VOLUME

29

NUMBER

MARCH

2011

JOURNAL OF CLINICAL ONCOLOGY

O R I G I N A L

R E P O R T

Pre-Enucleation Chemotherapy for Eyes Severely Affected

by Retinoblastoma Masks Risk of Tumor Extension and
Increases Death From Metastasis
Junyang Zhao, Helen Dimaras, Christine Massey, Xiaolin Xu, Dongsheng Huang, Bin Li, Helen S.L. Chan,
and Brenda L. Gallie
From the Beijing Tongren Eye Centre,
Beijing Tongren Hospital, Capital Medical University, Beijing, China; The
Hospital for Sick Children; Campbell
Family Institute for Cancer Research,
Ontario Cancer Institute/Princess
Margaret Hospital, University Health
Network; and the University of Toronto,
Toronto, Ontario, Canada.
Submitted September 7, 2010;
accepted December 10, 2010;
published online ahead of print at
www.jco.org on January 31, 2011.
Supported by the Ontario Ministry of
Health and Long Term Care
(OMOHLTC); the Kalmar Family Trust
for the One Retinoblastoma World
Network; and the Canadian Retinoblastoma Society. The views expressed do
not necessarily reflect those of the
OMOHLTC.
Authors disclosures of potential conflicts of interest and author contributions are found at the end of this
article.
Corresponding author: Brenda L. Gallie,
MD, FRCSC, Campbell Family Cancer
Research Institute, Ontario Cancer Institute/Princess Margaret Hospital, University Health Network, University of
Toronto, 610 University Ave, Toronto,
Ontario, Canada M5G 2M9; e-mail:
brenda@gallie.ca.
2011 by American Society of Clinical
Oncology

Purpose
Initial response of intraocular retinoblastoma to chemotherapy has encouraged primary chemotherapy instead of primary enucleation for eyes with clinical features suggesting high risk of
extraocular extension or metastasis. Upfront enucleation of such high-risk eyes allows pathologic
evaluation of extraocular extension, key to management with appropriate surveillance and
adjuvant therapy. Does chemotherapy before enucleation mask histologic features of extraocular
extension, potentially endangering the childs life by subsequent undertreatment?
Methods
We performed retrospective analysis of 100 eyes with advanced retinoblastoma enucleated with,
or without, primary chemotherapy, in Beijing Tongren Hospital, retrospectively, from October 31,
2008. The extent of retinoblastoma invasion into optic nerve, uvea, and anterior chamber on
histopathology was staged by pTNM classification. The treatment groups were compared for
pathologic stage (Cochran-Armitage trend test) and disease-specific mortality (competing
risks methods).
Results
Children who received chemotherapy before enucleation had lower pTNM stage than primarily
enucleated children (P .01). Five patients who received pre-enucleation chemotherapy died as
a result of extension into brain or metastasis. No patients who had primary enucleation died. For
children with group E eyes, disease-specific survival (DSS) was lower with pre-enucleation
chemotherapy (n 45) than with primary enucleation (n 37; P .01). Enucleation longer than
3 months after diagnosis was also associated with lower DSS (P .001).
Conclusion
Chemotherapy before enucleation of group E eyes with advanced retinoblastoma downstaged
pathologic evidence of extraocular extension, and increased the risk of metastatic death from
reduced surveillance and inappropriate management of high-risk disease, if enucleation was
performed longer than 3 months after diagnosis.
J Clin Oncol 29:845-851. 2011 by American Society of Clinical Oncology

0732-183X/11/2907-845/$20.00
DOI: 10.1200/JCO.2010.32.5332

INTRODUCTION

Metastasis of retinoblastoma is rare in the developed

world, where early diagnosis achieves cure for most
children.1 Cure of retinoblastoma outside the eye is
rare. Metastatic retinoblastoma in treated with systemic and intrathecal/intraventricular chemotherapy, stem-cell transplantation, and in some cases
radiation, with some success.2-8
Retinoblastoma confined to the eye can be
cured by simple enucleation, and usually no further
treatment is necessary. The enucleated eye is studied
for histopathologic evidence of tumor invasion to
evaluate risk of tumor spread outside the eye, classi-

