Methods · Web viewWhole-body MRI (WBMRI), lumbar puncture (LP) and bilateral bone marrow aspirate...

Clinical predictors at diagnosis of low risk histopathology in unilateral

advanced retinoblastoma

Stephanie N. Kletke, MD1, Zhao Xun Feng, BSc2, Lili-Naz Hazrati, MD, PhD, FRCPC3, Brenda L. Gallie,

MD, FRCSC1,2,4, Sameh E. Soliman, MD2,5*

Authors’ Affiliations

1 Department of Ophthalmology and Vision Sciences, University of Toronto, Toronto, Canada;

2 Department of Ophthalmology and Vision Sciences, The Hospital for Sick Children, Toronto, Canada;

3 Department of Pediatric Laboratory Medicine, The Hospital for Sick Children, Toronto, Canada;

4 Departments of Molecular Genetics and Medical Biophysics, University of Toronto, Toronto, Canada;

5 Department of Ophthalmology, Faculty of Medicine, Alexandria University, Alexandria, Egypt.

*Corresponding Author: Sameh E. Soliman, 555 University Avenue, Room 7265, Toronto, Canada,

M5G 1X8. [email protected]

Financial Support: None

Conflicts of interest: None

Running Head: Low-risk Histopathology in Unilateral cT2b Retinoblastoma

Number of Figures and Tables: 2 figures, 2 tables and 2 supplementary tables.

Keywords: unilateral retinoblastoma; Group D; histopathology; cancer; primary enucleation; vitreous

seeds.

1

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

Soliman et al. Low-risk Histopathology in Unilateral cT2b Retinoblastoma

Abstract (348/350)

Objective: Attempted eye salvage for unilateral (cT2b/Group D) retinoblastoma may increase risk of

tumor spread compared to primary enucleation. Identification of clinical features predictive of low

histopathologic risk would support safe trial salvage.

Design: Retrospective, non-comparative single-institutional observational case series

Participants: Children with unilateral cT2b /Group D retinoblastoma (January 2008-February 2018)

who were managed with primary enucleation at the Hospital for Sick Children (SickKids), Toronto,

Canada.

Methods: Data included clinical features (intraocular pressure (IOP), optic nerve obscuration, macular

involvement, tumor seeding and serous retinal detachment (RD) >1 quadrant), timing to enucleation,

histopathological findings, genetic testing and follow-up (metastasis and death).

Main outcome measures: Primary outcome was low-risk (LR) (pT1/pT2) versus high-risk (HR)

(pT3/pT4) (8th Edition American Joint Committee on Cancer) histopathology. Clinicopathologic

correlations were evaluated. Secondary outcome measures were positive predictive values

(probability that certain clinical features would predict LLR histopathology) and, negative

predictive values (probability that absence of these clinical features would predict HLR

histopathology), specificity and sensitivity values.

Results: Thirty-eight eyes were eligible with mean age of 21 months (range: 2-48) at diagnosis. All had

vitreous seeding and normal IOP. The median diagnosis to enucleation interval was 4 days (range, 0–14)

without parental refusals. Histopathology diagnosed 4/38–10.5% HR and 34/38–89.5% LR eyes. HR eyes

demonstrated massive choroidal invasion (4/38–10.5%), or trans-scleral, extraocular and post-laminar

optic nerve invasion (1/38–2.6%). Clinical findings included macular involvement (31/38–82%),

complete optic nerve obscuration (27/38–74%), and RD (28/38–74%). The proportion of eyes with HR

2

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

Zhao Xun Feng, 03/22/19,

The confidence interval for specificity is too big to provide any useful information. It is from 8% to 100%


histopathology was 13% (4/31; 95%CI 1% to 25%) with macular involvement, 15% (4/27; 95%CI 1%

to 28%) with complete optic nerve obscuration, and 14% (4/28; 95%CI 1% to 27%) with RD. The

predictability of LR histopathology was 100% (7/7; 95%CI 47% to 100%) with macular sparing

(7/7), 100% (10/10; 95%CI 63% to 100%) with optic nerve visibility (10/10) and/or 100% (10/10;

95%CI 63% to 100%) with <1 quadrant of RD (10/10) with 100% specificity. One child lacking all

three clinical LR predictive features had HR histopathology (pT3a) developed metastases and died; other

children are alive and well (mean follow-up 65 months).

Conclusion: Presence of macular sparing, optic-nerve visibility and/or <1 quadrant of RD predicted

are highly predictive of LR histopathology at primary enucleation, suggesting that any of these features

predict safe eye salvage. No clinical sign predicted HR histopathology.

3

42

43

44

45

46

47

48

49

50

51

Zhao Xun Feng, 03/22/19,

The abstract exceeded the word limit, perhaps we can remove this sentence since it doesn’t relate to our primary and secondary objectives?


