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

1

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

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 227495 / 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)

2

25

26

27

28

29

30

31

32

33

34

Abstract: (2963120/350 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 over the course of

clinical management (number of OCTs per eye median 5, range 1-15). per eye (median

5, range 1-15). Age at presentation and the presence of a heritable RB1 mutation

significantly correlated with 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-

3

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

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 and 17%

of directive sessions.

Conclusions: OCT improves accuracy of clinical evaluation in retinoblastoma

management.

4

59

60

61

62

63

64

65

66

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.

5

67

68

69

70

71

72

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 6

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

91

92

93

94

95

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 stage 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:

7

96

97

98

99

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

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

8

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

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.

9

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

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).

10

165

166

167

168

169

170

171

172

173

174

175

176

177

178

179

180

181

182

183

184

185

186

187

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 most often used by mainly in tertiary level ocular oncology centers and learning

institutes due to its relative high cost, limited feasibility with high case volumes and limited

11

188

189

190

191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

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 believeobserved that OCT improved clinical diagnostic accuracy and clinical staging

(Fig 6) for retinoblastoma when we used OCT. . In familial retinoblastoma, detection of tiny

small 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 unnecessarily 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

12

212

213

214

215

216

217

218

219

220

221

222

223

224

225

226

227

228

229

230

231

232

233

234

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

treatment modality.

OCT of recurrent tumor (Fig.9) shows a solid homogenous, elevated, rounded or ovoid hyper

reflectivity within a calcific regression. (Fig.9). It might be difficult in certain tumors with mixed

or fish-flesh regression to differentiate active from inactive areas except by stability over serial

OCT sessions. In our series, we opted to be more cautious and label all suspicious areas as active

as they can be easily treated with laser while small rather than wait for 4-5 weeks for follow up

and face a larger recurrence that might require a higher treatment burden.

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. A prospective

study would verify these results, if we could ensure the presence of an OCT machine in each

EUA, which would be costly and extend the time of EUA. 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.

13

235

236

237

238

239

240

241

242

243

244

245

246

247

248

249

250

251

252

253

254

255

256

257

258

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.

14

259

260

261

262

263

264

265

266

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.

15

267

268

269

270

271

272

273

274

275

276

277

278

279

280

281

282

283

284

285

286

287

288

289

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.

16

290

291

292

293

294

295

296

297

298

299

300

301

302

303

304

305

306

307

308

309

310

311

312

313

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.

17

314

315

316

317

318

319

320

321

322

323

324

325

326

327

328

329

330

331

332

333

334

335

336

337

338

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.

18

339

340

341

342

343

344

345

346

347

348

349

350

351

352

353

354

355

356

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.

19

357

358

359

360

361

362

363

364

365

366

367

368

369

370

371

372

373

374

375

376

377

378

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 20

379

380

381

382

383

384

385

386

387

388

389

390

391

392

393

394

395

396

397

398

399

400

401

402

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).

21

403

404

405

406

407

408

409

410

411

412

413

414

415