fied by the American Joint Committee on Cancer

(AJCC) pTNM classification.9 Patients displaying
unfavorable pathology may be treated with adjuvant chemotherapy and/or active surveillance to
identify extraocular retinoblastoma when cure is
still feasible.10
Intraocular retinoblastoma groups A to D, International Intraocular Retinoblastoma Classification (IIRC),11 can initially respond dramatically to
chemotherapy. This has resulted in a trend to also
use chemotherapy for more severely affected eyes in
IIRC group E, which exhibit clinical signs associated
with high risk for extraocular disease. This approach
is often attributed to parental reluctance to remove
2011 by American Society of Clinical Oncology

Downloaded from jco.ascopubs.org on March 27, 2013. For personal use only. No other uses without permission.
Copyright 2011 American Society of Clinical Oncology. All rights reserved.

845

Zhao et al

the childs eye, or the expectation that chemotherapy might make the
surgical removal of the eye safer. However, chemotherapy may downstage pathologic features associated with risk for extraocular extension, which otherwise would suggest adjuvant therapy and/or
surveillance. This could result in undertreatment and loss of opportunity for cure.
We performed a retrospective case study of 100 enucleated IIRC
group D and E eyes that were either primarily enucleated or treated
with prior primary chemotherapy, to determine the effect of preenucleation chemotherapy on the histologic detection of risk for extraocular extension and patient outcome.

METHODS
Data Collection and Ethics
All data were collected retrospectively with approval from the Capital
Medical University, Beijing Tongren Hospital Ethics Board. Data was analyzed
with approval of the research ethics board of the University Health Network,
Toronto, Canada. Anonymized clinical data were recorded including sex, date
of birth, age at diagnosis, disease laterality (unilateral/bilateral), IIRC group at
diagnosis, family history of retinoblastoma, follow-up duration, chemotherapy regimen, evidence of extraocular extension, and date of death.
Inclusion and Exclusion Criteria
Patients were selected based on the IIRC group of their enucleated eye
confirmed at the initial staging examination under anesthesia, and date of
enucleation. The 100 IIRC group D or E eyes enucleated at the Beijing Tongren
Hospital, Beijing, China, between May 19, 2006, and October 31, 2008, were
included in this study. The eyes were enucleated either as primary treatment or
after chemotherapy.
No patient had clinical evidence of metastasis at presentation by lumbar puncture (LP), bone marrow aspiration, and MRI and/or computed
tomography. Approximately 24 to 48 hours before each chemotherapy
cycle, examination under anesthesia, and prechemotherapy cryotherapy
(single freeze-thaw to one or two spots in peripheral retina) were performed to
enhance chemotherapy penetration into the eye.12 Children were excluded
from this study if there was evidence of clinical metastasis at presentation, or if
they had been treated with chemotherapy plus focal laser and/or therapeutic
triple freeze-thaw cryotherapy, since this indicates more extensive effort to
save the eye than simple pre-enucleation chemotherapy.
Histologic Assessment
The extent of tumor in optic nerve, choroid and anterior chamber were
scored based on the written histopathology reports and confirmed by retrospective rereview of hematoxylin and eosinstained slides. The AJCC pTNM9
stage was for each eye was based on extent of tumor invasion into optic nerve,
choroid, and/or anterior chamber. Evidence of metastasis was collected from
MRI, computed tomography, LP, and/or bone marrow aspiration reports. For
some analyses, children were grouped into low (pT1), moderate (pT2), or high
(pT3, pT4) risk for metastasis (Table 1).
Statistical Analysis
Disease-specific survival (DSS) was measured from date of diagnosis to
date of death from disease in the study eye. Death not due to disease or not
from the study eye was treated as a competing risk. Patients alive were censored
at last follow-up. Univariable analyses of DSS were conducted using Grays test
and plots of 1 minus the cumulative incidence function. Other univariable
analyses were conducted using Fishers exact test, the Wilcoxon rank sum
test, and the Cochran-Armitage trend test, and all P values reported for these
tests are exact P values. P values reported for all tests were two sided. There was
no missing data. Competing risks analyses13 were performed using the cmprsk
package in R 2.10.1 (http://www.r-project.org/). All other analysis was generated using SAS version 9.2 (SAS Institute, Cary, NC).
846

