Optical Coherence Tomography guided decisions in retinoblastoma

management

Sameh E. Soliman, MD,1,2 Cynthia VandenHoven,1 Leslie D. MacKeen,1 Elise Héon, MD,

FRCSC,1,3 Brenda L. Gallie, MD, FRCSC1,3-5

Authors affiliations

1Department of Ophthalmology and Vision Sciences, Hospital for Sick Children, Toronto,

Canada.

2Department of Ophthalmology, Faculty of Medicine, University of Alexandria, Alexandria,

Egypt.

3Department of Ophthalmology & Vision Sciences, Faculty of Medicine, University of Toronto,

Toronto, Ontario, Canada.

4Departments of Molecular Genetics and Medical Biophysics, Faculty of Medicine, University of

Toronto, Toronto, Ontario, Canada.

5Division of Visual Sciences, Toronto Western Research Institute, Toronto, Ontario, Canada.

Corresponding author:

Sameh E. Soliman, 555 University Avenue, room 7265, Toronto, ON, M5G 1X8.

samehelsayedsoliman@yahoo.com

Authors’ contributions

Concept and design: Soliman, VandenHoven, MacKeen, Héon, Gallie

Data collection: Soliman, VandenHoven, MacKeen.

Figure construction: Soliman, VandenHoven.

Analysis and interpretation: Soliman, VandenHoven, MacKeen, Héon, Gallie.

Critical review: Soliman, VandenHoven, MacKeen, Héon, Gallie

Overall responsibility: Soliman, VandenHoven, MacKeen, Héon, Gallie

Financial Support: None

Conflict of Interest: No financial conflicting relationship exists for any author.

Running head: OCT guided retinoblastoma management

Word count: 2170 / 3000 words

Numbers of figures and tables: 9 figures and 3 tables; 1 supplementary table

Key Words: retinoblastoma, Optical coherence Tomography, OCT, Cancer,

Meeting presentation: American Academy of Ophthalmology Annual Meeting presentation

(Chicago 2016, Monday 17th October 2016)

Abstract: (296/350 words)

Purpose: Assess the role of handheld Optical Coherence Tomography (OCT) in guiding

management decisions during diagnosis, treatment and follow-up of eyes affected by

retinoblastoma.

Design: Retrospective non-comparative single institution case series.

Participants: All children newly diagnosed with retinoblastoma from January 2011 to

December 2015 who had an OCT imaging session during their active treatment at The Hospital

for Sick Children (SickKids) in Toronto, Canada. OCT sessions for fellow eyes of unilateral

retinoblastoma without any suspicious lesion and those performed more than six months after

the last treatment were excluded.

Methods: Data collected included: age at presentation, sex, family history, RB1 mutation

status, 8th edition TNMH Cancer staging and International Intraocular Retinoblastoma

Classification (IIRC), and number of OCT sessions per eye. Details of each session were

scored for indication-related details (informative or not) and assessed for guidance (directive or

not), diagnosis (staging changed, new tumors found or excluded), treatment (modified, stopped

or modality shifted), or follow-up modified.

Main outcome measures: Frequency of OCT-guided management decisions, stratified by

indication and type of guidance (confirmatory versus influential).

Results: Sixty-three eyes of 44 children had 339 OCT sessions per eye (median 5, range 1-

15). Age at presentation and the presence of a heritable RB1 mutation significantly increased

the number of OCT sessions. Indications included evaluation of post-treatment scar (55%) or

fovea (16%), and posterior pole scanning for new tumors (11%). Of all sessions 92% (312/339)

were informative; 19/27 non-informative sessions had large, elevated lesions; of these, 14/19

were T2a or T2b (IIRC Group C or D) eyes. In 94% (293/312) of the informative sessions, OCT

directed treatment decisions (58%), diagnosis (16 %) and follow-up (26%). OCT influenced and

changed management from pre-OCT clinical plans in 15% of all OCT sessions.

Conclusions: OCT improves accuracy of clinical evaluation in retinoblastoma management.

Précis: (35/35 words; 226/460 characters)

We determined impact of handheld optical coherence tomography in retinoblastoma

management: 94% of 339 OCT sessions contributed indication-related details in 63 affected

eyes/ 44 patients; 86% significantly guided care; and 15% influenced change in management.