Unilateral retinoblastoma eyes staged cT2 (8th Edition American Joint Committee on Cancer (AJCC)

TNMH, tumor, node, metastasis and heritable trait)1 and Group D (International Intraocular

Retinoblastoma Classification, IIRC)2 poses a management challenge. Attempted eye salvage using

primary intra-arterial (IAC)3-7 or systemic chemotherapy8 (both with focal consolidation) is commonly

suggested. However, primary enucleation is an effective option to minimize risk of extraocular extension

and metastasis. The Canadian National Retinoblastoma Strategy Guidelines for Care published in 2009

recommend enucleation of affected unilateral Group D eyes.8

Recently, multiple treatment modalities are suggested to improve success of eye salvage, including

intravitreal chemotherapy (IVitC),9-11 IAC,3-7 periocular chemotherapy,12 and tumor endoresection via pars

plana vitrectomy (PPV).13 The primary concern is whether attempted eye salvage increases the risk of

extraocular tumor dissemination. Our aim was to identify clinical features of primarily enucleated

unilateral cT2 (Group D) eyes that at diagnosis predict low histopathologic risk, in order to guide “safe”

trial eye salvage.

Methods

Study Design

A retrospective, non-comparative, single institutional observational study was conducted in

accordance with the guidelines of the Declaration of Helsinki. Institutional Research Ethics Board

approval was obtained.

Eligibility

Children diagnosed with unilateral Group D retinoblastoma managed with primary enucleation of the

affected eye at the Hospital for Sick Children (SickKids), Toronto, Canada between January 2008

(following submission and implementation of the Canadian guidelines8) through February 2018 were

4

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

67

68

69

70

71

72

73


evaluated. Exclusion criteria were unilateral retinoblastoma of any other clinical stage, bilaterally affected

children, and cT2 eyes that were secondarily enucleated following trial salvage.

Data Collection

Clinical and Radiological Features

Medical records, including fundus photographs from examinations under anesthesia (EUA), were

reviewed for age at diagnosis and enucleation, laterality, clinical features at presentation (intraocular

pressure (IOP), tumor seeding, optic nerve obscuration (defined as inability to completely see the borders

of the optic nerve head, and subcategorized as “partial” if any portion of the nerve was seen or

“complete” if there was total obscuration), macular involvement and serous retinal detachment (RD),

parental consent to the proposed treatment, eye staging by IIRC, molecular genetic analysis, follow-up

duration, adjuvant treatments received, metastasis and death. Eyes were retrospectively staged by the 8th

Ed. AJCC TNMH.1 Baseline magnetic resonance imaging (MRI) or computed tomography (CT) of the

brain and orbit were reviewed.

Histopathologic Features

Histopathology reports and representative slides were reviewed for all children. Presence of choroidal

invasion was documented as “none”, “focal” (<3 mm) or “massive” (>3 mm in maximum diameter),

based on consensus definitions from the International Retinoblastoma Staging Working Group.14 Invasion

under the retinal pigment epithelium (RPE) but not through Bruch’s membrane was identified. Optic

nerve invasion was categorized as “none”, “prelaminar”, “post-laminar but not to the optic nerve

resection margin” and “tumor at the transected end”.14 Scleral invasion, anterior segment involvement and

extraocular disease were identified. Enucleated eyes were retrospectively staged by the 8th Ed. AJCC

pTNM1 (Table 1).

Outcome Measures

5

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

91

92

93

94

95

96


The primary outcome was the presence of low-risk (LR) histopathology (pT1/pT2) or high-risk (HR)

histopathology (pT3/pT4).1 High-risk histopathologic features included massive choroidal invasion, post-

laminar invasion of the optic nerve head, scleral invasion and extraocular extension.

Clinicopathologic correlation was evaluated. Positive predictive value was the probability of a clinical

feature to predict LR histopathology. Negative predictive value was the probability that absence of this

clinical feature to predict absence of LR (i.e. HR) Histopathology. Sensitivity and specificity for each

clinical sign to predict LR histopathology were calculated.

Statistical Analysis

Results were summarized using frequency/percentage for categorical variables and mean, median,

standard deviation and range for continuous variables. Groups were compared using Fisher’s exact test

for categorical variables and Student’s t-test for continuous variables. All P-values reported were two-

sided and significance was judged at the 5% level. All analyses were performed using SPSS Version 25

(IBM Corp).

Results

Demographic and Clinical Features

Thirty-eight (Supplementary Table 1) primarily enucleated Group D eyes of 38 children (presenting

age mean 21 months, range 2–48) with unilateral retinoblastoma were included (63% right, 37% left). All

eyes were staged cT2b (8th Edition AJCC).1 There was only one unilateral Group D eye during the study

period not offered enucleation at presentation due to initial staging as Group C. She was offered trial

salvage instead and is excluded from this study. The proper staging was ascertained by the OCT finding

of shallow subclinical detachment that involved the whole inferior retina upstaging it to Group D.15

At presentation, all eyes had normal IOP. Vitreous seeding was present in all eyes. Tumor involved the

macula in 31/38–82%. Children with macular involvement tended to be younger at diagnosis than

6

97

98

99

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119


children with macular sparing (mean 20 vs 28 months, respectively) (p=0.09). The optic nerve was

obscured in 28/38–74% (1/28 partial, 27/28 complete obscuration) and RD (>1 quadrant) was present in

28/38–74%. Retinal detachment impaired accurate assessment of subretinal seeding in some eyes.