2011 by American Society of Clinical Oncology

Table 1. pTNM Risk for Metastasis

Risk and pTNM
Low
pTX
pT0
pT1
Moderate
pT2a
pT2a
pT2b

High
pT3a
pT3a
pT3b

pT4a
pT4b

Definition
Primary tumor cannot be assessed
No evidence of primary tumor
Tumor confined to eye with no optic nerve or
choroidal invasion
Tumor superficially invades optic nerve head
but does not extend past lamina cribrosa
Tumor exhibits focal choroidal invasion
Tumor superficially invades optic nerve head
but does not extend past lamina cribrosa
and exhibits focal choroidal invasion
Tumor invades optic nerve past lamina
cribrosa but not to surgical resection line
Tumor exhibits massive choroidal invasion
Tumor invades optic nerve past lamina
cribrosa but not to surgical resection line
and exhibits massive choroidal invasion
Tumor invades optic nerve to resection line
but no extra-ocular extension identified
Tumor invades optic nerve to resection line
and extra-ocular extension identified

RESULTS

Patients
Histopathology reports of 100 enucleated retinoblastoma eyes
from 100 children (61 male, 39 female; 28 bilateral, 72 unilateral; 18
group D, 82 group E) were retrospectively staged for pTNM staging
features. Re-examination of histopathology by a second reviewer
(B.L.G.) showed 100% concurrence with low-, moderate-, and highrisk staging (Table 1).
Enucleation was the primary treatment for 45 eyes. Chemotherapy was the primary treatment for 55 eyes enucleated without
other treatments. The two groups showed no significant differences between sex, unilateral versus bilateral, IIRC group of the
eye(s), or age at diagnosis (Table 2). The bilaterally affected child
who died due to disease was diagnosed at age 24 months (overall
median age bilateral diagnosis 15 months; range, 4 to 77); the
unilaterally affected children who died were diagnosed at ages 9,
35, and 37 months (overall median age unilateral diagnosis 27
months; range, 4 to 77). Median follow-up was longer for patients
alive at last follow-up treated with primary enucleation (median,
31.2 months; minimum/maximum 10.3 to 44.9) than with preenucleation chemotherapy (median, 25 months; minimum/maximum 11.5 to 38.4; Table 2).
Only one of 28 bilateral patients (patient 33) had bilateral group
E disease (Table 2, Fig 1). The study eye was removed at diagnosis.
After enucleation, five cycles of chemotherapy were given, but the
child was lost to follow-up 19 months after initial diagnosis, so there is
no further outcome for the retained group E eye or current status. The
remaining patients had groups A to D disease in the remaining eyes
(Fig 1). Each patient contributed only one eye to this study.
JOURNAL OF CLINICAL ONCOLOGY

Downloaded from jco.ascopubs.org on March 27, 2013. For personal use only. No other uses without permission.
Copyright 2011 American Society of Clinical Oncology. All rights reserved.

Increased Mortality From Retinoblastoma

Table 2. Features of Children and Enucleated Eyes

Pre-Enucleation Chemotherapy
No
Parameter

Yes

No.

No.

Sex
Female
Male

17
28

38
62

22
33

40
60

.84

Group worst eye

D
E

8
37

18
82

10
45

18
82

1.00

Laterality
Unilateral
Bilateral

33
12

73
27

39
16

71
29

.83

Age at diagnosis, months

D
Unilateral
Bilateral
Total
E
Unilateral
Bilateral
Total
All patients
Unilateral patients
Bilateral patients
Diagnosis to enucleation, months
D
Unilateral
Bilateral
Total
E
Unilateral
Bilateral
Total
All patients
Follow up, months

No.

6
2
8

13
4
18

27
10
37
45
33
12

60
22
82

6
2
8

13
4
18

27
10
37
45
44

60
22
82

Minimum to
Maximum

No.