Optical Coherence Tomography (OCT) is well established to play an important role in ophthalmic patient

assessment, improving diagnostic accuracy and therapeutic decisions for a variety of ocular and retinal

conditions1-4 including ocular oncology.5,6 Handheld OCT performed while the supine child is under

anesthesia has deepened understanding of the features of retinoblastoma, the most common pediatric

ocular malignancy.7-10

OCT is shown valuable in retinoblastoma for detection of small invisible tumors,5,11-13 foveal

evaluation,14,15 localization and microstructure of tumor seeds,16 and detection of optic nerve

infiltration.10,17 It is documented to help in assessment of tumor anatomy, scar edges and simulating

conditions (e.g. retinoma or astrocytoma).5,18-20

However, handheld OCT is still not commonly used except in highly specialized ocular oncology

centers.7,21 The current Canadian Guidelines21 for retinoblastoma management define a center using

handheld OCT as a tertiary center.

In this study, we evaluate the influence of handheld OCT in guiding the management decisions in

children with retinoblastoma.

Methods

Study design

This study is a retrospective review of children with retinoblastoma who were managed in the Hospital

for Sick Children (SickKids), Toronto, Ontario, Canada from January 2011 to December 2015. Ethics

approval was obtained and the study follows the guidelines of the Declaration of Helsinki.

Eligibility

The records of all children with retinoblastoma examined with OCT imaging during management were

reviewed. Fellow eyes of unilateral retinoblastoma without any suspicious lesion studied at a single OCT

session at presentation were excluded. OCT sessions performed 6 months after the last treatment were

excluded.

Data collection

The data collected included age at presentation, sex, family history, laterality, International Intraocular

Retinoblastoma Classification (IIRC)22 at presentation, genetics results, indication for OCT, number of

OCT sessions per eye, and total active duration treatment (time from diagnosis until last treatment). The

extent cancer in each eye was retrospectively defined by the 2017 8th edition AJCC TNMH cancer

staging.23

OCT Session and Systems

We defined an OCT session as imaging of one eye for one or more indications, during an examination

under anesthesia. During the course of the study, two generations of handheld OCT systems were utilized:

Bioptigen® Envisu C2200 and Envisu C2300 (Bioptigen, Inc. Leica Microsystems, Morrisville, NC

USA). We did not compare the machines. We did not receive sponsorship or financial support to conduct

our research. At any point of time, one machine was available for both clinic and operating room. All

scans were captured by one of two highly skilled medical imaging specialists (authors CV and LM),

following a standardized methodology for good longitudinal reproducibility.

Technical considerations and indications

OCT was performed with operator at 12 o’clock position of the supine patient. The OCT scanner was

pivoted approximately 1 cm above the cornea, the optimal working distance, aiming the scanning beam

through the pupillary center.24 By manually holding the scanner, the operator was able to increase the

probe to eye working distance in real time while scanning over the apex of larger lesions. Image quality

and scan brightness was optimized by a combination of factors: manual adjustment of the OCT

spectrometer reference arm settings in accordance to the patient’s axial length; optimizing the focus for

the child’s refraction;24 and frequent application of 0.9% NaCl solution to prevent corneal dryness.

The handheld OCT produces a variety of scan configurations. For this study cohort, we routinely

obtained volumetric scans composed of non-averaged OCT scans (1000 A-scans x 100 B-scans per

volume). The accumulation of individual 100 B-scan produced the associated C-scan fundus image

otherwise called the Sum Voxel Projection (SVP). The OCT’s accompanying SVP image provided

critical information about the quality of the scan and so the OCT operator could respond in real-time with

positional adjustments to improve subsequent scans. To clarify pathology localization calipers were

placed on the OCT B-scan image to reveal the retinal position on the SVP image and measure tumor

height (Fig 1). Although algorithms might be applied to improve image quality via oversampling and

averaging of multiple scans,25 we routinely captured single line volume scans as they achieved rapid and

high quality images with ample clinical detail. To assess the posterior pole (Fig 2) for pre-clinical or

“invisible” tumor in infants less than 6 months of age, we used the widest volumetric scan settings

available. We performed 9 mm x 9 mm scans (Envisu C2200 system) and 12 mm x 12 mm scans (Envisu

C2300 system) of fovea, optic nerve, temporal, superior and inferior quadrants. If a tumor was identified,

the scan was repeated with the tumor centered within the OCT frame (Fig 2, 3).