Concomitant presence of obscured optic nerve and macular involvement was found in 26/38–68%,

obscured optic nerve and RD in 25/38–66%, macular involvement and RD in 28/38–74% and all three

clinical findings in 25/38–66%. Conversely, concomitant finding of visible optic nerve, macular sparing

and <1 quadrant of RD was present in 5/38–13%. The presence of macular involvement, optic nerve

obscuration and RD were positively correlated [macula and optic nerve (p=0.01), macula and RD (p<

0.001), optic nerve and RD (p=0.002)]. Four of 28 eyes with optic nerve obscuration demonstrated

possible optic nerve enhancement on baseline imaging of the brain/orbit. There was no radiological

evidence of extraocular or intracranial involvement in any child.

The median interval from diagnosis to enucleation was 4 days (range, 0–14). Primary enucleation was

performed during the staging EUA for all except one child, for whom enucleation was delayed due to low

partial thromboplastin time. All parents consented to enucleation as the primary treatment.

Histopathologic Features

Choroidal involvement included “none” (26/38–68.4%), “focal” (8/38–21.1%), and “massive” (4/38–

10.5%). Six eyes (15.8%) demonstrated tumour under the RPE without invasion of Bruch’s membrane.

Optic nerve involvement included “none” (10/38–26.3%), “prelaminar invasion” (27/38–71.1%), and

“post-laminar invasion but not to the optic nerve resection margin” (1/38–2.6%). There were no cases of

tumor involvement to the resected margin. One eye (2.6%) demonstrated histopathologic evidence of

anterior segment involvement (pT2b). One eye (2.6%) had trans-scleral and extraocular extension (pT4).

One eye (2.6%) had concomitant massive choroidal invasion, post-laminar optic nerve invasion and

extraocular extension.

High-Risk Pathology Eyes (Figure 1)

7

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143


Histopathology review identified 4/38–10.5% HR eyes and 34/38–89.5% LR eyes. HR eyes

demonstrated massive choroidal invasion (4/38–10.5%), and trans-scleral, extraocular and post-laminar

optic nerve invasion (1/38–2.6%). Children with HR versus LR eyes had a higher median age at diagnosis

(31 versus 20 months, p=0.035, Mann-Whitney test). Children <24 months tended to be LR (p=0.04,

Fisher Exact test). The predictability (positive predictive value) of a child <24 months to be LR was

100% (95%CI 82% to 100%), while children ≥24 months to be HR was 22% (95%CI 3% to 41%)

(Table 2). Presenting signs included leukocoria (3/4–75%) and strabismus (1/4–25%). Baseline MRI

brain and orbits showed no evidence of optic nerve, extraocular or intracranial involvement in children

with HR eyes. There was no evidence of metastases at presentation.

Clinicopathologic Correlation

Optic nerve obscuration was not significantly associated with post-laminar optic nerve invasion in this

cohort (p=1.000). Macular involvement was not significantly associated with massive choroidal invasion

(p=0.557) or scleral invasion (p=1.000). Serous RD was not significantly associated with massive

choroidal invasion (p=0.287) or scleral invasion (p=1.000, Supplementary Table 2). None of the eyes

showing enhanced optic nerve on MRI at presentation had post-laminar nerve invasion (p=1.000). All of

the 13 eyes with at least one of one of the following three signs: of macular sparing, optic nerve

visibility or <1 quadrant of RD showed LR histopathology. Twenty-one eyes (84%, 21/25) lacked all 3

signs and showed LR histopathology (p=0.28) (Figure 2). The predictability (positive predictive value)

that an eye had LR histopathology was 100% (95%CI 47% to 100%) with macular sparing, 100%

(95%CI 63% to 100%) with optic nerve visibility and, 100% (95%CI 63% to 100%) with <1 quadrant

of RD. and Eyes withany combination of two or three of these clinical findings all was also 100%

predictive ofhad LR histopathology with 100% specificity (i.e. truly have LR histopathology) (Table 2).

The negative predictive value that an eye lack LR histopathology (i.e. HR histopathology) was 13%

(95%CI 1% to 25%) with macular involvement, 14% with any degree of optic nerve obscuration

(95%CI 1% to 27%), 15% (95%CI 1% to 28%) with complete optic nerve obscuration and 14%

8

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168


(95%CI 1% to 27%) with RD. The predictability of HR histopathology with concomitant findings of

obscured optic nerve and macular involvement was 15% (95%CI 2% to 29%), obscured optic nerve

and RD was 16% (95%CI 2% to 30%), macular involvement and RD was 14% (95%CI 1% to 27%),

and all three clinical findings was 16% (95%CI 2% to 30%). The sensitivity (i.e. rule out LR

histopathology) was low (table 2).

Molecular Analysis

Molecular genetic testing was performed on tumor samples from 37 enucleated eyes while one eye

(HX) was untested. The two tumor RB1 pathogenic variants were identified in 34/37 eyes. Blood tests for

these pathogenic variants showed that the patients were H116 (7/38–18%) or H0* (27/38–71%); 3 patients

were HX (heritability cannot be verified). The H1 children all had known RB1 pathogenic variants

associated with unilateral disease: mosaicism (3/7), low penetrance RB1 pathogenic variant (3/7) and 13q

deletion syndrome (1/7). The children that showed high-risk pathology were H1 (1 pT4 eye, Figure 1A),

H0*16 (pT3a, Figure 1B) and HX (2 pT3a eyes, one died, Figure 1C, D).

Follow-up, Metastasis and Death

At mean follow-up of 65 months, one child (2.6%) lacking all three clinical LR predictive features

showed HR histopathology (pT3a, massive choroidal invasion) and developed metastases and died. Bony

metastases were found 1 year following retinoblastoma diagnosis.17 The child received six cycles of

systemic chemotherapy, autologous bone marrow transplant and focal radiation. While ocular pathology

was initially interpreted as LR, internal retrospective review identified a previously not seen area of

massive choroidal invasion. Metastatic surveillance remained negative until 1 year later, when

intracranial dural metastases were identified on MRI. The child died 18 months after metastasis diagnosis.

The other children in this cohort are alive and well. No child was lost to follow-up.

9

169

170

171

172

173

174

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190


Discussion

The International Intraocular Retinoblastoma Classification (IIRC) (2005) staged eyes clinically

Group A (very low risk) through E (very high risk) to predict outcomes following systemic

chemoreduction and focal therapy.2 The 8th Edition TNMH was based on evidence from an international

survey,1 now recommended for retinoblastoma staging. Advanced intraocular disease includes Group E

eyes [with phthisis bulbi (cT3a), anterior segment tumor invasion (cT3b), rubeosis irides with neovascular

glaucoma (cT3c), hyphema and/or massive vitreous hemorrhage (cT3d) and aseptic orbital cellulitis

(cT3e)], and Group D eyes [with significant RD (cT2a) and/or seeding (vitreous and/or subretinal, cT2b)].

High-risk histopathologic features of the enucleated eye predictive of increased metastatic risk are

pT3/pT4, which include massive choroidal invasion,18,19 post-laminar invasion of the optic nerve head

with or without a positive margin,18,20 scleral invasion and extraocular extension (Table 1).1

Isolated massive choroidal invasion is controversial as a risk factor for metastasis. Shields et al.19

found that patients with choroidal invasion were more likely to develop metastases, but this was only

significant in the presence of concomitant optic nerve invasion. While some groups report significantly

reduced metastatic risk with postenucleation adjuvant therapy for massive choroidal invasion,21 others

report good prognosis without adjuvants.22,23 For isolated massive choroidal invasion, Kim24 estimated an

extraocular relapse rate of 5.5% versus 0% without and with adjuvant therapy, respectively. Similar

survival rates for focal versus massive choroidal invasion have been reported without adjuvants, provided

effective treatments for relapse are available.25 Toxicity associated with systemic adjuvant therapy

requires consideration during decision-making.26 In our center, massive choroidal invasion is considered a

HR feature requiring adjuvant chemotherapy.8

The first goal of treatment for advanced unilateral retinoblastoma is to save the child’s life and prevent

extraocular tumour dissemination; secondary goal is to save a seeing eye. Primary enucleation is a safe

and cheap option, allowing an early return to normal life,27 fewer interventions and EUAs,28 less

10

191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

214


socioeconomic impact,29 and histopathologic review of disease extent to guide further therapy. Primary

enucleation is current practice for cT3 (Group E) eyes and many cT2b (Group D) eyes (Group D). In our

cohort all parents accepted our recommendation for primary enucleation. Parental acceptance of

enucleation is influenced by physician education and counselling.30 Domingo et al. studied the

psychosocial factors affecting parental decision-making regarding enucleation and suggested that

“explanation on the course and prognosis of retinoblastoma may make the attitude of the parents more

favorable toward seeking surgery”.31

Potential for useful vision is important but salvage of a unilateral blind eye for cosmesis may not be

justified, given good prosthesis movement with myoconjunctival enucleation.32 Prolonged attempts at

globe salvage may delay diagnosis of HR features exposing the child to risk of metastasis. Pre-

enucleation chemotherapy may downstage pathological findings, delay enucleation and obstruct

recognition of HR disease.33 With no randomized controlled trial evidence to guide management of cT2

eyes, the clinician and family can balance the impact of potential years of trial salvage with hidden risks,33

against vision potential and quality of life for the child and family. Systemic chemotherapy, IAC, IVitC,

periocular chemotherapy and PPV are available for attempted eye salvage. The success of IVitC to

control vitreous disease9-11 justifies trial salvage for a unilateral cT2b eye, as long as metastatic risk is

minimal. The incidence of metastasis post-IAC is vague,5 as 3 metastatic deaths34 were reported as a sole

outcome for the collective data of 6 centers on 1177 eyes of 1139 patients without reporting how many

children had metastasis.35,36 This management debate clearly shows the value of predictors of either HR or

LR histopathology at diagnosis to facilitate decision making.

Clinical features at presentation reported to predict HR histopathology predominantly describe cT3

(Group E) eyes, not cT2b eyes, and include older age, symptoms >6 months, hyphema, pseudohypopyon,

orbital cellulitis, secondary glaucoma and buphthalmos.37-39 Furthermore, exophytic growth pattern, tumor

thickness >15 mm and vitreous hemorrhage predict optic nerve invasion,20 and iris neovascularization is

associated with choroidal invasion.19,40 Yousef et al.41 concluded that clinical staging alone (TNM 7th ed.,

11

215

216

217

218

219

220

221

222

223

224

225

226

227

228

229

230

231

232

233

234

235

236

237

238

239


IIRC or Reese Ellsworth classification) is insufficient to predict HR histopathology. This encouraged us

to study individual clinical findings as predictors of HR/LR histopathology. Useful predictive,

reproducible clinical findings would be those easily identified at initial staging EUA, the time of decision

of trial salvage versus primary enucleation. We excluded subjective findings such as presenting

complaint, duration of symptoms, pattern of growth (endophytic, exophytic or mixed) and presence of

subretinal seeding under detached retina. The value of MRI in predicting HR histopathology is

minimal.42,43

Approximately 2–33% of Group D eyes are expected to harbour HR histopathologic features in

primary enucleated eyes.27,39,41,44-47 However, the heterogeneous literature is limited by non-consensus in

defining HR histopathology features, variable classifications and inclusion of primarily and secondarily

enucleated eyes and children with unilateral and bilateral disease, and eyes with no staging recorded. In

our cohort of primarily enucleated unilateral cT2b eyes, 10.5% had HR histopathology. Macular

involvement, optic nerve obscuration and >1 quadrant of RD each was not helpful to predict HR

histopathology (predictive value 13%, 14%, or 14%, respectively). However, presence of macular

sparing, visible optic nerve and <1 quadrant of RD werewas each and any of their combinations

highly predictive of LR histopathology. with high specificity (100%). This suggests that such eyes are

appropriate for cautious trial salvage, especially if all three findings are present.

Fabian et al.46 reported on 40 primarily enucleated IIRC Group D eyes (all cT2b, 37 unilateral). At

presentation, 95% (38/40) had foveal involvement, 95% (38/40) had optic disc obscuration and 97% had

RD, compared to 82%, 74% and 74%, respectively in our cohort. They reported absence of vitreous seeds

as the sole significant predictor of HR based on p=0.042,46 whereas all eyes in our study had vitreous

seeds. Small sample sizes render tests of significance inaccurate as one extra entry can shift the p-value

significantly.

We used predictive values as well as significance tests to interpret our data. Significance tests showed

non-significant associations with wide confidence intervals due to the small sample size and low rate of

12

240

241

242

243

244

245

246

247

248

249

250

251

252

253

254

255

256

257

258

259

260

261

262

263

264


positive events. When we combined our data with Fabian et al.46 (78 eyes, Table 2), absence of vitreous

seeds was not significant (p=0.05) and other factors showed a lower p-value than reported by others.

When we applied predictive values, we showed visible optic nerve was 100% (95%CI 69% to 100%)

predictive of LR histopathology and non-involved fovea was 100% (95%CI 59% to 100%) predictive of

LR histopathology. we showed 100% predictive of LR histopathology with visible optic nerve and non-

involved fovea. Absence of vitreous seeds showed 71% (95%CI 48% to 95%) predictability of having

LR histopathology. Berry et al.47 reported that at diagnosis 15% of eyes with optic nerve obscuration

(69/102), and 0% with visible optic nerve (33/102) had post-laminar invasion following primary

enucleation, suggesting a possible clinicopathologic association. This is in accordance with 100%the

high predictability for LR histopathology with visible optic nerve in our cohort. The degree of optic

nerve obscuration could not be assessed as a predictor given only one eye demonstrated partial nerve

obscuration.

The mean age for unilateral retinoblastoma is 24 months.16 Children diagnosed at a younger age

(<24 months) showed 100% (95%CI 82% to 100%) predictability of LR eyes in our cohort. This is

probably due to earlier diagnosis at a less advanced intraocular stage. Fabian et al.46 showed similar

results after excluding the one HR group D eye diagnosed at 1 month of age as it was in the context of

bilateral retinoblastoma. Age should be treated cautiously as a clinical predictor since the subjective

duration of symptoms and delay in diagnosis might confound its value.

Our proposed clinical predictors are at initial diagnosis and there is no evidence to suggest their

usefulness in decision-making for recurrent or refractory cases. These predictors do not predict outcomes

as LR eyes at diagnosis are not guaranteed to remain LR if unresponsive to initial treatment. The decision

of extended duration trial salvage may depend on realistic outcomes including metastatic risks, treatment

morbidity, socioeconomic impact and visual potential taking into consideration that the other eye is

perfectly normal.

13

265

266

267

268

269

270

271

272

273

274

275

276

277

278

279

280

281

282

283

284

285

286

287

288


The limitations of this study include its retrospective design and relatively small sample size.

However, our stringent inclusion criteria of only unilateral, primarily enucleated Group D eyes achieved a

homogenous study population. Another limitation was the low rate of positive events, which limits

statistical analysis of associations between clinical findings and histopathologic features of the included

eyes. Furthermore, our analysis is relevant only to the point of diagnosis and does not address progress of

histopathologic risk with extended treatments.

Conclusion

Macular sparing, optic nerve visibility and <1 quadrant of RD at presentation were highly predictive of

LR, and may predict which advanced eyes are suitable for trial salvage decision. HR histopathology was

found in 10.5% of primarily enucleated unilateral cT2b (Group D) eyes. No clinical findings were found

to be predictive of HR histopathology. Previous reports have demonstrated acceptable salvage rates in

unilateral Group D eyes. We recommend taking into consideration the previous aforementioned factors

before undertaking a decision for more cautious trial salvage in eyes with macular involvement, optic

nerve obscuration and >1 quadrant RD. Debates on management of unilateral cT2 eyes would be solved

by robust, multicenter collaborative studies involving many children to establish clinicopathologic

correlations and multifaceted outcomes.

References

1. 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.

2. Murphree AL. Intraocular retinoblastoma: the case for a new group classification.

Ophthalmology clinics of North America. 2005;18:41-53.

14

289

290

291

292

293

294

295

296

297

298

299

300

301

302

303

304

305

306

307

308

309

310


3. Munier FL, Mosimann P, Puccinelli F, et al. First-line intra-arterial versus intravenous

chemotherapy in unilateral sporadic group D retinoblastoma: evidence of better visual outcomes,

ocular survival and shorter time to success with intra-arterial delivery from retrospective review

of 20 years of treatment. Br J Ophthalmol. 2016.

4. Shields CL, Manjandavida FP, Lally SE, et al. Intra-arterial chemotherapy for retinoblastoma in

70 eyes: outcomes based on the international classification of retinoblastoma. Ophthalmology.

2014;121(7):1453-1460.

5. Yousef YA, Soliman SE, Astudillo PP, et al. Intra-arterial Chemotherapy for Retinoblastoma: A

Systematic Review. JAMA ophthalmology. 2016;134(6):584-591.

6. Gobin YP, Dunkel IJ, Marr BP, Brodie SE, Abramson DH. Intra-arterial chemotherapy for the

management of retinoblastoma: four-year experience. Arch Ophthalmol. 2011;129(6):732-737.

7. Suzuki S, Yamane T, Mohri M, Kaneko A. Selective ophthalmic arterial injection therapy for

intraocular retinoblastoma: the long-term prognosis. Ophthalmology. 2011;118(10):2081-2087.

8. 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.

9. 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.

10. 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.

11. Berry JL, Shah S, Bechtold M, Zolfaghari E, Jubran R, Kim JW. Long-term outcomes of Group

D retinoblastoma eyes during the intravitreal melphalan era. Pediatr Blood Cancer. 2017.

15

311

312

313

314

315

316

317

318

319

320

321

322

323

324

325

326

327

328

329

330

331

332

333

334


12. Mallipatna AC, Dimaras H, Chan HS, Heon E, Gallie BL. Periocular topotecan for intraocular

retinoblastoma. Arch Ophthalmol. 2011;129(6):738-745.

13. Zhao J, Li Q, Wu S, et al. Pars Plana Vitrectomy and Endoresection of Refractory Intraocular

Retinoblastoma. Ophthalmology. 2018;125(2):320-322.

14. Sastre X, Chantada GL, Doz F, et al. Proceedings of the consensus meetings from the

International Retinoblastoma Staging Working Group on the pathology guidelines for the

examination of enucleated eyes and evaluation of prognostic risk factors in retinoblastoma.

Archives of pathology & laboratory medicine. 2009;133(8):1199-1202.

15. Soliman SE, VandenHoven C, MacKeen LD, Heon E, Gallie BL. Optical Coherence

Tomography-Guided Decisions in Retinoblastoma Management. Ophthalmology.

2017;124(6):859-872.

16. Soliman SE, Racher H, Zhang C, MacDonald H, Gallie BL. Genetics and Molecular Diagnostics

in Retinoblastoma--An Update. Asia Pac J Ophthalmol (Phila). 2017;6(2):197-207.

17. Racher H, Soliman S, Argiropoulos B, et al. Molecular analysis distinguishes metastatic disease

from second cancers in patients with retinoblastoma. Cancer Genet. 2016;209(7-8):359-363.

18. Khelfaoui F, Validire P, Auperin A, et al. Histopathologic risk factors in retinoblastoma: a

retrospective study of 172 patients treated in a single institution. Cancer. 1996;77(6):1206-1213.

19. Shields CL, Shields JA, Baez KA, Cater J, De Potter PV. Choroidal invasion of retinoblastoma:

metastatic potential and clinical risk factors [see comments]. British Journal of Ophthalmology.

1993;77(9):544-548.

20. Shields CL, Shields JA, Baez K, Cater JR, De-Potter P. Optic nerve invasion of retinoblastoma.

Metastatic potential and clinical risk factors. Cancer. 1994;73(3):692-698.

16

335

336

337

338

339

340

341

342

343

344

345

346

347

348

349

350

351

352

353

354

355

356


21. Honavar SG, Singh AD, Shields CL, et al. Postenucleation adjuvant therapy in high-risk

retinoblastoma. Arch Ophthalmol. 2002;120(7):923-931.

22. Uusitalo MS, Van Quill KR, Scott IU, Matthay KK, Murray TG, O'Brien JM. Evaluation of

chemoprophylaxis in patients with unilateral retinoblastoma with high-risk features on

histopathologic examination. Arch Ophthalmol. 2001;119(1):41-48.

23. Chantada GL, Dunkel IJ, de Davila MT, Abramson DH. Retinoblastoma patients with high risk

ocular pathological features: who needs adjuvant therapy? Br J Ophthalmol. 2004;88(8):1069-

1073.

24. Kim JW. Retinoblastoma: evidence for postenucleation adjuvant chemotherapy. Int Ophthalmol

Clin. 2015;55(1):77-96.

25. Bosaleh A, Sampor C, Solernou V, et al. Outcome of children with retinoblastoma and isolated

choroidal invasion. Arch Ophthalmol. 2012;130(6):724-729.

26. Perez V, Sampor C, Rey G, et al. Treatment of Nonmetastatic Unilateral Retinoblastoma in

Children. JAMA ophthalmology. 2018;136(7):747-752.

27. Mallipatna AC, Sutherland JE, Gallie BL, Chan H, Heon E. Management and outcome of

unilateral retinoblastoma. J AAPOS. 2009;13(6):546-550.

28. Fabian ID, Stacey AW, Johnson KC, et al. Primary enucleation for group D retinoblastoma in the

era of systemic and targeted chemotherapy: the price of retaining an eye. Br J Ophthalmol. 2017.

29. Soliman SE, Dimaras H, Souka AA, Ashry MH, Gallie BL. Socioeconomic and psychological

impact of treatment for unilateral intraocular retinoblastoma. Journal Francais D Ophtalmologie.

2015;38:550—558.

30. Olteanu C, Dimaras H. Enucleation Refusal for Retinoblastoma: A Global Study. Ophthalmic

Genet. 2014:1-7.

17

357

358

359

360

361

362

363

364

365

366

367

368

369

370

371

372

373

374

375

376

377

378

379


31. Domingo RED, Toledo MSW, Mante BVL. Psychosocial Factors Influencing Parental Decision

to Allow or Refuse Potentially Lifesaving Enucleation in Children with Retinoblastoma. Asia

Pac J Oncol Nurs. 2017;4(3):191-196.

32. Shome D, Honavar SG, Raizada K, Raizada D. Implant and prosthesis movement after

enucleation: a randomized controlled trial. Ophthalmology. 2010;117(8):1638-1644.

33. Zhao J, Dimaras H, Massey C, et al. Pre-enucleation chemotherapy for eyes severely affected by

retinoblastoma masks risk of tumor extension and increases death from metastasis. Journal of

clinical oncology : official journal of the American Society of Clinical Oncology. 2011;29(7):845-

851.

34. Abramson DH, Shields CL, Jabbour P, et al. Metastatic deaths in retinoblastoma patients treated

with intraarterial chemotherapy (ophthalmic artery chemosurgery) worldwide. Int J Retina

Vitreous. 2017;3:40.

35. Soliman SE, Dimaras H, Gallie B, Shaikh F. Re: Metastatic deaths in retinoblastoma patients

treated with intraarterial chemotherapy (ophthalmic artery chemosurgery) worldwide. Int J

Retina Vitreous. 2018;4:19.

36. Soliman SE, Gallie BL, Shaikh F. Intra-arterial Chemotherapy for Retinoblastoma-Reply. JAMA

ophthalmology. 2016;134(10):1203.

37. Kashyap S, Meel R, Pushker N, et al. Clinical predictors of high risk histopathology in

retinoblastoma. Pediatr Blood Cancer. 2012;58(3):356-361.

38. Chantada GL, Gonzalez A, Fandino A, et al. Some clinical findings at presentation can predict

high-risk pathology features in unilateral retinoblastoma. J Pediatr Hematol Oncol.

2009;31(5):325-329.

39. Kaliki S, Srinivasan V, Gupta A, Mishra DK, Naik MN. Clinical Features Predictive of High-

Risk Retinoblastoma in 403 Asian Indian Patients: A Case-Control Study. Ophthalmology. 2015.

18

380

381

382

383

384

385

386

387

388

389

390

391

392

393

394

395

396

397

398

399

400

401

402

403


40. Chawla B, Sharma S, Sen S, et al. Correlation between clinical features, magnetic resonance

imaging, and histopathologic findings in retinoblastoma: a prospective study. Ophthalmology.

2012;119(4):850-856.

41. Yousef YA, Al-Hussaini M, Mehyar M, et al. Predictive Value of Tnm Classification,

International Classification, and Reese-Ellsworth Staging of Retinoblastoma for the Likelihood of

High-Risk Pathologic Features. Retina. 2015.

42. Hiasat JG, Saleh A, Al-Hussaini M, et al. The predictive value of magnetic resonance imaging of

retinoblastoma for the likelihood of high-risk pathologic features. Eur J Ophthalmol.

2018:1120672118781200.

43. de Jong MC, de Graaf P, Noij DP, et al. Diagnostic performance of magnetic resonance imaging

and computed tomography for advanced retinoblastoma: a systematic review and meta-analysis.

Ophthalmology. 2014;121(5):1109-1118.

44. Wilson MW, Qaddoumi I, Billups C, Haik BG, Rodriguez-Galindo C. A clinicopathological

correlation of 67 eyes primarily enucleated for advanced intraocular retinoblastoma. The British

journal of ophthalmology. 2011;95(4):553-558.

45. Kaliki S, Tahiliani P, Mishra DK, Srinivasan V, Ali MH, Reddy VA. OPTIC NERVE

INFILTRATION BY RETINOBLASTOMA: Predictive Clinical Features and Outcome. Retina.

2016;36(6):1177-1183.

46. Fabian ID, Stacey AW, Chowdhury T, et al. High-Risk Histopathology Features in Primary and

Secondary Enucleated International Intraocular Retinoblastoma Classification Group D Eyes.

Ophthalmology. 2017;124(6):851-858.

47. Berry JL, Kogachi K, Jubran R, Kim JW. Loss of fundus view as an indication for secondary

enucleation in retinoblastoma. Pediatr Blood Cancer. 2017.

19

404

405

406

407

408

409

410

411

412

413

414

415

416

417

418

419

420

421

422

423

424

425

426


Tables

Table 1. American Joint Committee on Cancer (AJCC) pathological staging 8th Edition.

Table 2. Predictive values versus significance (p-value) in analysis of Fabian et al 46 and current sample.

Supplementary Table 1. Clinicopathologic features of the studied sample (n=38).

Supplementary Table 2. Significance of association between clinical findings and histopathologic features

of enucleated eyes.

Figure Legend

Figure 1. (A) Left, wide-field fundus image showing a right multilobulated tumor with overlying

serous retinal detachment (RD) and subretinal seeding with obscured optic nerve and macular

involvement with tumor. Middle left, Histopathological section under low magnification through the optic

nerve demonstrating extra-scleral (yellow-box) and post-lamina cribrosa invasion (green-box), but not to

the optic nerve resection margin (pT4). Middle right, high magnification of trans-scleral and extra-scleral

invasion (arrowheads). Right, High magnification showing post-laminar invasion (arrowheads). Whole-

body MRI (WBMRI), lumbar puncture (LP) and bilateral bone marrow aspirate (BMA) were negative for

malignancy. He underwent six cycles of vincristine, etoposide, carboplatin (VEC) and cyclophosphamide,

followed by orbital irradiation. Child is alive and disease free after 3 years follow-up.

(B) Left, wide-field fundus image demonstrating a large inferior tumor with overlying RD, vitreous

and subretinal seeds. The optic nerve was obscured and the macular involved with tumor. Middle and

right, Histopathological sections under low and intermediate magnification showing massive choroidal

invasion (arrowheads) beyond the confines of the retinal pigment epithelium (RPE), with no evidence of

scleral invasion (pT3a). There was prelaminar optic nerve invasion. LP and BMA were negative for

malignancy and she received four cycles of VEC. The child is alive and disease free after 3 years follow

up.

20

427

428

429

430

431

432

433

434

435

436

437

438

439

440

441

442

443

444

445

446

447

448

449


(C) Left, wide-field fundus image of demonstrating a large tumor obscuring the nerve and macula,

with associated hemorrhage, RD and diffuse vitreous seeding. There was no anterior segment extension

evident on ultrasound biomicroscopy. Histopathology was confirmed to be massive choroidal invasion

(arrowheads) (pT3a). The child received systemic adjuvant chemotherapy and alive after 29 months

follow up.

(D) Left, wide-field fundus image demonstrating a large tumor with associated RD, subretinal and

focal vitreous seeding. There was no visualization of the optic nerve and the macula was involved. Initial

review was consistent with low-risk histopathology. One year later the child presented with fever and

pain, and WBMRI identified a paraspinal tumor. Molecular analysis confirmed metastasis.17 Bilateral

BMA were involved with tumor cells. MRI showed no orbital or intracranial disease and LP was

negative. Internal review of the ocular pathology, including further choroidal sections, showed an area of

massive choroidal invasion (yellow-box) (pT3a). The child received 6 cycles of VEC and cyclosporine,

followed by autologous bone marrow transplant and focal irradiation. He was diagnosed with dural-based

metastases 1 year later. Despite radiotherapy, the child died 18 months after presentation with metastases.

LC– lamina cribrosa, S – sclera, RPE – retinal pigment epithelium, arrowheads denotes tumor

invasion boundaries, blue boxes denotes highlighted areas massive choroidal and/or scleral invasion,

green box denotes highlighted areas of optic nerve invasion

Figure 2: Schematic presentation of the predictive clinical signs of low-risk histopathology and their

clinicopathologic correlation.

21

450

451

452

453

454

455

456

457

458

459

460

461

462

463

464

465

466

467

468

469

Methods · Web viewWhole-body MRI (WBMRI), lumbar puncture (LP) and bilateral bone marrow aspirate...

Documents

Transcript of Methods · Web viewWhole-body MRI (WBMRI), lumbar puncture (LP) and bilateral bone marrow aspirate...