Median

Minimum to
Maximum

31
6
27

8-37
4-9
4-37

6
4
10

11
7
18

23
10
19

9-36
3-16
3-36

27
15
26
26
28
13

4-77
1-41
1-77
1-77
4-77
1-41

33
12
45
55
39
16

60
22
82

27
21
25
23
27
19

2-122
2-41
2-122
2-122
2-122
3-41

0.3
0.5
0.3

0-12
0-1
0-12

6
4
10

11
7
18

9
6
9

2-14
0.8-13
0.8-14

0.1
0.4
0.1
0.1
31

0-12
0-3
0-12
0-12
10-45

33
12
45
55
50

60
22
82

3
4
3
3
25

0.2-19
0.9-18
0.2-19
0.2-19
12-38

Median

.92
.73
.37

.001

Fishers exact test.

Wilcoxon rank sum test.
Follow-up for patients alive at last follow-up.

Pre-Enucleation Chemotherapy
Pre-enucleation chemotherapy every 3 to 4 weeks consisted of
one to 12 cycles (median, 3 cycles) of carboplatin, etoposide, and
vincristine/carboplatin, teniposide, and vincristine, with cyclosporine
doses lower than therapeutic doses previously published.14,15 Eyes
were removed when the tumors progressed with no possibility of
useful vision.
Time From Diagnosis to Enucleation
Median time from diagnosis to enucleation was 0.1 months
for primarily enucleated eyes, and 3 months for eyes that received
pre-enucleation chemotherapy (P .001; Table 2). Group E eyes
that received pre-enucleation chemotherapy were removed sooner
(median, 2.8 months) than group D eyes (median, 8.6 months),
indicating the poor response of group E eyes to therapeutic intervention (Table 2).
www.jco.org

Primary Outcomes
Five children died of metastasis. No children who had primary
enucleation (Figs 1, 2), and no child with a group D eye in the
study, died of tumor. For group E eyes in the chemotherapy group,
four children (8.9%) died due to disease in the study eye by last
follow-up. For all children who died, the time from diagnosis until
enucleation was longer than 3 months. In the no chemotherapy
group, none of the patients died due to disease by last follow-up,
one patient (2.7%) experienced a competing risk (died from chemotherapy complications).
Patient 22 presented with leukocoria and bilateral groups E/D
retinoblastoma at age 21 months. The group E eye was removed after
2.5 months after seven cycles of chemotherapy. No adverse features
were present on pathology (pT1). Four more cycles of chemotherapy
were given for the group D eye. The optic nerve was found enlarged 1.5
years after diagnosis, further treatment was refused, and the child died
2011 by American Society of Clinical Oncology

Downloaded from jco.ascopubs.org on March 27, 2013. For personal use only. No other uses without permission.
Copyright 2011 American Society of Clinical Oncology. All rights reserved.

847

Zhao et al

Post-Enucleation Chemotherapy
(No. of cycles)
1

10 11 12

A
L
M
L

0.6

0.4

0.2

Enucleation

100

100

100

Pre-enucleation
chemotherapy (n = 45)

100

95

NA

10

20

30

40

Time From Diagnosis (months)

1.0

Disease-Specific Survival (%)

at 12 months
at 24 months
at 36 months
No pre-enucleation
chemotherapy (n = 37)

Gray's test P = .01

0.8

0.6

0.4

3 months between
diagnosis and enucleation (n = 58)

0.2

> 3 months between diagnosis

and enucleation (n = 24)

Disease-Specific Survival (%)

at 12 months at 24 months at 36 months
100

100

100

100

90

Gray's test P < .001

10

20

30

40

Time From Diagnosis (months)

Fig 2. Disease-specific survival (DSS) for children with enucleated group E eyes.
(A) DSS for children with group E eyes who received chemotherapy before
enucleation (n 45) compared with those who did not (n 37). (B) DSS for
children whose eyes were enucleated longer than 3 months (n 24) or 3
months (n 58) from diagnosis.

Fig 1. Cycles of pre-enucleation and postenucleation chemotherapy. Patients

(identified by numbers) sorted by laterality (unilateral/bilateral), risk of extraocular
disease on pathology (L low, white background; M moderate, light gold
background; or H high, dark gold background) and number of chemotherapy
cycles (blue bars, range from 0 to 12 cycles). Patients who died as a result of
disease are indicated in red outline with patient number in red. IIRC, International
Intraocular Retinoblastoma Classification.