For foveal or perifoveal tumors, the foveal center was located by a horizontal macular volumetric scan.

When needed, a vertically oriented scan was performed with the scanner is held the same physical

configuration while the SVP image was rotated 90 degrees indicating the scan direction change (Fig 4).

For parafoveal scans, the scanner was pointed towards the area of interest. Increased resolution for

small lesions was obtained by reducing the scan volume area to 8 x 8, or 6 x 6, maximizing the number of

A-scans/line. To assess the mid-periphery and beyond, a scleral depressor was used to rotate the eye,

while angling the scanner perpendicular to the retinal plane (Fig 5).

Assessment

An OCT session was assessed Informative if it provided sufficient data about the main indication;

Directive if the information obtained guided management decisions affecting diagnosis, treatment or

follow-up. Directive guidance that confirmed the pre-OCT clinical decision was considered

Confirmatory, and Influential if it changed a pre-OCT clinical decision. Every OCT session during the

active treatment phase of each child was assessed. Guidance was provided for diagnosis, treatment or

follow-up (Tables 2 and 3).

Diagnosis sessions were scored Confirmatory when OCT confirmed a clinically suspicious tumor

mass or clinical eye IIRC22 Group, including children less that 6 months of age known to carry an RB1

mutant allele; and Influential when OCT excluded tumor in clinically suspicious area(s), changed IIRC22

Group, or detected an invisible tumor during posterior pole screening.

Treatment sessions were scored Confirmatory when OCT confirmed a clinically suspicious new or

recurrent tumor or showed anatomic details (fovea, scarring, seeds, traction, etc.) supporting the planned

treatment; and Influential when OCT revealed an unsuspected recurrent tumor within a tumor scar or

showed anatomic details mandating changing the treatment modality or plan.

Follow-up sessions were considered Confirmatory when the OCT showed no change from the last

scan in absence of active treatment; and Influential when OCT showed anatomic details excluding

activity, leading to alteration in treatment plan.

Results:

Patient Demographics and numbers of OCTs

We reviewed 339 OCT sessions for 63 eyes of 44 children with retinoblastoma; 26 were male. Eight

children (10 eyes) were under active treatment; one child (one eye) was lost to follow up when they

moved outside Canada. The median number of OCT sessions per eye was 5 (range: 1-15 sessions),

significantly higher for familial (7) than non-familial (4) eyes (p=0.001, Mood’s Median test). Younger

children at presentation received significantly more OCT sessions (r=-0.26, p=0.04). The most common

indication for OCT was tumor scar evaluation (186/339, 55%), followed by foveal assessment and

posterior pole screening (16% and 11% respectively) (Table 2).

OCT Impact on Care

Informative versus Non-informative

OCT was Informative in 92% of sessions (312/339) (Table 2). Large or highly elevated lesions

rendered OCT technically challenging and Uninformative in 19/27 sessions (Table 3, Fig 1); 14/19 were

cT2a23 or cT2b23 (IIRC22 Group D or C) at presentation. In two eyes/children, OCT became Uninformative

after multiple previously Informative OCTs, due to progression of central tumor (one) and tractional

retinal detachment (one).

Directive versus Non-Directive OCT

OCT was Directive in 86% (293/339) of all OCT sessions and 94% (293/312) of Informative sessions

(Table 2), guiding treatment (168/312, 54%), diagnosis (46/312, 15%), or follow up (79/312, 25%).

Nineteen OCT sessions were Non-Directive, mainly because the OCT was not performed to assess a

clinical decision (17/19) or performed for academic interest (2/19) (Table 3).

