100 000 inhabitants [38]. The genetic changes that occur most frequently are point mutations in the IDH1 and IDH2 genes, which encode isocitrate, these mutations being correlated with the young age of onset and with a favorable prognosis, most glioblastomas being the wild type of IDH. Molecular changes also include mutations in the receptor tyrosine kinase (RTK) pathway of the RAS pathway of the p53 gene [39].

Neuroblastomas

They are neuroendocrine malignancies that develop from neural crest cells and are the most common type of solid extracranial cancer of childhood, with onset up to five years. They usually occur sporadically, but family cases with an incidence of 1–2% are also reported [40,41].

Only a few familial cases of neuroblatomas in which gene mutations occur have been reported in the literature. Tyrosine kinase receptors (TrkA, TrkB and TrkC) and their ligands, nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF), neurotrophin-3 (NT-3), and Aurora kinase A (AURKA) are involved in the mechanisms of regulation of differentiation, cell proliferation, apoptosis, angiogenesis, and metastasis.

The presence of TrkA (synonymous with NTRK1) in neuroblastoma is a favorable prognostic marker, TrkA receptors stimulating cell differentiation in the presence of NGF ligand. TrkB and BDNF are expressed in aggressive forms of neuroblastoma accompanied by amplification of the MYCN oncogene [42]. AURKA pathway signals direct oncogenesis and promote tumorigenesis. If it is low, the MYCN expression will decrease and the survival time will increase, but in overexpression, the survival time decreases. AURKA stimulates cell proliferation, cell growth and mediates VEGF angiogenesis. Amplification of the MYCN oncogene is an unsatisfactory prognostic factor in neuroblastoma [43].

There have also been identified three non-sense germline mutations (G1128A, R1192P and R1275Q) in the tyrosine kinase domain of the protein encoded by the anaplastic lymphoma kinase (ALK) oncogene. In one family case, there were reported two non-germline mutations (R100L and R141G) located in the similar homeobox 2B (PHOX2B) [44].

There have been identified various chromosomal abnormalities, such as:

(i) Amplification of the MYCN oncogene (located at the level of chromosome 2) is found in 18–38% of primary tumors, as well as in other cell lines of neuroblastomas [45,46,47,48]. This is amplified at the 2p24 resident site which is present in many neuroblastoma’ cell lines, such as CHP134, KPNYS, IMR-5, GOTO and NB17, but also in other solid tumors, such as lung carcinoma, adrenal tumors, breast cancer, and esophagus. The amplified segments at the respective chromosomal site show homogeneous staining (hsr).

(ii) 1p36 deletion is the most common deletion encountered in neuroblastoma. In this region there is the KIF1B gene, which is a tumor suppressor gene, its mutations being identified in several family cases, thus showing that the gene is involved in the appearance and progression of neuroblastoma. Amplification of MYCN oncogene together with 1p36 deletion plays an important role in the vascularity of these tumors. There are, also, other genes in this region of chromosome 1, involved in the pathogenesis of neuroblastoma. Abel et al. highlighted the DNA fragmentation factor subunit alpha (DFFA) gene (1p36) in several cases of Scandinavian neuroblastomas [49].

(iii) Chromosome 17 abnormalities have been identified since 1980. Mutual translocation between chromosome 1 and 17 der(1)t(1p; 17q), which results in loss of 1p and gain of 17q terminal (unbalanced translocation), occurs frequently in primary tumor, but also in neuroblastoma cell lines. The baculoviral IAP repeat containing 5 (BIRC5) gene located at 17q25.3 is expressed in many cases of primary neurofibroblastomas or in various survivin neuroblast cell lines.

(iv) Loss of 11q heterozygosity (at the 11q23.3 locus) is a negative prognostic factor representing neuroblastoma metastasis. This deletion may be accompanied by other deletions: 3p, 4p, 14q and is inversely correlated with MYCN oncogene amplification.

We can thus talk about three genetically distinct subgroups of neuroblastomas. According to Brodeur et al., we have the following groups: infants in the early stage and with a very good prognosis whose tumors are triploid, but no 1p deletion is observed and without amplification of the MYCN oncogene; children older than one year in stage 3 or 4 disease with diploid or tetraploid tumors, without amplification of MYCN, but with 1p loss; older children, with advanced disease, with di- or tetraploid tumors, in which we have both MYCN and lipooligosaccharide (LOS) amplification of chromosome 1 with a negative prognosis [50].

⧉ Retinoblastoma (RB)

RB represents a rare form of cancer with primary localization in the eye (retina); it is also called retinal glioma. It is the most common ocular tumor of the child representing 6% of all tumors in children with onset in early childhood (up to five years) formed by bialelic activation of a tumor suppressor gene. The incidence is 15 000–20 000 live births [51]. Although the incidence of the disease is relatively constant, there are areas of the Globe where it is higher, especially in underdeveloped areas of the world. This is explained by the fact that in these parts of the world the infection with the human papillomavirus (HPV) is widespread. HPV-infected cells produce an oncogenic protein, which binds to pRB by inactivating it.

RB can be evaluated on a genetic and laterality basis, it can be hereditary or sporadic, unilateral or bilateral. Tumor unilaterality occurs in 60% of cases and bilateral involvement is present in the other 40% of cases. Bilateral forms of RB have familial aggregation as opposed to unilateral ones that occur sporadically.

The molecular changes involved in pathogenesis are complex. The most common are germline or somatic mutations in the RB1 gene located on the chromosome 13q14.1-q14.2, this being the first tumor suppressor gene identified and expressed in various tissues [52]. It encodes the pRB protein which has multiple roles, the most important being the cell cycle stopping in the G1 phase, but also intervenes in cell differentiation, apoptosis control, chromosomal stability by temporarily stopping the cell cycle. Along with p107 and p130, pRB belongs to a family of proteins that bind to E2F transcription factors, their activity being controlled by phosphorylation. In human cancers, frequent mutations occur in the RB, while p107 and p130 are rarely inactivated [53]. The mutations that occur range from large deletions to non-sense mutations, splicing mutations. Germline mutations occur in 75–80% of cases of bilateral RB and 15–25% in cases of unilateral RB [54]. Another aspect encountered is that in unilateral forms there is a loss of RB1 function (biliary mutations) in 98% of cases, the remaining 2% having the normal RB1 gene, but have changes in the MYCN oncogene [55].

Epigenetic mechanisms also play an important role in the occurrence of these RBs including DNA methylation, miRNA histone modification, lncRNA, ATP-dependent chromatin remodeling [56,57].

The p53 gene also plays a role in the pathogenesis of RB, being the gene with the most common mutations in human cancer. It controls the cell cycle and causes apoptosis in the retinal cone precursor cells (where the RB cell line originates). It controls the cell cycle and apoptosis in the retinal cone precursor cells, from which the RB cell line originates [58,59,60].

⧉ Wilms tumor (WT) or nephroblastoma

Nephroblastoma is a malignant embryonic tumor that originates in kidney cells, more precisely from blastemal nephrogenic cells, and often has different differentiation patterns. The incidence is one in 10 000 newborns. The relatively constant incidence of nephroblastomas in all geographical areas suggests that environmental factors do not play a major role in the development of these tumors. There have been variations in racial incidence, with the tumor being more common in the African–American population and less common among Asian children.

Nephroblastoma affects children of both sexes, with no predilection for either sex and with equal frequency in both kidneys. Over 95% of cases occur in children under 10 years of age, although there are presentations in adulthood.

In about 20% of patients, mutations are identified in the WT1 gene, a gene located on chromosome 11 (11p13) [61,62]. Approximately 50% of patients with nephroblastoma who have mutations in the WT1 gene also associated mutations in the catenin-beta 1 (CTNB1) gene (encoding the beta-catenin oncoprotein) located at the 3p22 level. The WTX gene [also called APC membrane recruitment protein 1 (AMER1)] is located at the Xq11.1 level, mutations in this gene occurring in 15% of cases. Both genes encode proteins that stimulate cell growth and division. At the short arm of chromosome 11 are the insulin-like growth factor 2 (IGF2) and H19 genes, both responsible for the appearance of this tumor form (reduced activity of the gene that usually stops cell growth and increased IGF2 activity that stimulates cell growth causes loss of cell growth control and intensifies growth and cell division) and eventually leads to the appearance of the tumor. In the chromosome 11p15.1, there is a second locus WT2 (imprinted region) responsible for the appearance of WT. Almost 3% of children with WT have germ epigenetic changes at 11p.15.5.