2 years after diagnosis with brain metastases from the group D (nonstudy) eye, considered a competing risk.
Patient 38 was diagnosed age 24 months with bilateral groups
E/D retinoblastoma 9 months after leukocoria was first noticed. After
eight cycles of chemotherapy, the group E eye was removed. Pathology
showed tumor at the cut end of optic nerve (pT4b). Despite six more
cycles of chemotherapy, the child died of brain metastasis, 2 years
after diagnosis.
Patient 68 was diagnosed at age 35 months with unilateral group
E retinoblastoma. Despite six cycles of chemotherapy and externalbeam radiotherapy, the tumor was not controlled and the eye was
removed. Pathology showed moderate risk (tumor past the lamina
848

0.8

Disease-Specific Survival

Bilateral

D
A
D
D
B
C
D
D
D
B
B
A
D
B
E
A
D
D
D
D
B
B

1.0

Disease-Specific Survival

Enucleation

Bilateral
IIRC D
Unilateral

IIRC E

19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100

B
A
D
C
B
A

11 10

Unilateral

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18

12

Risk

Pre-Enucleation Chemotherapy
(No. of cycles)
OTHER
EYE

2011 by American Society of Clinical Oncology

cribrosa but not to the cut end of optic nerve, pT2a). Despite adjuvant
therapy of three cycles of chemotherapy and six doses of intrathecal
chemotherapy, the child died of brain and spinal metastases 2.5 years
after diagnosis.
Patient 93 was diagnosed with unilateral group E retinoblastoma at age 37 months. The eye was removed after four cycles of
chemotherapy. Pathology showed tumor at the cut end of optic
nerve and massive choroidal invasion (pT4b). Seven more cycles of
chemotherapy were given but the child died of brain metastasis 1.5
years after diagnosis.
Patient 94 was diagnosed at age 9 months with unilateral group E
retinoblastoma. Six cycles of chemotherapy were given, but recurrent
intraocular disease prompted enucleation. Pathology showed no tumor past the lamina cribrosa of the optic nerve but massive choroidal
and scleral invasion (pT4b). Three more cycles of chemotherapy were
given but the child died 1 year after diagnosis.
The childrens survival times were measured from date of diagnosis. Because time intervals between diagnosis and enucleation were
longer for patients with pre-enucleation chemotherapy (3.4 v 0.1
months for patients without pre-enucleation chemotherapy), there
is a design bias, favoring children who received pre-enucleation
chemotherapy with regard to survival time. If pre-enucleation chemotherapy had no effect on survival, survival times for patients with
JOURNAL OF CLINICAL ONCOLOGY

Downloaded from jco.ascopubs.org on March 27, 2013. For personal use only. No other uses without permission.
Copyright 2011 American Society of Clinical Oncology. All rights reserved.

Increased Mortality From Retinoblastoma

pre-enucleation chemotherapy would tend to be longer than for those

without, simply because they had to live through longer diagnosisuntil-enucleation periods to be included in the study. Despite this
built-in bias, the DSS was significantly lower for children with group E
eyes who received chemotherapy before enucleation than for those
who did not (Fig 2A; 24 month DSS was 95% and 100%, respectively,
Grays test P .01), or if enucleation was delayed by longer than 3
months (Fig 2B; 24 month DSS was 90% and 100%, respectively,
Grays test P .001).
In the longer than 3 months group, four patients (16.7%) died
due to disease by last follow-up, and one patient died from disease
from the other eye (competing risk). The remaining patients were
censored. In the 0- to 3-month group, none of the patients died due to
disease by last follow-up, one patient (1.7%) experienced a competing
risk and the remaining 57 patients (98.3%) were censored at last
follow-up. The median time for DSS cannot be estimated for the 0- to
3-month group since there has not been enough follow-up for their
cumulative incidence function to reach 0.50. The estimated median
DSS for the longer than 3 month group is 30.6 months.
Similarly, the number of cycles of chemo for all pre-enucleation
chemotherapy group E patients significantly affected DSS, comparing
outcome for those who received more than four cycles to those who
received 4 cycles (Grays test P .01). The median time to diseasespecific death in the more than four cycles group was 30.6 months.
Histologic Risk Factors of Metastasis
Overall, pre-enucleation chemotherapy was associated with
lower risk staging (P .01) with the pTNM stage collapsed into low-,
moderate-, and high-risk groups (Table 1). This was significant for the
D eyes (P .02), but a similar trend for E eyes was not statistically
significant (P .09; Table 3). Patient 68 died of intracranial extension
of unilateral retinoblastoma, yet the enucleated eye showed a risk level
of pT2a, which would not suggest adjuvant treatment. For the preenucleation chemotherapy group, pTNM risk (low/moderate/high)
was negatively correlated with time between diagnosis and enucleation (Spearman rank correlation coefficient 0.30; 95% CI, 0.47
to 0.11; P .002).