Confirmatory versus Influential OCT

Of Directive OCT sessions, 243/293 (83%) were Confirmatory: for treatment 141 (58%), diagnosis 39

(16%) or follow-up 63 (26%) (Table 2). Of Directive OCT sessions, 50/293 (17%) were Influential: for

treatment 27/293 (11%), diagnosis 7/293 (3%) or follow-up 16/293 (7%) (Table 2). The most Influential

OCT sessions were for scar and foveal evaluation (Table 3, Figs 1 to 9). We have previously published

one Influential OCT which showed tumor over optic nerve head.17

OCT provided limited information in eyes with that were staged cT2 (TNMH 8th edition23) (IIRC22

Group C, D) or with large tumors, due to absorption of optical signal by dense lesions and lesion

elevation beyond the scan capacity.24 Eyes staged cT123 (IIRC22 Groups A and B) were easily scanned up

to the mid periphery26 (Fig 5). OCT assessed well the location of tumor with respect to retina: intra-

retinal, pre-retinal, vitreal or subretinal (Fig 6). This supported accurate TNMH23,27 or IIRC22 staging, for

example, suspected tumor separate from the primary tumor was shown by OCT to be subretinal tumor

extension, not an independent new tumor (Fig 6C). This influenced the diagnosis from multifocal tumor

to seeding of a unifocal tumor. The verification of tumor seeds by OCT16 also affected the choice of

treatment modality (i.e., intra-vitreal chemotherapy)28,29.

Discussion

OCT in retinal imaging has been shown effective to guide management (diagnostic and therapeutic)

decisions in multiple conditions, including macular holes,2 macular edema1 (diabetic and vascular) and

age related macular degeneration.3,4 Multiple reports have shown how useful OCT can be to differentiate

ocular tumors and simulating lesions.5,6,9-12,14-16,18-20,26 Currently, hand-held OCT is used by mainly in

tertiary level ocular oncology centers and learning institutes due to its relative high cost and limited

published benefits.21 The current study highlights an important role of OCT in guiding management of

retinoblastoma in 85% of sessions by confirming (83%) or even changing (17%) pre-OCT clinical

decisions.

We show that OCT improved clinical diagnostic accuracy and clinical staging (Fig 6) for

retinoblastoma. In familial retinoblastoma, detection of tiny and sometimes invisible tumors5,11 by OCT

(Fig 2-3) facilitated tumor control by only focal therapy, achieving minimal retinal damage and better

final visual outcomes.27 In unilateral retinoblastoma, OCT helped differentiate suspicious lesions from

retinoblastoma (Fig 7) in the normal eye. Previously, this depended on clinical opinion or B-scan

ultrasonography, which does not show the inner architecture of retina and lesion. Without in-vivo

evidence of the nature of these suspicious lesions, such lesions may have been treated with focal therapy,

potentially falsely labeling the child as bilateral, heritable retinoblastoma, imposing multiple unnecessary

examinations under anesthesia and life-long surveillance for second cancers.21,30

OCT evaluated well important anatomic landmarks such as the fovea and the optic nerve disc, which

affected our treatment and follow up choices. Foveal pit detection (Fig 4) provided information about

anticipated visual potential with perifoveal tumors.14 Foveal localization respective to the tumor affected

choice of treatment modality (chemotherapy versus primary focal therapy), which laser to use (532 nm

versus 810 nm laser) and technique (ie, sequential targeted laser therapy from the tumor side opposite the

fovea, Fig 8). An intact fovea during or after treatment suggested benefit of early amblyopia therapy.31,32

In peripapillary tumors,10,17,33 OCT appearance may raise suspicion of optic nerve invasion but sometimes

failed to distinguish tumor from papilledema. OCT improved clinical judgment during tumor scar

evaluation, and distinguished gliosis and scar from tumor recurrence (Fig 9), especially useful with white

choroidal scars where visualization of recurrence is challenging,33 thereby altering choice of treatment

modality.

The current study is limited by being a single center, retrospective study, with absence of correlation

to a quantifiable outcome. It was not practical to correlate OCT sessions with outcomes such as eye

salvage, vision salvage, life salvage, which are affected by many other factors (tumor location, number

and type, stage at presentation, complications of treatments, treatment duration, etc.). The presence of a

single OCT machine limited the number of sessions in some eyes due to unavailability related to

maintenance or concomitant use by other surgeons. Training and academic interest may have increased

the number of OCT sessions performed for some eyes, and we took this into account in scoring the impact

of the OCT session.

Cost is an important consideration during technology assessment. We did not collect this data during

our study. We hypothesize that OCT imaging results in decreased the treatment costs in multiple

situations, such as earlier detection of tumors in familial cases reducing the need for systemic therapies.

