Mohammad AskarPublic working copy · medical review pending

Paediatric intraocular tumours · 218

Childhood intraocular tumours: recognise, refer and protect life

Leukocoria and suspicious childhood ocular inflammation need specialist tumour assessment before intraocular intervention.

Public working copy · medical review pending. Source, translation and medical checks remain open. Read the limitations in the text and references. These educational materials do not replace individual medical assessment. Published: 2026-11-18T19:45:52+01:00. Internal project references lead to this notice. They are not publicly accessible primary sources; the source limitations remain stated in the text.

Stage the tumour, assess germline risk and select treatment in a multidisciplinary centre; eye preservation cannot take priority over survival.

Terminology and draft status

Terminology review is not complete. Selected actual WHO UMD queries and their limits are recorded in terminology.md; they do not approve the whole translation.

01Recognise the signal and refer safely

Retinoblastoma is the commonest primary intraocular malignancy in childhood. The original “You never walk alone” describes joint ophthalmic oncology, paediatric oncology, genetics and specialist radiology care, not an examination-board rule. Delay can threaten survival and the eye. 1 2

Leukocoria may be noticed on photographs, whose processing can alter the visible reflex; the original 50% frequency is cohort-dependent and untraced. Assess reflexes from appropriate angles and arrange a sufficiently complete examination, often under anaesthesia with peripheral fundus documentation. A new squint warrants prompt red-reflex/fundus assessment but is not itself proof of cancer. 1

02Separate laterality from heritable risk

Most tumours follow biallelic RB1 inactivation on chromosome 13q, consistent with Knudson’s two-hit model; rare RB1-wild-type MYCN-amplified unilateral tumours are an exception. A constitutional first hit can predispose to multifocal/bilateral disease, but unilateral disease can also be heritable. A13q deletion can involve dysmorphism; neither appearance nor a normal parent examination establishes genotype. 1 3

Original 40% heritable/60% nonheritable is an approximate epidemiologic distinction, not a way to classify one child by laterality. In a heterozygous germline RB1 carrier,50% variant transmission per pregnancy is not the same as certainty of an affected child. Heritable disease carries second-malignancy risk; confirmed nonheritable disease does not mean zero ordinary lifetime cancer risk. Retinoma/retinocytoma is a benign potential precursor, not necessarily spontaneous regression; offer genetic counselling, testing and family assessment. 1 3

Separate laterality from heritable risk
FeatureHeritable modelNonheritable model
GeneticsConstitutional RB1 first hitUsually tumour-local hits, rare exceptions
LateralityOften bilateral/multifocal, can be unilateralUsually unilateral/unifocal, not classification by appearance alone
Second tumoursIncreased risk, e.g. osteosarcomaNo zero-cancer-risk guarantee
Transmission50% variant risk per pregnancy in usual heterozygous modelNo germline transmission if genuinely absent; testing matters
ProportionOriginal approximately 40%, cohort contextOriginal approximately 60%, not a laterality test

03Use complementary imaging and avoid tumour dissemination

Ultrasound and MRI are complementary. Ultrasound can identify an intraocular mass and calcification; absence of calcium does not exclude diffuse infiltrating retinoblastoma. Specialist orbital/brain MRI assesses extent, optic nerve and intracranial disease, including trilateral disease (the original “trilinear” is incorrect). Surface coils do not make every MRI superior to ultrasound for every question. Avoid unnecessary ionising-radiation CT, especially with heritable predisposition; needed emergency imaging is an individual specialist decision. 1 2

Do not perform routine intraocular tumour biopsy or vitrectomy while retinoblastoma remains suspected. Diffuse disease can mimic uveitis/endophthalmitis with pseudohypopyon. Coats more often has unilateral telangiectasia and lipid/xanthocoria, but sex, yellow versus white reflex, angiographic leakage or absent obvious solid mass alone cannot safely exclude tumour. A painful blind eye of uncertain cause requires specialist oncologic planning, not automatic unplanned enucleation. 1

04Use stage to frame treatment choices

ICRB estimates intraocular eye-salvage prospects; AJCC staging addresses broader tumour extent and prognosis. Group A has small favourable tumours, B includes broader intraretinal disease than the original “all small” wording. C/D distinguish limited versus more extensive seeding/fluid patterns; E is very advanced intraocular disease with poor salvage potential. These letters do not replace full assessment or a life-preserving plan when extraocular/optic-nerve extension is suspected. 1 2

05Choose local and systemic treatment together

Systemic chemotherapy remains important alongside focal therapy, intra-arterial chemotherapy(IAC), intravitreal chemotherapy and selected enucleation. IAC via the ophthalmic artery requires expert endovascular care. The original forehead blush is not a required therapeutic endpoint; femoral occlusion, alopecia/arterial distribution and leg compartment syndrome are cited procedural concerns with untraced rates, not inevitable events. 1 2