Table 3. Pre-Enucleation Chemotherapy and Risk of Metastasis by pTNM

Pre-Enucleation Chemotherapy
pTNM Risk for
Metastasis
All eyes
High
Moderate
None
Group D
High
Moderate
None
Group E
High
Moderate
None

www.jco.org

No

Yes

No.

No.

9
31
5

20
69
11

7
28
20

13
51
36

.01

1
7
0

12
88
0

0
5
5

0
50
50

.02

8
24
5

22
65
14

7
23
15

16
51
33

.09

Post-Enucleation Chemotherapy
Seventy-four percent (74 of 100) of patients received chemotherapy (median four cycles; minimum, one cycle; maximum, 12 cycles)
after enucleation (Fig 1). The postenucleation chemotherapy regimen
was the same as the pre-enucleation chemotherapy regimen. Overall,
treatment with postenucleation chemotherapy correlated to degree
(low/moderate/high) of histologic risk (P .02), however, the number of cycles given was not consistent for the pTNM the risk groups.
Fifty-two percent (13 of 25) of children with low-risk pathology and
81.3% (48 of 59) with moderate-risk pathology received postenucleation chemotherapy, although no published data suggests that low or
moderate-risk justifies adjuvant therapy.16 Nineteen percent (three of
16) of children with high-risk pathology did not receive postenucleation chemotherapy, which conventionally might be clinically indicated.16 Univariable tests did not suggest that the decision to give
postenucleation chemotherapy was associated with laterality or
IIRC group.

DISCUSSION

Next to India, China has the most new patients with retinoblastoma.1,17 Accrual of such a large a cohort of enucleated eyes in a short
period of time (100 eyes in 2.5 years) is nearly impossible in other
countries, due to the low incidence of retinoblastoma.1,17
We understand that a combination of desire to save the eye
and vision, parental refusal of recommended enucleation, and the idea
that treatment might make removal of the eye safer was rationale
for pre-enucleation chemotherapy. However, we show that preenucleation chemotherapy masked unfavorable pathology that would
warrant further therapy and/or surveillance, and gave the illusion of
tumor response while extraocular disease progressed to death.
The AJCC pTNM staging is designed to assess the risk of
tumor dissemination and guide further treatment for the child. In
this study, the postenucleation chemotherapy regimen was not
consistent with the pTNM risk (Fig 1). For bilateral patients, the
chemotherapy might have been given to treat the other eye, but
unilaterally affected children with no histopathologic risk factors
also received postenucleation chemotherapy.
We speculate that if the four children who died of metastasis from
their study eye had undergone immediate enucleation, the cancer
might have not yet extended outside the eye, or if high-risk pathologic
risk factors had been recognized, postenucleation surveillance and
adjuvant therapy might have avoided metastatic deaths. The preenucleation chemotherapy given in this study poorly treated tumors
already outside the eye, and failed prevent extension outside the eye.
High-risk optic nerve invasion was not treated with orbital radiotherapy although this might conventionally have been considered in addition to the postenucleation chemotherapy.
Delay of treatment is commonly known to contribute to poor
outcomes.18 In addition, the pre-enucleation chemotherapy masked
the true extent of extraocular extension, and may have rendered postenucleation chemotherapy ineffective by promoting multidrug resistance.19 One patient died despite low-risk pathology (patient 68,
pT2a) in the enucleated eye, suggesting that high-risk clinical and
pathology features that led to death might have been obscured by the
pre-enucleation chemotherapy. We also observed pathologic downstaging of group D eyes treated with pre-enucleation chemotherapy.
2011 by American Society of Clinical Oncology