In unilateral cases with suspicious lesion in the fellow eye, OCT reduced the number of required

examinations under anesthetic for follow-up. OCT detected earlier scar recurrences treated with focal

rather than costly systemic therapies. A cost-effectiveness study is suggested.

In conclusion, multiple studies have reported OCT signs of retinoblastoma at presentation. To our

knowledge, this is the first study to evaluate the impact of OCT on guiding management decisions of

active retinoblastoma. Hand-held OCT is recommended in the investigative armamentarium of any

tertiary ocular oncology center to provide precision of retinoblastoma management.

Acknowledgement

There are no conflicts of interests or disclosures. BLG is the unpaid medical director of Impact Genetics.

References

1. Panozzo G, Parolini B, Gusson E, et al. Diabetic macular edema: an OCT-based classification. Semin Ophthalmol. 2004;19(1-2):13-20.

2. Rubowitz A. Classification of macular holes. Ophthalmology. 2007;114(10):1956-1957; author reply 1957.

3. Quellec G, Lee K, Dolejsi M, Garvin MK, Abramoff MD, Sonka M. Three-dimensional analysis of retinal layer texture: identification of fluid-filled regions in SD-OCT of the macula. IEEE Trans Med Imaging. 2010;29(6):1321-1330.

4. Introini U, Casalino G, Querques G, Gimeno AT, Scotti F, Bandello F. Spectral-domain OCT in anti-VEGF treatment of myopic choroidal neovascularization. Eye (Lond). 2012;26(7):976-982.

5. 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(1):59-65.

6. Medina CA, Plesec T, Singh AD. Optical coherence tomography imaging of ocular and periocular tumours. Br J Ophthalmol. 2014;98 Suppl 2:ii40-46.

7. Gallie BL, Soliman S. Retinoblastoma. In: Lambert B, Lyons C, eds. Taylor and Hoyt's Paediatric Ophthalmology and Strabismus. Vol 5th Edition. Oxford, OX5 1GB, United Kingdom: Elsevier, Ltd.; In Press.

8. Dimaras H, Corson TW, Cobrinik D, et al. Retinoblastoma. Nature Reviews Disease Primers. 2015:15021.

9. Lee H, Proudlock FA, Gottlob I. Pediatric Optical Coherence Tomography in Clinical Practice-Recent Progress. Invest Ophthalmol Vis Sci. 2016;57(9):OCT69-79.

10. Mallipatna A, Vinekar A, Jayadev C, et al. The use of handheld spectral domain optical coherence tomography in pediatric ophthalmology practice: Our experience of 975 infants and children. Indian Journal Of Ophthalmology. 2015;63(7):586-593.

11. Berry JL, Cobrinik D, Kim JW. Detection and Intraretinal Localization of an 'Invisible' Retinoblastoma Using Optical Coherence Tomography. Ocul Oncol Pathol. 2016;2(3):148-152.

12. Saktanasate J, Vongkulsiri S, Khoo CT. Invisible Retinoblastoma. JAMA ophthalmology. 2015;133(7):e151123.

13. Bremner R. Retinoblastoma, an inside job. Cell. 2009;137(6):992-994.14. Samara WA, Pointdujour-Lim R, Say EA, Shields CL. Foveal microanatomy

documented by SD-OCT following treatment of advanced retinoblastoma. J AAPOS. 2015;19(4):368-372.

15. Hasanreisoglu M, Dolz-Marco R, Ferenczy SR, Shields JA, Shields CL. Spectral Domain Optical Coherence Tomography Reveals Hidden Fovea Beneath Extensive Vitreous Seeding From Retinoblastoma. Retina. 2015;35(7):1486-1487.

16. Munier FL. Classification and management of seeds in retinoblastoma. Ellsworth Lecture Ghent August 24th 2013. Ophthalmic Genet. 2014;35(4):193-207.

17. Yousef YA, Shroff M, Halliday W, Gallie BL, Heon E. Detection of optic nerve disease in retinoblastoma by use of spectral domain optical coherence tomography. J AAPOS. 2012;16(5):481-483.