The original IVC means intravitreal chemotherapy here, not intravenous systemic treatment. Melphalan and topotecan are both used for vitreous seeds in specialist protocols. Original universal topotecan superiority, RPE-toxicity advantage and mandatory standard are not established by the inspected evidence. Route, dose, burden and toxicity matter; do not substitute one drug automatically. 1 2

Radiation, including proton therapy, is selected with attention to growth and second-cancer risks, particularly in germline RB1 disease. The original Essen “ultima ratio” is an unverified local practice attribution, not proof of zero secondary tumours or a universal sequence. Enucleation may be the safest option in an unsalvageable advanced painful eye; suspected optic-nerve/extraocular invasion needs an oncologic strategy rather than a blanket routine-enucleation rule. 1 2

06Recognise a ciliary-body masquerade

Medulloepithelioma usually arises from nonpigmented ciliary epithelium and may present with lens subluxation, cataract, cysts or secondary glaucoma/buphthalmos. Although often unilateral, it can be associated with heritable DICER1 predisposition: the original “no heredity” is incorrect. The original 10-year-old boy with presumed disc oedema and diffuse vitreous infiltration remains an unverified teaching case, not a diagnostic pattern. 1 4

Plaque brachytherapy can preserve selected eyes, but is not automatically the preferred treatment for every medulloepithelioma. Tumour size, extent, visual potential and malignant features guide plaque, resection or enucleation in a specialist centre; DICER1 assessment may change family surveillance. 1 4

Recognise a ciliary-body masquerade
FeatureRetinoblastomaCoatsMedulloepithelioma
Age tendencyOriginal<3 years, older cases possibleOriginal 5–10 years/boys, not exclusiveOriginal 2–10 years, not exclusion boundaries
GeneticsRB1(13q), rare MYCN exceptionsNo single established germline causeDICER1 association
HeredityOriginal 40% approximate, testing requiredUsually sporadicCan be heritable, unlike original “none”
PresentationLeukocoria/strabismus or masqueradeLipid/xanthocoria/telangiectasiaSubluxation/glaucoma/cysts
ImagingExamination, ultrasound and orbital/brain MRIPeripheral exudation/leakage; exclude tumourCiliary-body mass; specialist imaging
TreatmentSelected systemic/local/IAC/intravitreal/enucleationSelected laser/cryo; original plaque not routine Coats therapySelected plaque/resection/enucleation, no universal preference

07Keep new diagnostics and global outcomes in context

Aqueous-humour cfDNA analysis can detect somatic copy-number alterations(SCNAs) and variants including MYCN-related changes in selected specialist pathways. It requires anterior-chamber sampling and therefore punctures the eye: the original “without puncture” is wrong. Selected clinical validation does not make every assay a routine universal replacement for examination, staging or germline testing. Avoid unspecialised ocular sampling when retinoblastoma is suspected. 1 6

Original mortality<1% in high-income versus>28% in low-income settings has an untraced denominator/time horizon. A prospective international cohort of 4064 children diagnosed in 2017 across 149 countries reported 3-year survival 99.5%(95% CI98.8–100) versus 57.3%(52.1–63), respectively. These survival estimates are not the same endpoint as crude recorded deaths. Income/access comparisons are observational and confounded, not proof of one causal intervention; earlier specialist access remains a practical care goal. 1 5

What are the main learning points?

Stage the tumour, assess germline risk and select treatment in a multidisciplinary centre; eye preservation cannot take priority over survival.

When is assessment urgent?

A white pupil, new squint, unexplained glaucoma or atypical hypopyon in a child needs prompt specialist assessment.

References and access status

  1. 1
    Immutable original export; paediatric intraocular tumour export

    Complete export read; original attribution and any recording/transcript or appointment not independently verified.

  2. 2
    NCI professional retinoblastoma PDQ: selected treatment passages

    Selected indexed/official passages read 17November 2026; no complete source, imaging or operative-protocol appraisal.

  3. 3
    GeneReviews: Retinoblastoma, selected indexed genetic summary

    Selected indexed/official passages read 17November 2026; no complete source, imaging or operative-protocol appraisal.

  4. 4
    GeneReviews: DICER1-related tumour predisposition, selected indexed CBME passages

    Selected indexed/official passages read 17November 2026; no complete source, imaging or operative-protocol appraisal.

  5. 5
    Global retinoblastoma outcome study; prospective international cohort; PMID35839812

    Selected indexed/official passages read 17November 2026; no complete source, imaging or operative-protocol appraisal.

  6. 6
    Prospective implementation of an aqueous humour liquid biopsy platform; selected indexed abstract

    Selected indexed/official passages read 17November 2026; no complete source, imaging or operative-protocol appraisal.

Original lecture link