Downloaded from jco.ascopubs.org on March 27, 2013. For personal use only. No other uses without permission.
Copyright 2011 American Society of Clinical Oncology. All rights reserved.

849

Zhao et al

Shorter follow-up times for children who received preenucleation chemotherapy suggests an increasing clinical practice
trend to administer pre-enucleation chemotherapy. We show that this
trend carries a severe risk to the life for children with retinoblastoma.
In comparison, primary enucleation of even group E eyes is curative.
The fewer group D than E eyes enucleated in the study time
period reflects that some group D eyes can be saved, consistent with
their less severe involvement. Success to salvage group D eyes
ranges from 30% to 50% at retinoblastoma treatment centers worldwide.19
In this treatment center, approximately 99% of group E eyes
were not saved by chemotherapy. If a group E eye was not enucleated, it was because parents refused treatment and did not return
for follow-up, and these children likely died. Several published
papers that claim cure of group E eyes, actually report eyes that are
incorrectly downstaged to be group D eyes.20,21 Our data reconfirms the urgency to primarily enucleate group E eyes.11,22 Perhaps
only when clinical trials show cure of 70% of group D eyes and
early-detected extraocular retinoblastoma, should we attempt to
save group E eyes.
Most children whose enucleation was delayed because of chemotherapy did not develop metastasis. Formal studies are necessary to
study the overall impact of chemotherapy and repeated examinations
under anesthesia on families and children. Particularly, any immediate and future systemic adverse effects and risks and financial costs to
family and health resources must be weighed against the remote possibility of successful treatment of severely involved eyes. One child
who had not received pre-enucleation chemotherapy died from postenucleation chemotherapy toxicity.
Unlike in other pediatric cancers, no significant clinical trials as
yet support the care of children with retinoblastoma.23 Starting in the
1950s, radiation therapy for intraocular retinoblastoma swept the
world after dramatic responses in intraocular tumors.24 No clinical
trial evaluated the long-term outcomes of radiation. Only 20 years
REFERENCES
1. Gallie BL, Zhao J, Vandezande K, et al: Global
issues and opportunities for optimized retinoblastoma care. Pediatr Blood Cancer 49:1083-1090,
2007
2. Dunkel IJ, Aledo A, Kernan NA, et al: Successful treatment of metastatic retinoblastoma. Cancer
89:2117-2121, 2000
3. Gunduz K, Gunalp I, Yalcindag N, et al: Causes
of chemoreduction failure in retinoblastoma and
analysis of associated factors leading to eventual
treatment with external beam radiotherapy and enucleation. Ophthalmology 111:1917-1924, 2004
4. Kremens B, Wieland R, Reinhard H, et al:
High-dose chemotherapy with autologous stem cell
rescue in children with retinoblastoma. Bone Marrow Transplant 31:281-284, 2003
5. Namouni F, Doz F, Tanguy ML, et al: Highdose chemotherapy with carboplatin, etoposide and
cyclophosphamide followed by a haematopoietic
stem cell rescue in patients with high-risk retinoblastoma: A SFOP and SFGM study. Eur J Cancer [A]
33A:2368-2375, 1997
6. Rodriguez-Galindo C, Wilson MW, Haik BG, et
al: Treatment of metastatic retinoblastoma. Ophthalmology 110:1237-1240, 2003
850