18. Shields CL, Manalac J, Das C, Saktanasate J, Shields JA. Review of spectral domain-enhanced depth imaging optical coherence tomography of tumors of the retina and retinal pigment epithelium in children and adults. Indian Journal Of Ophthalmology. 2015;63(2):128-132.

19. Pierro L, De Francesco S, Hadjistilianou D, et al. Spectral-domain optical coherence tomography appearance of a posterior pole retinoma. J Pediatr Ophthalmol Strabismus. 2014;51(5):320.

20. Malhotra PP, Bhushan B, Mitra A, Sen A. Spectral-domain optical coherence tomography and fundus autofluorescence features in a case of typical retinocytoma. Eur J Ophthalmol. 2015;25(6):e123-126.

21. Canadian Retinoblastoma S. National Retinoblastoma Strategy Canadian Guidelines for Care: Strategie therapeutique du retinoblastome guide clinique canadien. Can J Ophthalmol. 2009;44 Suppl 2:S1-88.

22. Murphree AL. Intraocular retinoblastoma: the case for a new group classification. Ophthalmology clinics of North America. 2005;18:41-53.

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

24. Maldonado RS, Izatt JA, Sarin N, et al. Optimizing hand-held spectral domain optical coherence tomography imaging for neonates, infants, and children. Invest Ophthalmol Vis Sci. 2010;51(5):2678-2685.

25. Scott AW, Farsiu S, Enyedi LB, Wallace DK, Toth CA. Imaging the infant retina with a hand-held spectral-domain optical coherence tomography device. Am J Ophthalmol. 2009;147(2):364-373 e362.

26. Choudhry N, Golding J, Manry MW, Rao RC. Ultra-Widefield Steering-Based Spectral-Domain Optical Coherence Tomography Imaging of the Retinal Periphery. Ophthalmology. 2016;123(6):1368-1374.

27. Soliman SE, Dimaras H, Khetan V, et al. Prenatal versus Postnatal Screening for Familial Retinoblastoma. Ophthalmology. 2016;123(12):2610-2617.

28. 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(8):1078-1083.

29. Munier FL, Soliman S, Moulin AP, Gaillard MC, Balmer A, Beck-Popovic M. Profiling safety of intravitreal injections for retinoblastoma using an anti-reflux procedure and sterilisation of the needle track. Br J Ophthalmol. 2012;96(8):1084-1087.

30. Villani A, Shore A, Wasserman JD, et al. Biochemical and imaging surveillance in germline TP53 mutation carriers with Li-Fraumeni syndrome: 11 year follow-up of a prospective observational study. The lancet oncology. 2016.

31. Watts P, Westal C, Colpa L, et al. Visual results in children treated for macular retinoblastoma. Eye. 2002;16(1):75-80.

32. Lengyel D, Klainguti G, Mojon DS. [Does amblyopia therapy make sense in eyes with severe organic defects?]. Klinische Monatsblatter fur Augenheilkunde. 2004;221(5):386-389.

33. Astudillo PP, Chan HS, Heon E, Gallie BL. Late-diagnosis retinoblastoma with germline mosaicism in an 8-year-old. J AAPOS. 2014;18(5):500-502.

Figure Legends

Figure 1. Central tumors. (A) A perifoveal tumor mass (cT1b23, IIRC22 Group B) was isodense within the

retinal layers; the exact foveal location was evident (yellow *); maximal tumor height of 0.75 mm

(Informative, Directive and Influential in guiding laser treatment) was over-estimated on B-scan

ultrasonography. (B) A peripapillary tumor (cT1b23, IIRC22 Group B) not involving the fovea measured

1.36 mm in height on B-scan ultrasonography; OCT provided no additional data (Non-informative). (C) A

juxtafoveal tumor (cT1b23, IIRC22 Group B) measured 1.65 mm in height on B-scan ultrasonography;

OCT showed intact overlying retinal layers and minimal surrounding subretinal fluid (Informative,

Directive and Confirmatory for diagnosis). (D) OCT on a large central tumor (cT1b23, IIRC22 Group B)

measuring 3.08 mm in height on B-scan ultrasonography was Confirmatory; OCT was Non-informative

regarding both tumor internal architecture and overlying retinal layers. In (B-D) tumors, calipers could

not be accurately utilized to measure tumor thickness, as the outer tumor boundary was ill defined.

Figure 2. OCT screening of posterior quadrants (superior, temporal, inferior, and nasal). (A, B) An

invisible lesion was found (white *) in the inferior quadrant scan; (C) reimaging centralized on the

suspicious area (green 12mm x 12mm box) showed an isodense small tumor within the inner nuclear

layer (Informative, Influential for diagnosis and treatment).

Figure 3. First diagnosis of small tumors. (A-D) After detection on posterior pole screening, small intra-

retinal elevated isodense round tumors centralized on the inner nuclear layer (cT1a23, IIRC22 Group A)

were confirmed (Informative, Influential for diagnosis and treatment).

Figure 4. Perifoveal tumors. The exact location of the foveal center (yellow *) was located in horizontal

(green line) and vertical (dotted green line) scans with the foveal pit at the intersection. The foveal center

was (A) on top of tumor, (B) partially involved or (C) adjacent to tumor (Informative, Influential for

diagnosis and treatment).

Figure 5: Pre-equatorial lesions. The eyes were deviated in the required direction with complimentary

tilting of the OCT scanner; peripheral indentation with scleral depressor was helpful. (A) OCT of a

peripheral nasal elevated isodense lesion. (B) OCT to evaluate a tumor tag (yellow *) vs vitreous seed

revealed an unsuspected nearby edge recurrence (arrow) (Informative, Directive, Influential for diagnosis

and treatment); (C) two months after both active tumors were treated, clinical exam and OCT showed that

the tumor tag (white *) extending into vitreous had increased in size, while the edge recurrence (arrow)

was a flat scar (Informative, Directive, Confirmatory); further laser and cryotherapy ablated the tumor

tag.)

Figure 6: Suspected tumor seeds. (A) Multiple white small masses in the macular area of an eye

harboring a large nasal tumor were shown by OCT to be preretinal vitreous seeds (Informative/ Directive/

Influential for diagnosis and treatment). (B) Multiple yellowish spots in an eye with treated

retinoblastoma were shown on OCT to be calcified with shadowing (arrows); an isodense inner nuclear

layer lesion (white *) was considered an active new tumor, thereby treated with laser (Informative/

Directive/ Influential for diagnosis and treatment). (C) A white lesion (arrow) inferior to large central

tumor with shallow retinal detachment in unilateral retinoblastoma was considered likely to be a separate

primary tumor, so the eye was staged cT2a23 (IIRC22 Group C); OCT showed this to be subretinal seeding

within shallow retinal detachment, changing initial staging to cT2b23 (IIRC22 Group D) changing

treatment (Informative/ Directive/ Influential for diagnosis and treatment).

Figure 8. Sequential targeted Laser therapy (STLT) in juxtafoveal retinoblastoma. The child presented

with a cT2b23 (IIRC22 Group D) eye with two large tumors; the central tumor was juxtafoveal; (A) after

six cycles of systemic chemotherapy, the fovea was visible on OCT; STLT was initiated using 532 nm

laser starting from the edge farthest from the fovea sequentially moving inwards (direction of the arrows)

avoiding the tumor nearest to the fovea; (B) appearance 6 months after starting STLT; (C) fovea was

further away from the tumor edge 12 months after starting STLT; (D) 18 months after starting STLT OCT

showed a flattened lesion with preserved fovea; 18 months after last treatment the tumor remained the

same (Informative/ Directive/ Confirmatory (Influential) for diagnosis, treatment, follow-up). Fovea

marked by yellow *.

Figure 9. Evaluation of tumor scars. (A) OCT of a clinically suspected recurrence in scar (white *)

showed an isodense elevation of indicating active tumor; the adjacent scar showed an unsuspected similar

edge recurrence; both were treated with laser. (B) OCT detected tumor activity (arrow) hidden within

areas of calcification. (C) OCT of two clinically suspicious white areas showed that the upper white area

(white *) was scar (gliosis) and the lower area (white *) was a tumor. (Informative/ Directive/ Influential

(Confirmatory) for diagnosis and follow-up).

Table legends

Table 1. Clinical, Genetic and Tumor characteristics.

Table 2: Stratification of different OCT assessments in diagnosis, treatment and follow up with indication for OCT imaging.

Table 3. Causes of different OCT assessment layers.