Chromosomal changes, with genetic material gain on chromosome 1q, are an indicator of a negative prognosis. This cytogenetic change at level 1q is most common in nephroblastoma and is seen in 30% of tumors. At level 16q, in addition to changes at level 1p, there are other tumor suppressor genes [63].

Changes in miRNA processing genes (miRNAPG) are present in approximately 20% of WT cases. The most frequently involved are Drosha ribonuclease III (DROSHA), DRCR8, Dicer 1, ribonuclease III (DICER1), and exportin 5 (XPO5) genes [64,65,66,67]. These genes, through their products, directly control the maturation of miRNA from the initial pre-miRNA transcriber to the final miRNA in the cytoplasm. Germline mutations of miRNAPG involving DICER1 and DIS3-like 3’–5’ exoribonuclease 2 (DIS3L2) genes may be encountered. Germline mutations in the DICER1 gene cause DICER1 syndrome – one allele is inactivated by germline mutation and the other allele has a somatic mutation that affects one of the RNase IIIb metal ion-binding sites (pleuro-pulmonary blastoma, cystic nephroma, ovarian cancer, thyroid, WT) [68]. Mutations in the DIS3L2 gene cause Perlman’s syndrome (polyhydramnios, macrosomia, bilateral kidney tumors – hamartomas with or without nephroblastomas) [69,70,71].

Nephroblastoma can occur in various genetic syndromes, but there are also non-syndromic causes. The most common genetic syndromes associated with WT are: WAGR syndrome (WT, aniridia, urogenital system abnormalities and mental retardation), Denys–Drash syndrome (congenital nephropathy, WT, gonadal disgenesia), Frasier syndrome (pseudohermaphroditism, glomerulopathy). All of these syndromes have mutations in the WT1 gene. Patients with Beckwith–Wiedemann syndrome may develop nephroblastoma through the mechanism of imprinting the IGF2 and H19 genes. Other syndromes that associate WT are: Simpson–Golabi–Behmel syndrome (macroglossia, macrosomia, kidney tumors), in which there are mutations in the glypican 3 (GPC3) and glypican 4 (GPC4) genes, these mutations increasing the risk of nephroblastoma; CLOVES syndrome (congenital lipomas, macrosomia, vascular malformations, epidermal nevi, skeletal abnormalities), in which mutations of the phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha (PIK3CA) gene occur; Sotos syndrome (macrosomia, cerebral gigantism, heart malformations, scoliosis, convulsions, nephroblastoma) with mutations in the nuclear receptor binding SET domain protein 1 (NSD1) gene; Bloom syndrome (sensitivity to ultraviolet radiation, WT); Li–Fraumeni syndrome (breast cancer, osteosarcoma, soft tissue cancer, leukemia, kidney cancer, WT, brain cancer); Alagille syndrome (congenital heart malformations, facial dysmorphism, vertebral, renal abnormalities); Bohring–Opitz syndrome (craniofacial dysmorphism, hypertrichosis, severe mental retardation).

Non-syndromic cases of WT are: (i) Family forms are found in only 2% of patients, although the number of genes involved is increased. Two genes are involved: FWT1 (17q12-q21) and FWT2 (19q13.4). Rarely, germline mutations may occur in the WT1 gene in which some family members may associate genitourinary malformations. (ii) Sporadic aniridia occurs because of mutation in one of the paired box 6 (PAX6) copies. About 50% of individuals with sporadic aniridia associate with WT.

⧉ Rhabdomyosarcomas (RMS)

RMS are malignant tumors starting with striated muscles and represent slightly more than 20% of all sarcomatous tumors of the child.

According to the macroscopic and histological aspect, we have three types of RMS: embryonal (ERMS), alveolar (ARMS) and pleomorphic (PRMS). The embryonal form is most common from birth to about 15 years, alveolar between 10–25 years, while pleomorphic RMS predominates in adults and older people and is extremely rare [72]. RMS occurs more frequently in male gender, sex ratio of 1.5:1 [72,73]. The most common location is in the head and neck (45%), especially orbits (30% of the total RMS in the head), followed by the neck, chest (40%) and limbs (20–25%). [74]. Due to the complex therapeutic modalities, the survival rate is currently over 70%, five years after diagnosis [75]. If we refer to the severity of clinical manifestations, the embryonal type has a more favorable prognosis (43% at five years) compared to the ARMS type which is the most severe form with increased risk of metastasis (25.9%) and five-year survival rate is up to 29% depending on the type of existing genomic rearrangements (different gene fusions) [76].

The most important histological subtypes in children are ARMS and ERMS, the molecular changes being different in the two types [77].

In ARMS, there is either mutual translocation between chromosomes 2 and 13, t(2; 13)(q35; q14) or between chromosomes 1 and 13, t(1; 13)(q36; q14), in both cases resulting in the fusion of paired box 3 (PAX3) genes – forkhead box O1 (FOXO1), respectively paired box 7 (PAX7) genes – FOXO. Following these recombinations, certain transcription factors will initiate the process of tumor genesis encoded proteins by stimulating cell growth and proliferation and stopping terminal differentiation [78].

Different genomic changes occur in the ERMS: mutations with loss of p53 function, activation of the RAS pathway or mutations in the myogenic differentiation 1 (MYOD1) gene [79,80,81]. The mutation of the MYOD1 gene together with the fusions of the vestigial like family member 2 (VGLL2) gene causes the appearance of an aggressive tumor form, sclerosing RMS [82].

Ricker et al. described different gene variants associated with ERMS, such as BUB1 mitotic checkpoint serine/threonine kinase B (BUB1B), DICER1, and FGFR1. The BUB1B gene mutation has a negative prognosis and leads to other cancers (gastric, breast, prostate) [83]. Decreased expression of DICER1 correlated with miRNA concentration during organogenesis has an important role in the occurrence of RMS and can occur in both familial and sporadic forms. Shukla et al. reported a percentage of 4.9% cases of RMS, in which mutations of the PI3KCA gene located on chromosome 3 occur [84].

Among the genetic syndromes that associate this tumor type, we mention Li–Fraumeni syndrome, neurofibromatosis, Beckwith–Wiedemann syndrome and Costello syndrome.

⧉ Bone tumors

Bone tumors develop by abnormal growth of cells inside the bone; they can be benign (osteoid osteoma and osteoblastoma) or malignant [osteosarcoma and Ewing’s sarcoma (EWS)].

Osteoid osteoma

Osteoid osteoma is a very painful benign bone tumor, usually located at the level of the bone shaft with an incidence of 12%. It can rarely turn into an osteoblastoma. It usually occurs in late childhood and in young adults. Osteoblastoma is a much rarer tumor (3% of benign bone tumors). The sex ratio is 2:1 in favor of males. It is most commonly located in the spine and sacrum. From a morphological point of view, the two tumor types are similar.

Molecular changes are similar in both forms with rearrangements in the Fos proto-oncogene (FOS) (AP-1 transcription factor subunit) 87% of cases and FOSB 3%, on chromosome 14. Both genes belong to the family of transcription factors FOS (which has a total of four members: FOS, FOSB, FOSL1, and FOSL2), with a role in the differentiation, proliferation, and transformation of osteoblasts. In some cases, FOS gene expression has been associated with cellular apoptosis [85].

Among malignant tumors, osteosarcoma and EWS are the most common bone tumors in children.

⧉ Osteosarcoma

Osteosarcoma is a severe malignant bone tumor with rapid metastasis and a low prognosis of survival at five years, about 71% [86]. It mainly affects young children and adolescents, the age of onset being 10–15 years, but cases can also occur at younger ages [87]. It has a slightly higher incidence in boys; sex ratio is 1.5: 1.

It is characterized from a molecular point of view by a high degree of genomic instability explained either by the occurrence of the phenomenon of chromothripsis or by the existence of gene mutations with a role in DNA repair. The most common genetic syndromes predisposed to osteosarcoma are: Li–Fraumeni syndrome and RB. Mutations in RecQ-like (RECQL) helicases (important role in DNA repair mechanisms) may occur in: Rothmund–Thomson syndrome (RECQL4 gene) or there may be syndromes in which germline mutations occur in RECQL helicases: RAPADILINO syndrome, Baller–Gerold syndrome, Werner syndrome and Bloom syndrome [88].

Histologically, there are different types of osteosarcomas, but the most common type is high-grade osteosarcoma. Complex changes occur at the level of chromosomes most often occurring mutual translocations, amplifications, deletions, all these chromosomal abnormalities causing a high degree of genomic instability.

Genomic instability in osteosarcoma is explained by the mechanism of chromothripsis in which thousands of chromosomal ruptures and rearrangements occur in a single or several chromosomes that are then grouped and oriented together in a random, chaotic manner. This phenomenon was first described by Stephens et al. showing the presence of chromothripsis in 3% of all cancers and 30% in osteosarcomas [89]. In addition to this phenomenon, many genes are involved in maintaining genomic stability, changes in tumor protein p53 (TP53) and RB1 being the most important in the occurrence of osteosarcoma. The most commonly affected is the TP53 gene, with genome-wide sequencing studies showing the presence of mutations in this gene in 47–90% of osteosarcoma cases, with translocations in the first intronic region of the TP53 gene being most commonly observed, followed by mutations of the RB1 gene [90]. It is known that this gene is involved in blocking the cell cycle in the G1 phase of the cell cycle (it stops the passage in the S phase until the cell is ready). Loss of the function of this gene causes an uncontrolled cell division [91]. Kovac et al. showed that in 29–47% of cases of osteosarcoma, somatic mutations can occur at RB1 level [92].

In addition to mutations in the two genes, there are other genomic changes such as germline and somatic mutations in the ATRX chromatin remodeler (ATRX) gene, with 29% of osteosarcoma cases showing somatic mutations in this helicase with a role in DNA stability [93]. Its mutations also cause alpha-thalassemia X-linked intellectual disability (ATR-X) syndrome, and in some forms of osteosarcomas there are mutations in breast cancer 1 (BRCA1) and breast cancer 2 (BRCA2) genes [94].

EWS is a malignant, aggressive tumor that begins in childhood or in young adults. Although remarkable progress is currently being reported in its treatment, mortality remains high, with death occurring in approximately 40% of patients [95].

In 85% of patients with EWS, reciprocal translocation occurs between chromosomes 11 [where the Fli-1 proto-oncogene, ETS transcription factor (FLI1) gene exists] and 22 (where the EWS gene exists) t(11, 22)(q24; q12) resulting in an EWS–FLI1 fusion gene. This abnormality is found in over 85% of diagnosed EWS. Other reciprocal translocations t(21; 22)(q22; q12) were also observed, giving rise to an EWS–ERG fusion gene or t(7; 22), t(17; 22) and t(2; 22), and inv(22). In all cases, proteins with abnormal function result in abnormal cell proliferation through continuous activation of insulin-like growth factor 1 (IGF1) [96].

⧉ Conclusions

Congenital solid tumors are heterogeneous containing several cell types, but the initiation and continuous growth usually depend on a single population of neoplastic cells and carry the same genetic or epigenetic abnormality. Genetic risk factors for cancer are incriminated in both adults and children, while environmental factors, incriminated in adults, are insignificant in children. The genetic conditions that can anticipate the neoplasm are diverse and they must be identified as early as possible, in order to monitor them in order to capture, as soon as possible, the neoplastic process. Tumor progression is achieved by certain mutations that activate oncogenes and suppress the activity of tumor suppressor genes. Different combinations of mutations may occur at random in patients with the same form of cancer, so the medical treatment needs to be targeted to that mutation. Thus, the understanding and identification of molecular mechanisms that take place in various forms of neoplasms are the basis for identifying molecules that intervene targeted and customized in the therapy of this extremely complicated disease. The identification and monitoring of these cases involves a multidisciplinary team: pediatrician, geneticist, oncologist, surgeon, chemo- and radiotherapist, psychologist.

Conflict of interest

The authors declare that they have no conflict of interests.