2011 by American Society of Clinical Oncology

later was it noticed that more people had died of radiation-induced

secondary cancers than of retinoblastoma.25 Chemotherapy replaced
primary radiation in the 1990s,19 again with dramatic responses in
intraocular retinoblastoma, but again, no multicenter clinical trials
have systematically evaluated negative outcomes, such as the increased
mortality we now document retrospectively. Another dramatic therapy, intra-arterial chemotherapy, swept the world for all stages of
intraocular retinoblastoma, including group E eyes.26 The data we
report here raise the alarming potential that children with group E
tumors treated with intra-arterial chemotherapy face an increased risk
of death from metastasis, when simple enucleation of the eye would
have saved their life with minimal morbidity for unilaterally affected children.
Also alarming is the possibility that the remainder of the children
on this study is not out of danger, since with a longer follow-up more
children may ultimately die of metastases.
AUTHORS DISCLOSURES OF POTENTIAL CONFLICTS
OF INTEREST
The author(s) indicated no potential conflicts of interest.

AUTHOR CONTRIBUTIONS
Conception and design: Junyang Zhao, Helen Dimaras, Helen SL Chan,
Brenda L. Gallie
Administrative support: Bin Li
Provision of study materials or patients: Junyang Zhao
Collection and assembly of data: Junyang Zhao, Helen Dimaras, Xiaolin
Xu, Dongsheng Huang, Bin Li, Brenda L. Gallie
Data analysis and interpretation: Junyang Zhao, Helen Dimaras,
Christine Massey, Helen SL Chan, Brenda L. Gallie
Manuscript writing: All authors
Final approval of manuscript: All authors

7. Sandri A, Besenzon L, Acquaviva A, et al:

Eight drugs in one day chemotherapy in a nonfamilial bilateral retinoblastoma with recurrent cerebrospinal fluid metastases. Pediatr Hematol Oncol
15:557-561, 1998
8. Dimaras H, Heon E, Budning A, et al: Retinoblastoma CSF metastasis cured by multimodality
chemotherapy without radiation. Ophthalmic Genet
30:121-126, 2009
9. Finger PT, Harbour JW, Murphree AL, et al:
Retinoblastoma, in Edge SB, Byrd DR, Carducci MA,
et al (eds): AJCC Cancer Staging Manual. New York,
NY, Springer, 2010, pp 561-568
10. Uusitalo MS, Van Quill KR, Scott IU, et al:
Evaluation of chemoprophylaxis in patients with
unilateral retinoblastoma with high-risk features on
histopathologic examination. Arch Ophthalmol 119:
41-48, 2001
11. Linn Murphree A: Intraocular retinoblastoma:
The case for a new group classification. Ophthalmology clinics of North America 18:41-53, 2005
12. Wilson TW, Chan HS, Moselhy GM, et al:
Penetration of chemotherapy into vitreous is increased by cryotherapy and cyclosporine in rabbits.
Arch Ophthalmol 114:1390-1395, 1996
13. Gray RJ: A class of K-sample tests for comparing the cumulative incidence of a competing risk.
The Annals of Statistics 16:1141-1154, 1988

14. Chan HS, DeBoer G, Thiessen JJ, et al: Combining cyclosporin with chemotherapy controls intraocular retinoblastoma without requiring radiation.
Clin Cancer Res 2:1499-1508, 1996
15. Gallie BL, Budning A, DeBoer G, et al: Chemotherapy with focal therapy can cure intraocular
retinoblastoma without radiotherapy. Arch Ophthalmol 114:1321-1328, 1996
16. Chantada GL, Fandino AC, Guitter MR, et al:
Results of a prospective study for the treatment of
unilateral retinoblastoma. Pediatr Blood Cancer 55:
60-66, 2010
17. Kivela T: The epidemiological challenge of the
most frequent eye cancer: Retinoblastoma, an issue
of birth and death. Br J Ophthalmol 93:1129-1131,
2009
18. Kimani K, Ouma B, Gallie B, et al: Ratis challenge:
A vision for Africa. Report from the First Kenyan National
Retinoblastoma Strategy Meeting. Daisys Eye Cancer
Fund, 2008. http://www.daisyseyecancerfund.org/Files/
Reports/ratischallenge3.pdf
19. Chan HS, Gallie BL, Munier FL, et al: Chemotherapy for retinoblastoma. Ophthalmology Clinics
of North America 18:55-63, 2005
20. Novetsky DE, Abramson DH, