What is the pattern of inheritance for retinoblastoma?

Retinoblastoma is the most common childhood intraocular malignancy that affects one or both eyes.1 Because of the strong links between clinical care and genetic causation,2 retinoblastoma is considered the prototype of heritable cancers.3 Worldwide, about 8000 children are newly diagnosed with retinoblastoma every year (1/16,000 live births)1,4 but most have no access to knowledge of the important role genetics plays in many aspects of retinoblastoma: clinical presentation, choice of treatment modalities, and follow-up for both child and family. We now highlight the genetic etiology of retinoblastoma in the context of individual children and families, led by the questions commonly asked by parents.

WHAT IS RETINOBLASTOMA?

Retinoblastoma is a cancer that arises because both copies of the RB1 gene that normally suppresses retinoblastoma are lost from a developing retinal cell in fetuses, babies, and young children. Retinoblastoma can affect one (unilateral) or both eyes (bilateral) and, in 5% of children with heritable retinoblastoma (H1),5 is associated with a midline brain tumor (trilateral).6 Without timely and effective treatment, retinoblastoma may spread through the optic nerve to the brain, or via blood, particularly to bone marrow, and result in death.

HOW CAN CANCER OCCUR AT SUCH A YOUNG AGE?

The cell of origin of retinoblastoma is most likely a developing cone photoreceptor precursor cell that has lost both alleles of the RB1 tumor suppressor gene and remains in the inner nuclear layer of the retina (Fig. 1), perhaps because it is unable to migrate to the outer retina and function normally.1,7,8 The cell susceptible to developing cancer is present in the retinas of young children, from before birth, up to around 7 years of age. Rarely, retinoblastoma is first diagnosed in older persons, who likely previously had an undetected small tumor (retinoma) present from childhood that later became active.9,10 In high-income countries, the mean age at presentation is 12 months in bilateral disease and 24 months in unilateral disease, whereas significant delay in low-income countries impacts negatively on the children.11

FIGURE 1.:

Early awareness of risk for retinoblastoma optimizes therapy and outcomes. This child was examined because his sibling (triplets) presented with retinoblastoma. His right eye appeared normal, but optical coherence tomography (OCT) revealed 2 invisible tumors (* and ↓) (A-C), which were treated with laser therapy only; OCT after laser shows coagulation of tumors, visible in the retinal image (D-F).

WHAT CAUSES RETINOBLASTOMA?

No one knows what really causes the genomic damage to the RB1 gene, but retinoblastoma arises at a constant rate in all races irrespective of local environment. In nearly 50% of patients, the first copy of the RB1 gene is damaged in most, or all, normal cells of the patient. A retinal tumor develops when the second copy of the RB1 gene is also damaged in a developing retinal cell.1 The RB1 gene resides on chromosome 13q14 and encodes the retinoblastoma protein (pRB), an important regulator of cell division cycle in most cell types, and the first tumor suppressor gene discovered.12 Normally, cell division is inhibited by hypophosphorylated pRB binding to E2F molecules and blocking their transactivation of RB1, E2F, and other promoters of molecules that support cell division.13-15 To resume cell division, cyclindependent kinases re-phosphorylate pRB, activating promoters of key proteins important in cell division.16 Loss of pRB therefore allows uncontrolled cell division. In many cell types, loss of the RB1 gene is compensated by increased expression of other related proteins. However, in susceptible cells such as retinal cone cell precursors, compensatory mechanisms are insufficient, allowing uncontrolled cell division to initiate cancer.8

WHAT CAUSES RETINOBLASTOMA TO BE UNILATERAL VERSUS BILATERAL?

In heritable retinoblastoma (also called germline retinoblastoma), the first RB1 allele is mutated (M1) in nearly all cells, including germline reproductive cells, whereas the second allele is mutated (M2) in the retinal cells that become cancer, usually in both eyes (Figs. 2A, B). The most common M2 event is loss of the normal RB1 allele and duplication of the mutated M1 allele, in that 2 copies of the mutated RB1 gene remain; a mutational event referred to as loss of heterozygosity (LOH).17,18 Heritable retinoblastoma encompasses 45% of all reported cases19-21 with bilateral (80%), unilateral (15%), or trilateral (5%) tumors.1 Germline RB1 mutations carry the risk of second primary cancers, most commonly leiomyosarcoma, osteosarcoma, and melanoma.22 These patients may benefit from regular surveillance for such cancers over their lifetime.

FIGURE 2.:

Pedigrees illustrating inheritance patterns for RB1 mutations. A, Full penetrance and expressivity for null mutation: father (bilaterally enucleated) and 2/2 bilaterally affected offspring (I-1 unilateral enucleation vision 1.0 remaining eye); all H1 with 11 base pair deletion in exon 12; DER = 2. B, No family history, new null mutation: triplets1 diagnosed with bilateral retinoblastoma at age 2.5 months due to c.1345G>T (p.Gly449Ter) RB1 missense mutation resulting in no pRB; parents showed no evidence of the mutation but were considered H0* because there remains a less than 1% risk of mosaicism in either parent; older sibling is negative for the mutation, therefore H0; DER = 2. C, 100% penetrance and variable expressivity: grandfather (I-1) was diagnosed with bilateral retinoma when his daughter (II-1) was diagnosed with bilateral retinoblastoma due to c.1960G>T (p.Val654Leu) RB1 missense mutation; she (II-1) later developed meningioma in the radiation field and breast cancer; her brother (II-3) and daughter (III-2) inherited the mutation and developed unilateral retinoblastoma; DER = 1.5. D, Parent of origin low penetrance62: c.607+1G>T RB1 splice mutation that shows higher penetrance when inherited from the father (ie, II-1, III-1, III-3; DER = 1) than from the mother (ie, III-1, III-3; DER = 0), likely due to increased expression from the maternal than the paternal mutant RB1 allele60 (overall DER = 0.7); IV-1 had a small unilateral tumor but died at 11 years of age due to radiation-induced secondary malignancies; IV-5 has not been tested but developed thyroid cancer.

Of nonheritable retinoblastoma, 98% have both RB1 M1 and M2 events within a retinal cell. In the remaining 2%, retinoblastoma is induced by somatic amplification of the MYCN oncogene, in the presence of normal RB1 genes.23

WHAT CAUSED THESE MUTATIONS? DID I CAUSE THEM?

No one is to blame for the mutations causing retinoblastoma. Many environmental forces induce DNA damage, including cosmic rays, X-rays, DNA viruses, ultraviolet irradiation, and smoking. The DNA damage may be point mutations, small and large deletions, promotor methylation shutting down RB1 expression, and rarely, chromothripsis.24,25 The majority of RB1 mutations arise de novo, unique to a specific patient or family (Fig. 2B). However, some recurrent mutations are found in unrelated individuals, such as those that affect 11 hyper-mutable CpG DNA sequence sites, which make up 22% of all RB1 mutations.26,27

A de novo RB1 germline mutation may arise either pre- or post-conception. Pre-conception mutagenesis of RB1 usually occurs during spermatogenesis, perhaps because cell division (an opportunity for mutation) is very active during spermatogenesis but not during oogenesis.28,29 Advanced paternal age increases risk for retinoblastoma,30 suggesting that genomic errors may increase in aging men. The affected child carries the de novo RB1 mutation in every cell, typically presenting with 4-5 tumors and bilateral retinoblastoma. In contrast, if mutagenesis occurs postconception, during embryogenesis, only a portion of cells will carry the RB1 mutation (ie, mosaicism).1

DOES ONLYRB1MUTATION CAUSE RETINOBLASTOMA?

There are 2 answers to this question: (i) loss of both RB1 alleles only causes retinoma, a benign precursor to retinoblastoma, and other genes are modified to cause progression to cancer;10 and (ii) 2% of unilateral retinoblastoma have normal RB1 and are caused by a different gene.

(i) Retinoma is a premalignant precursor to retinoblastoma with characteristic clinical features: translucent white mass, reactive retinal pigment epithelial proliferation, and calcific foci.9 Retinoma may be found retrospectively after a child develops retinoblastoma (Figs. 2C, 3). Pathology of retinoma reveals nonproliferative fleurettes.10,31 Comparison of adjacent normal retina, retinoma, and retinoblastoma showed loss of both RB1 alleles and early genomic copy number changes in retinoma that were amplified further in the adjacent retinoblastoma.10 Many retinoblastomas have underlying elements of retinoma. Retinoma can transform to retinoblastoma even after many years of stability, so they need lifelong monitoring. The alternate, confusing term “retinocytoma” was proposed 1 year later32 and is inappropriate because it had been used for several different entities, including active tumors with Flexner-Wintersteiner rosettes.33-36 In addition to loss of RB1, specific alterations in copy number of other genes are common in RB1-/- retinoblastoma. There are gains (4-10 copies) in oncogenes MDM4, KIF14 (1q32), MYCN (2p24), DEK, and E2F3 (6p22) and loss of the tumor suppressor gene CDH11 (16q22-24).3,37 Other less common genomic alterations in retinoblastoma tumors include differential expression of specific microRNAs,38 recurrent single nucleotide variants/insertion-deletions in the genes BCOR and CREBBP,39 and upregulation of spleen tyrosine kinase (SYK).40 In comparison with the genomic landscape of other cancers, retinoblastoma is one of the least mutated, and epigenetic modification of gene expression may play an important role in retinoblastoma progression.39-41SYK is a postulated methylating proto-oncogene required for retinoblastoma cell survival. It was found to be upregulated in retinoblastoma and suggested as a potential target of therapy.40,42 It was present in 100% of retinoblastomas and 0% of normal retina histologically by immunohistochemical staining of pathological specimens of enucleated eyes with retinoblastoma.43

FIGURE 3.:

Retinoma, the benign precursor of retinoblastoma. Bilateral stable translucent retinal masses with calcification and associate retinal pigment epithelial disturbance were diagnosed to be retinoma in the grandparent I-1 (Fig. 2C), only when his daughter (II-1) was diagnosed with bilateral retinoblastoma. He never received treatment and was alive and well at age 90 at last follow-up.

(ii) There is a newly recognized form of retinoblastoma with normal RB1 genes. Two percent of unilateral patients have RB1+/+ MYCNA tumors, in which the MYCN oncogene is amplified (28-121 DNA copies instead of the normal 2 copies).23 These children are diagnosed at a median age of 4.5 months compared with 24 months for nonheritable unilateral RB-/- patients, and the tumors are histologically distinct with advanced features at diagnosis.

COULD WE HAVE DISCOVERED RETINOBLASTOMA EARLIER?

The only way to find retinoblastoma tumor early is to examine the eye with specific expertise, which we cannot do for every child. The earliest signs of retinoblastoma detectable by parents are leukocoria (white pupil), either directly or in photographs (photo-leukocoria), and strabismus when the macula is involved by tumor. Facial features and various degrees of hypotony and mental retardation in 13q deletion syndrome can lead to discovery of retinoblastoma.44,45 If we examine the retina of patients with positive familial history early, we may discover the smallest visible tumors that are round, white retinal lesions that obscure the underlying choroidal pattern; invisible definitive tumors can be detected even earlier by optical coherence tomography (Fig. 1).46,47

Centrifugal growth results in small tumors being round; more extensive growth produces lobular growth, likely related to genomic changes in single (clonal) cells that provide a proliferative advantage.48 Next, tumor seeds spread out of the main tumor into the subretinal space or vitreous cavity. Vitreous seeding is associated with loss of chromosome 16q, including the cadherin 13 gene, a genomic change that may induce poor cohesive forces between tumor cells.49,50

Advanced vitreous tumor seeds can migrate to the anterior chamber producing a pseudohypopyon. Enlarging tumor can push the iris lens diaphragm forward causing angle closure glaucoma. Advanced tumors may induce iris neovascularization. Rapid necrosis of tumor can cause an aseptic orbital inflammatory reaction resembling orbital cellulitis, sometimes showing central retinal artery occlusion on pathology.48,51 Untreated, retinoblastoma spreads into the optic nerve and brain, or hematogenous spread occurs through choroid, particularly to grow in bone marrow. Tumor extension through sclera can present as orbital extension and proptosis.

DO ALL AFFECTED INDIVIDUALS WITHRB1MUTATIONS DEVELOP RETINOBLASTOMA?

Depending on the exact RB1 mutation, most, but not all, carriers of an RB1 mutation will develop retinoblastoma and other cancers throughout life.

Each offspring of a person carrying an RB1 mutated gene has 50% risk to inherit the RB1 mutated gene (Fig. 2). A measure of expressivity of a mutant retinoblastoma allele is the diseaseeye ratio (DER) (number of eyes affected with tumor divided by the number of carriers of the RB1 mutation).52 Nonsense and frame-shift germline mutations that lead to absent or truncated dysfunctional pRB result in 90% bilateral retinoblastoma (nearly complete penetrance, DER = 2). Partially functional RB1 mutated alleles show variable penetrance and expressivity (Fig. 2C) with fewer tumors and later onset,47 and some carriers never develop retinoblastoma (DER = 1-1.5). Reduced penetrance mutated RB1 alleles include (i) mutations in exons 1 and 2,53 (ii) mutations near the 3’ end of the gene in exons 24 to 27,54,55 (iii) splice and intronic mutations,56 and (iv) missense mutations.57 Counterintuitively, large deletions encompassing the RB1 gene and MED4 gene also cause reduced expressivity/penetrance because RB1-/- cells cannot survive in the absence of MED458 and the most common M2 event in retinoblastoma tumors is loss of the normal allele and reduplication of the mutated allele. In comparison, patients with large deletions with 1 breakpoint inside the RB1 gene typically present with bilateral disease.59

There are at least 2 specific RB1 mutant alleles showing a parent-of-origin effect: c.607+1G>T substitution60,61 (Fig. 2D) and c.1981C>T (p.Arg661Trp) missense mutation.62 Both may be explained by differential methylation of intron 2 CpG85, which skews RB1 expression in favor of the maternal allele.63 When the mutated allele is maternally inherited, there is sufficient tumor suppressor activity to limit retinoblastoma development to 10% of carriers. However, when the mutated RB1 allele is paternally transmitted, very little pRB is expressed, leading to retinoblastoma in 68% of carriers.

WHAT FACTORS AFFECT RETINOBLASTOMA CANCER STAGING?

Treatment and prognosis of retinoblastoma depend on the stage of disease at initial presentation. Factors predictive of outcomes include size, location of tumor origin, extent of subretinal fluid, presence of tumor seeds, and the presence of high-risk features on pathology.5 Multiple staging systems have predicted likelihood to salvage an eye without using radiation therapy, but published evidence is confusing because significantly different staging versions have been used.1,5 The 2017 TNMH classification is based on an international consensus and evidence from an international survey of 1728 eyes and separates well initial clinical and pathological features predictive of outcomes to save the eye, in retrospective comparison with 5 previous eye staging systems.5

DOES GERMLINE STATUS AFFECT RETINOBLASTOMA STAGING?

Retinoblastoma is the first cancer in which staging recognizes the impact of genetic status on outcomes in the 2017 TNMH staging (Fig. 4): presence of a positive family history, bilateral or trilateral disease, or high sensitivity positive RB1 mutation testing is stage H1; without these features before testing blood, HX; and H0 for those relatives shown to not carry the proband’s specific RB1 mutation (Fig. 2).5 We propose H0* for patients tested and having neither M1 nor M2 RB1 mutated alleles of the tumor detectable in blood and parents showing no evidence of the M1 allele of their offspring, but with remaining low risk (<1%) of undetectable mosaicism. Offspring of H0* persons who do not show their family’s RB1 mutation are H0; if they test H1 for the inherited mutation they need intensive surveillance for eye tumors from birth.

FIGURE 4.:

Eighth edition TNMH cancer staging for intraocular retinoblastoma.54 Clinical staging for eyes with retinoblastoma, compared with IIRC (A).38 Heritability stage for persons at risk to carry an RB1 germline mutation (B). H0* is our proposal to clarify the remaining small risk of undetectable mosaicism in parents of H1 patients who test negative for the RB1 pathogenic variant in their H1 child and in unilateral patients with no RB1 pathogenic variant detected in blood.

WHAT ARE THE MAIN GOALS OF RETINOBLASTOMA MANAGEMENT?

Saving life is the top priority of retinoblastoma treatment, followed by vision salvage; least important is eye salvage. The child’s job is to play and develop over a healthy life; many procedures and their complications that may span years for at best a 50% chance to save a blind eye with risk of tumor spread are not justified, especially when the other eye is normal.64,65 Multiple treatment modalities are currently present with promising tumor control in the short term (5 years) with minimal data on their long-term effects. Only after 30 years of being used on every child, the enormous impact of external beam irradiation (EBRT) was recognized: more children with bilateral retinoblastoma who were treated with radiation die of their second (or third, and so on) cancer than of retinoblastoma.66,67

However, often missing from choices in the complex care of children with retinoblastoma are truly informed parents. Essentially, the doctors decide what treatment they “feel” is best, based on little substantial evidence. There currently exists no easy way to show the parents prospectively the true “costs” of each treatment: the burden of invasive therapies and potential complications; the imposition of hours and days in hospitals and feeling ill on the child, whose real job in those critical, irreplaceable years is to play; the true costs including time off work and uncertainties; and the burden of “false hope” in the absence of real evidence. There are imminent solutions on the horizon: DePICTRB (depictrb. technainstitute.com) collects retinoblastoma-specific details including genetic results and makes them viewable online by the family and patient68 and the burgeoning field of patient-reported outcomes for research with accurate patient details. These new attitudes and tools may empower, in the future, well-informed choices by parents for their child and family.

WHAT ARE THE TREATMENTS AND WHAT GOVERNS THE CHOICE?

Choice of treatment depends on the tumor stage, genetic status, and laterality of disease. Focal therapy can only control cT1a eyes, but visually threatening or large cT1b tumors and cT2 eyes need chemotherapy (systemic or intra-arterial chemotherapy) to reduce the size of the tumor followed by consolidation focal therapies (laser therapy or cryotherapy) as initial treatment. Enucleation of eyes with advanced tumors (cT3) is usually a definitive cure, and if high risk features are found, appropriate treatment response has a chance for cure.1 Ancillary therapies for specific indications include plaque radiotherapy and periocular chemotherapy. Intravitreal chemotherapy for vitreous disease has recently dramatically improved safe eye salvage.69 Retinoblastoma with extraocular spread (cT4) requires more aggressive treatments with high dose systemic chemotherapy, surgery, and EBRT with or without autologous stem cell transplantation. For persons carrying RB1 mutations, EBRT is rarely indicated due to the high risk of inducing later second cancers.1

Laterality at presentation also directs treatment choices. In unilateral retinoblastoma, enucleation is the preferred definitive treatment modality if there is anticipated poor visual outcome of the affected eye.1 If useful vision is anticipated, chemotherapy (systemic or intra-arterial) combined with focal therapy might be utilized. In bilateral retinoblastoma, systemic chemotherapy followed by consolidation focal therapy is preferred when both eyes are salvageable (eighth edition TNMH cancer staging5 cT1b, cT2) [International Intraocular Retinoblastoma Classification (IIRC)48 B, C, D eyes]. When one eye is cT1a and the other is advanced, enucleation of the advanced eye and focal therapy for the cT1a eye might be more appropriate. When both eyes are nonsalvageable, bilateral enucleation is the definitive treatment to save life and minimize morbidity.

IS RETINOBLASTOMA LETHAL?

Untreated, retinoblastoma is lethal. If treated before metastasis occurs, the cure rate is nearly 100%. Occasionally, metastatic retinoblastoma may be confused with a second cancer; molecular demonstration of the RB1 mutations of the intraocular retinoblastoma will confirm metastases.70 Globally, the chance for cure is remote, and lack of knowledge of genetics results too often in the death of children who could have been saved if they had surveillance and definitive treatment when tumors were small. We look forward to the cancer genomic revolution focusing on metastatic retinoblastoma, leading to breakthrough drugs that make retinoblastoma a zero death cancer. However, delayed diagnosis and treatment due to lack of knowledge by ophthalmologists and parents, along with socioeconomic64 and cultural factors are the major causes of mortality. Asia and Africa have more than 70% mortality from retinoblastoma compared with less than 5% in developed countries.71

HOW CAN WE TEST FOR RETINOBLASTOMA MUTATIONS?

Genetic counseling is the psychosocial and educational process to help patients and families adapt to the genetic risk, the genetic condition, and the process of informed decision-making.72,73 Especially when genetic testing is not available or unaffordable, genetic counseling is very important. Concrete knowledge of genetic test outcomes supports informed and precise genetic counseling. Genetic testing supports precision medicine for those who need it (carry the RB1 mutated allele) and is more cost effective than examining all at-risk family members.27,74

If the proband is bilaterally affected, the probability of finding a germline mutation in the RB1 gene in DNA extracted from blood is high (97% in a comprehensive RB1-specialized laboratory). In 3% of bilateral retinoblastoma patients, the predisposing RB1 mutation cannot be detected in blood. In these instances, identification of M1 and M2 RB1 mutations in DNA from tumor can assist in the identification of a germline mutation, including low-level mosaic mutations.26,75,76

Similarly, to detect the 15% of unilateral patients carrying a germline mutation, the optimal strategy first tests tumor DNA and then looks for the mutations in blood. If the blood is not found to carry 1 of the tumor RB1 mutations, risk of germline status is reduced to less than 1% (Table 1) for parents, siblings, and cousins.76 Quality of genetic results depends on quality of DNA. Fresh or frozen tumor samples are ideal, whereas DNA from formalin fixed paraffin embedded tumors is suboptimal. Whole blood in EDTA or ACD anticoagulant typically provides high quality DNA.

TABLE 1:

Risks for Probands and Relatives to Develop Retinoblastoma and Second Cancers and Clinical Surveillance Plans85

The RB1 gene can be mutated in many ways, best identified by a series of techniques. The most appropriate technology depends on the clinical question being asked. Next-generation sequencing (NGS) may be the most effective screening strategy to find an unknown de novo mutation in an affected proband and may detect lower level mosaic mutations.77,78 To screen family members for a known sequencing-detectable RB1 mutation, targeted Sanger dideoxy sequencing is still more cost and time effective.

Large RB1 deletions or duplications that span whole exons or multiple exons typically cannot be detected by DNA sequencing. Multiplex ligation-dependent probe amplification (MLPA), quantitative multiplex polymerase chain reaction (QM-PCR), or array comparative genomic hybridization (aCGH) are used to identify RB1 deletions and duplications and other genomic copy number alterations, such as MYCN amplification. New developments in bioinformatic analysis methods suggest that NGS data can be interrogated for copy number variants,77,79 but sensitivity is not yet optimal. Somatic mosaicism can arise in either the presenting patient or their parent. Allele-specific PCR (AS-PCR) has excellent sensitivity when the RB1 mutation is known26 and can detect as low as 1% mosaicism.

Polymorphic microsatellite markers distributed throughout chromosome 13 can be used to detect a change from a heterozygous state in blood compared with the homozygous state in a tumor with LOH. Microsatellite marker analysis is also important in identity testing and in identifying maternal cell contamination in prenatal samples.

Epigenetic changes can also initiate retinoblastoma development.80 Hypermethylation of the RB1 promoter CpG island results in inhibition of RB1 gene transcription in 10-12% of retinoblastoma tumors, commonly involving both alleles.27,81 This epigenetic gene-silencing event primarily occurs in somatic cells, but heritable RB1 promoter mutations and translocations disrupting RB1 regulatory sites or translocations involving the X chromosome have been shown to cause constitutional RB1 promoter hypermethylation.82,83

Rarely, no RB1 mutation is identified in the coding, promoter, or flanking intronic sequence in blood from a bilateral patient. Deep intronic sequencing alterations that disrupt RB1 transcription by interfering with correct splicing in patients with retinoblastoma can be detected by analysis of the RB1 transcript by reverse transcriptase PCR (RT-PCR).56,84 RNA studies can also clarify pathogenicity of intronic sequence alterations. As costs decrease, whole genome sequencing may become the method of choice to uncover deep intronic changes.

Karyotype, fluorescent in situ hybridization (FISH), or array comparative genomic hybridization (aCGH) of peripheral blood lymphocytes can be used to identify large deletions and rearrangements in retinoblastoma patients, including patients suspected of 13q14 deletion syndrome.59 In parents of 13q14 deletion patients, karyotype analysis can be used to investigate for balanced translocations, which increase the risk for retinoblastoma in subsequent generations.44

WHAT IS DONE AFTER FINDING THERB1MUTATION?

Family members at risk to also carry the identified RB1 mutation are offered testing on blood samples (Fig. 2, Table 1).1,76,85 If the proband’s mutation is not found in either parent, a risk for low level mosaicism still exists. Offspring of any family member carrying the RB1 mutation can be tested during pregnancy or immediately after birth. If the proband is mosaic for the RB1 mutation, parents and siblings are not at risk, as mosaicism cannot be inherited. The children of a mosaic proband need to be tested as early as possible because if they do inherit the RB1 mutated allele, they will usually be at high risk of bilateral disease. If they do not carry the mutation, they are at population risk for retinoblastoma.

HOW DOES THERB1STATUS AFFECT FOLLOW-UP OF THE CHILD?

If the child does not carry an RB1 germline mutation, no examinations under anesthesia (EUAs) are required for the unaffected eye with follow-up only in the clinic (Table 1). If the child carries a germline mutation or is mosaic for the RB1 mutation, the possibility of new tumors renders frequent EUAs essential up to at least 3 years of age.86 The germline status also predisposes to other second malignant neoplasms, requiring frequent annual surveillance. The use of whole body magnetic resonance imaging as a method for surveillance is controversial, as a high false positive rate may provoke unnecessary invasive procedures. Development of surveillance programs beyond the pediatric age is a high priority for early detection of these often-lethal second RB1-induced cancers.67

CAN WE USE THE KNOWN MUTATION TO TEST MY FUTURE CHILDREN?

Prenatal genetic testing can be performed in the course of the pregnancy. Two early procedures are available: (i) chorionic villus sampling (CVS), performed between 11-14 weeks gestation, involves obtaining a sample of the placenta either transvaginally or transabdominally; and (ii) amniocentesis after 16 weeks gestation involves sampling amniotic fluid transabdominally. The procedure-associated risk of miscarriage with CVS is 1%, whereas amniocentesis is 0.1-0.5%. Maternal cell contamination is more frequent with CVS87 and is routinely assessed by the clinical molecular lab.

If the fetus does not carry the mutation, the pregnancy can proceed with no further intervention. If the fetus carries the familial mutation, the parents have several choices. Some may decide to stop the pregnancy, whereas others may decide to continue the pregnancy regardless of test results. If the parents are concerned by the risk of miscarriage, they can consider late amniocentesis between 30-34 weeks gestation when the major complication is early delivery rather than miscarriage.87 Prenatal or postnatal RB1 mutation testing will either show the baby to be H0 (no family RB1 mutation) or H1 (confirmed to carry the mutation). Early preterm delivery of an H1 baby achieves smaller tumors with higher treatment success, eye preservation, and visual outcome than delivery at full term (Fig. 1).47

In many countries, the option for prenatal genetic testing is not available, and some parents may choose not to do prenatal invasive testing. All offspring of an H1 parent have a 50% risk for retinoblastoma, so it is important that the pregnancy does not go past 40 weeks (Table 1, Fig. 5).47,76

FIGURE 5.:

Postnatal diagnosis of familial retinoblastoma too late to save good vision. Bilateral inherited retinoblastoma in a neonate with positive family history born full term (39 weeks) and examined at 5 days after birth: right eye stage cT2b (IIRC group C), left eye stage cT1b (IIRC group B) involving both foveas. Treatment with multiple periocular and systemic chemotherapies and 3 years of focal therapies salvaged both eyes with legal blindness. The child has been previously published (patient #2).47

CAN WE PLAN OUR NEXT PREGNANCY TO AVOID PASSING ON THISRB1MUTATION?

In some countries, preimplantation genetic diagnosis (PGD) with in vitro fertilization is an option.88 Conceptions are tested for the familial mutation at an early stage of embryonal development (typically at the 8 cell stage). Those that do not carry the RB1 mutation are implanted. The procedure is costly, ranging from $10,000-$15,000 Canadian dollars per cycle; in some countries, there may be full or partial coverage of the costs. The full medical implications of PGD are not yet understood; there is emerging evidence of a low risk for epigenetic changes in the conception as a result of the procedure. It is recommended that typical prenatal testing be pursued during the course of the pregnancy to confirm the results.88

ARE THESE TESTS AVAILABLE WORLDWIDE?

High-sensitivity RB1 mutation testing is established in core labs mainly in high-income countries.89 In low- and middleincome countries, genetic counseling as a specialty does not exist, so many families with retinoblastoma children do not have the opportunity to understand the benefits of genetic testing and counseling in retinoblastoma treatment and follow-up.1 In many places, existing health insurance does not cover costs of genetic testing. Given all these obstacles, there is limited application of genetic testing and genetic counseling worldwide, which leads to increased health care to deal with late diagnosis, along with increased economic burden on affected families. As an example, the Chinese government’s new policy to allow families to have additional children will make the need for genetic testing and counseling even more important.

In Egypt, genetic testing for retinoblastoma is not available, therefore, genetic counseling is the only way to address the issues facing retinoblastoma families.90 Ophthalmologists with insufficient training in retinoblastoma genetics are burdened with this task. Genetic counseling was found to decrease knowledge gaps remaining in the translation of this knowledge into appropriate screening action.

CONCLUSIONS

Genetics is a core element in care of retinoblastoma patients and their families. Action based on knowledge of genetics in retinoblastoma improves outcomes for the eye and points to life and treatment strategies that alleviate suffering and may reduce early mortality.90 The whole family benefits economically in health by precision in diagnosis of risk, based on genetic testing. Treatments accommodate the risks of therapies, and surveillance protocols can achieve earlier diagnosis of retinoblastoma and second cancers, leading to better outcomes. Knowledge of test results alleviates psychological burden on families moving forward with their life choices for the affected child and future siblings and exposure to environmental carcinogens for the whole family.

REFERENCES

1. Dimaras H, Corson TW, Cobrinik D, et al. Retinoblastoma. Nat Rev Dis Primers. 2015:15021.

2. Knudson AG. Mutation and cancer: statistical study of retinoblastoma. Proc Natl Acad Sci U S A. 1971;68:820-823.

3. Theriault BL, Dimaras H, Gallie BL, et al. The genomic landscape of retinoblastoma: a review. Clin Exp Ophthalmol. 2014;42:33-52.

4. Seregard S, Lundell G, Svedberg H, et al. Incidence of retinoblastoma from 1958 to 1998 in Northern Europe: advantages of birth cohort analysis. Ophthalmology. 2004;111:1228-1232.

5. Mallipatna A, Gallie BL, Chévez-Barrios P, et al. Retinoblastoma. In: Amin MB, Edge SB, Greene FL, eds. AJCC Cancer Staging Manual. 8th ed. New York, NY: Springer; 2017:819-831.

6. de Jong MC, Kors WA, de Graaf P, et al. Trilateral retinoblastoma: a systematic review and meta-analysis. Lancet Oncol. 2014;15:1157-1167.

7. Rootman DB, Gonzalez E, Mallipatna A, et al. Hand-held high-resolution spectral domain optical coherence tomography in retinoblastoma: clinical and morphologic considerations. Br J Ophthalmol. 2013;97:59-65.

8. Xu XL, Singh HP, Wang L, et al. Rb suppresses human cone-precursorderived retinoblastoma tumours. Nature. 2014;514:385-388.

9. Gallie BL, Ellsworth RM, Abramson DH, et al. Retinoma: spontaneous regression of retinoblastoma or benign manifestation of the mutation? Br J Cancer. 1982;45:513-521.

10. Dimaras H, Khetan V, Halliday W, et al. Loss of RB1 induces nonproliferative retinoma: increasing genomic instability correlates with progression to retinoblastoma. Hum Mol Genet. 2008;17:1363-1372.

11. Nyamori JM, Kimani K, Njuguna MW, et al. The incidence and distribution of retinoblastoma in Kenya. Br J Ophthalmol. 2012;96:141-143.

12. Friend SH, Bernards R, Rogelj S, et al. A human DNA segment with properties of the gene that predisposes to retinoblastoma and osteosarcoma. Nature. 1986;323:643-646.

13. Nevins JR. The Rb/E2F pathway and cancer. Hum Mol Genet. 2001;10: 699-703.

14. Cobrinik D. Pocket proteins and cell cycle control. Oncogene. 2005;24: 2796-2809.

15. Sage J, Cleary ML. Genomics: the path to retinoblastoma. Nature. 2012; 481:269-270.

16. Knudsen ES, Knudsen KE. Tailoring to RB: tumour suppressor status and therapeutic response. Nat Rev Cancer. 2008;8:714-724.

17. Dunn JM, Phillips RA, Zhu X, et al. Mutations in the RB1 gene and their effects on transcription. Mol Cell Biol. 1989;9:4596-4604.

18. Cavenee WK, Dryja TP, Phillips RA, et al. Expression of recessive alleles by chromosomal mechanisms in retinoblastoma. Nature. 1983;305:779-784.

19. MacCarthy A, Birch JM, Draper GJ, et al. Retinoblastoma in Great Britain 1963-2002. Br J Ophthalmol. 2009;93:33-37.

20. Moreno F, Sinaki B, Fandino A, et al. A population-based study of retinoblastoma incidence and survival in Argentine children. Pediatr Blood Cancer. 2014;61:1610-1615.

21. Wong JR, Morton LM, Tucker MA, et al. Risk of subsequent malignant neoplasms in long-term hereditary retinoblastoma survivors after chemotherapy and radiotherapy. J Clin Oncol. 2014;32:3284-3290.

22. MacCarthy A, Bayne AM, Brownbill PA, et al. Second and subsequent tumours among 1927 retinoblastoma patients diagnosed in Britain 1951-2004. Br J Cancer. 2013;108:2455-2463.

23. Rushlow DE, Mol BM, Kennett JY, et al. Characterisation of retinoblastomas without RB1 mutations: genomic, gene expression, and clinical studies. Lancet Oncol. 2013;14:327-334.

24. Lohmann DR. RB1 gene mutations in retinoblastoma. Hum Mutat. 1999;14: 283-288.

25. McEvoy J, Nagahawatte P, Finkelstein D, et al. RB1 gene inactivation by chromothripsis in human retinoblastoma. Oncotarget. 2014;5:438-450.

26. Rushlow D, Piovesan B, Zhang K, et al. Detection of mosaic RB1 mutations in families with retinoblastoma. Hum Mutat. 2009;30:842-851.

27. Richter S, Vandezande K, Chen N, et al. Sensitive and efficient detection of RB1 gene mutations enhances care for families with retinoblastoma. Am J Hum Genet. 2003;72:253-269.

28. Zhu XP, Dunn JM, Phillips RA, et al. Preferential germline mutation of the paternal allele in retinoblastoma. Nature. 1989;340:312-313.

29. Dryja TP, Morrow JF, Rapaport JM. Quantification of the paternal allele bias for new germline mutations in the retinoblastoma gene. Hum Genet. 1997;100:446-449.

30. Toriello HV, Meck JM. Statement on guidance for genetic counseling in advanced paternal age. Genet Med. 2008;10:457-460.

31. Tso MOM, Fine BS, Zimmerman LE. The nature of retinoblastoma. II. Photoreceptor differentiation: an electron microscopic study. Am J Ophthalmol. 1970;69:350-359.

32. Margo C, Hidayat A, Kopelman J, et al. Retinocytoma. A benign variant of retinoblastoma. Arch Ophthalmol. 1983;101:1519-1531.

33. Letarte F. [Considerations on a case of retinocytoma]. Laval Med. 1950;15: 900-904.

34. Miletto G. [New case of retinocytoma in an aged negro]. Med Trop (Madr). 1951;11:458-459.

35. Audibert J, Morel P. [Retinocytoma, puncture of the anterior chamber]. Bull Soc Ophtalmol Fr. 1960;2:87-88.

36. Saudax E, Ducret J. [Pseudo-hypoyon caused by retinocytoma]. Bull Soc Ophtalmol Fr. 1960;8:516-518.

37. Corson TW, Gallie BL. One hit, two hits, three hits, more? Genomic changes in the development of retinoblastoma. Genes Chromosomes Cancer. 2007;46: 617-634.

38. Huang JC, Babak T, Corson TW, et al. Using expression profiling data to identify human microRNA targets. Nat Methods. 2007;4:1045-1049.

39. Kooi IE, Mol BM, Massink MP, et al. Somatic genomic alterations in retinoblastoma beyond RB1 are rare and limited to copy number changes. Sci Rep. 2016;6:25264.

40. Zhang J, Benavente CA, McEvoy J, et al. A novel retinoblastoma therapy from genomic and epigenetic analyses. Nature. 2012;481:329-334.

41. Temming P, Corson TW, Lohmann DR. Retinoblastoma tumorigenesis: genetic and epigenetic changes walk hand in hand. Future Oncol. 2012;8: 525-528.

42. Murphree AL, Triche TJ. An epigenomic mechanism in retinoblastoma: the end of the story? Genome Med. 2012;4:15.

43. ElSabaa B, Soliman SE, Fadel S, et al. Evaluation of SYK protein expression in treated versus untreated retinoblastoma. Egyptian Journal of Pathology. 2016;36:1-5. In press.

44. Baud O, Cormier-Daire V, Lyonnet S, et al. Dysmorphic phenotype and neurological impairment in 22 retinoblastoma patients with constitutional cytogenetic 13q deletion. Clin Genet. 1999;55:478-482.

45. Bojinova RI, Schorderet DF, Addor MC, et al. Further delineation of the facial 13q14 deletion syndrome in 13 retinoblastoma patients. Ophthalmic Genet. 2001;22:11-18.

46. Soliman SE, VandenHoven C, MacKeen L, et al. Optical coherence tomography guided decisions in retinoblastoma management. Ophthalmology. March 15, 2017. [Epub ahead of print].

47. Soliman SE, Dimaras H, Khetan V, et al. Prenatal versus postnatal screening for familial retinoblastoma. Ophthalmology. 2016;123:2610-2617.

48. Murphree AL. Intraocular retinoblastoma: the case for a new group classification. Ophthalmol Clin North Am. 2005;18:41-53.

49. Gratias S, Rieder H, Ullmann R, et al. Allelic loss in a minimal region on chromosome 16q24 is associated with vitreous seeding of retinoblastoma. Cancer Res. 2007;67:408-416.

50. Gustmann S, Klein-Hitpass L, Stephan H, et al. Loss at chromosome arm 16q in retinoblastoma: confirmation of the association with diffuse vitreous seeding and refinement of the recurrently deleted region. Genes Chromosomes Cancer. 2011;50:327-337.

51. Valverde K, Pandya J, Heon E, et al. Retinoblastoma with central retinal artery thrombosis that mimics extraocular disease. Med Pediatr Oncol. 2002;38:277-279.

52. Lohmann DR, Brandt B, Hopping W, et al. Distinct RB1 gene mutations with low penetrance in hereditary retinoblastoma. Hum Genet. 1994;94: 349-354.

53. Sanchez-Sanchez F, Ramirez-Castillejo C, Weekes DB, et al. Attenuation of disease phenotype through alternative translation initiation in lowpenetrance retinoblastoma. Hum Mutat. 2007;28:159-167.

54. Bremner R, Du DC, Connolly-Wilson MJ, et al. Deletion of RB exons 24 and 25 causes low-penetrance retinoblastoma. Am J Hum Genet. 1997;61: 556-570.

55. Mitter D, Rushlow D, Nowak I, et al. Identification of a mutation in exon 27 of the RB1 gene associated with incomplete penetrance retinoblastoma. Fam Cancer. 2009;8:55-58.

56. Zhang K, Nowak I, Rushlow D, et al. Patterns of missplicing caused by RB1 gene mutations in patients with retinoblastoma and association with phenotypic expression. Hum Mutat. 2008;29:475-484.

57. Scheffer H, Van Der Vlies P, Burton M, et al. Two novel germline mutations of the retinoblastoma gene (RB1) that show incomplete penetrance, one splice site and one missense. J Med Genet. 2000;37:E6.

58. Dehainault C, Garancher A, Castera L, et al. The survival gene MED4 explains low penetrance retinoblastoma in patients with large RB1 deletion. Hum Mol Genet. 2014;23:5243-5250.

59. Mitter D, Ullmann R, Muradyan A, et al. Genotype-phenotype correlations in patients with retinoblastoma and interstitial 13q deletions. Eur J Hum Genet. 2011;19:947-958.

60. Klutz M, Brockmann D, Lohmann DR. A parent-of-origin effect in two families with retinoblastoma is associated with a distinct splice mutation in the RB1 gene. Am J Hum Genet. 2002;71:174-179.

61. Schuler A, Weber S, Neuhauser M, et al. Age at diagnosis of isolated unilateral retinoblastoma does not distinguish patients with and without a constitutional RB1 gene mutation but is influenced by a parent-of-origin effect. Eur J Cancer. 2005;41:735-740.

62. Eloy P, Dehainault C, Sefta M, et al. A parent-of-origin effect impacts the phenotype in low penetrance retinoblastoma families segregating the c.1981C>T/p.Arg661Trp mutation of RB1. PLoS Genet. 2016;12:e1005888.

63. Kanber D, Berulava T, Ammerpohl O, et al. The human retinoblastoma gene is imprinted. PLoS Genet. 2009;5:e1000790.

64. Soliman SE, Dimaras H, Souka AA, et al. Socioeconomic and psychological impact of treatment for unilateral intraocular retinoblastoma. J Fr Ophtalmol. 2015;38:550-558.

65. Soliman SE, Gallie BL, Shaikh F. Intra-arterial chemotherapy for retinoblastoma-reply. JAMA Ophthalmol. 2016;134:1203.

66. Eng C, Li FP, Abramson DH, et al. Mortality from second tumors among longterm survivors of retinoblastoma. J Natl Cancer Inst. 1993;85:1121-1128.

67. Temming P, Arendt M, Viehmann A, et al. How eye-preserving therapy affects long-term overall survival in heritable retinoblastoma survivors. J Clin Oncol. 2016;34:3183-3188.

68. Chiu HH, Dimaras H, Downie R, et al. Breaking down barriers to communicating complex retinoblastoma information: can graphics be the solution? Can J Ophthalmol. 2015;50:230-235.

69. Munier FL, Gaillard MC, Balmer A, et al. Intravitreal chemotherapy for vitreous disease in retinoblastoma revisited: from prohibition to conditional indications. Br J Ophthalmol. 2012;96:1078-1083.

70. Racher H, Soliman S, Argiropoulos B, et al. Molecular analysis distinguishes metastatic disease from second cancers in patients with retinoblastoma. Cancer Genet. 2016;209:359-363.

71. Chantada GL, Qaddoumi I, Canturk S, et al. Strategies to manage retinoblastoma in developing countries. Pediatr Blood Cancer. 2011;56: 341-348.

72. Uhlmann WR. Response to Robert G. Resta commentary (Unprepared, understaffed, and unplanned: thoughts on the practical implications of discovering new breast and ovarian cancer causing genes). J Genet Couns. 2009;18:524-526.

73. Shugar A. Teaching genetic counseling skills: incorporating a genetic counseling adaptation continuum model to address psychosocial complexity. J Genet Couns. November 28, 2016. [Epub ahead of print].

74. Noorani HZ, Khan HN, Gallie BL, et al. Cost comparison of molecular versus conventional screening of relatives at risk for retinoblastoma. Am J Hum Genet. 1996;59:301-307.

75. Astudillo PP, Chan HS, Heon E, et al. Late-diagnosis retinoblastoma with germline mosaicism in an 8-year-old. J AAPOS. 2014;18:500-502.

76. Canadian Retinoblastoma Society. National Retinoblastoma Strategy Canadian Guidelines for Care: Strategie Therapeutique du Retinoblastome Guide Clinique Canadien. Can J Ophthalmol. 2009;44 (Suppl 2):S1-S88.

77. Li WL, Buckley J, Sanchez-Lara PA, et al. A rapid and sensitive nextgeneration sequencing method to detect RB1 mutations improves care for retinoblastoma patients and their families. J Mol Diagn. 2016;18:480-493.

78. Chen Z, Moran K, Richards-Yutz J, et al. Enhanced sensitivity for detection of low-level germline mosaic RB1 mutations in sporadic retinoblastoma cases using deep semiconductor sequencing. Hum Mutat. 2014;35:384-391.

79. Devarajan B, Prakash L, Kannan TR, et al. Targeted next generation sequencing of RB1 gene for the molecular diagnosis of retinoblastoma. BMC Cancer. 2015;15:320.

80. Ohtani-Fujita N, Fujita T, Aoike A, et al. CpG methylation inactivates the promoter activity of the human retinoblastoma tumor-suppressor gene. Oncogene. 1993;8:1063-1067.

81. Zeschnigk M, Lohmann D, Horsthemke B. A PCR test for the detection of hypermethylated alleles at the retinoblastoma locus. J Med Genet. 1999;36: 793-794.

82. Jones C, Booth C, Rita D, et al. Bilateral retinoblastoma in a male patient with an X; 13 translocation: evidence for silencing of the RB1 gene by the spreading of X inactivation. Am J Hum Genet. 1997;60:1558-1562.

83. Quinonez-Silva G, Davalos-Salas M, Recillas-Targa F, et al. Monoallelic germline methylation and sequence variant in the promoter of the RB1 gene: a possible constitutive epimutation in hereditary retinoblastoma. Clin Epigenetics. 2016;8:1.

84. Dehainault C, Michaux D, Pages-Berhouet S, et al. A deep intronic mutation in the RB1 gene leads to intronic sequence exonisation. Eur J Hum Genet. 2007;15:473-477.

86. Valenzuela A, Chan HSL, Heon E, et al. A language for retinoblastoma: guidelines through standard operating procedures. In: Reynolds J, ed. Pediatric Retina. New York: Marcel Dekker Inc; 2010:205-234.

87. Akolekar R, Beta J, Picciarelli G, et al. Procedure-related risk of miscarriage following amniocentesis and chorionic villus sampling: a systematic review and meta-analysis. Ultrasound Obstet Gynecol. 2015;45: 16-26.

88. Xu K, Rosenwaks Z, Beaverson K, et al. Preimplantation genetic diagnosis for retinoblastoma: the first reported liveborn. Am J Ophthalmol. 2004;137: 18-23.

89. One Retinoblastoma World. 2015. Available at: http://1rbw.org/. Accessed August 31, 2015.

90. Soliman SE, ElManhaly M, Dimaras H. Knowledge of genetics in familial retinoblastoma. Ophthalmic Genet. 2016:1-7. [Epub ahead of print].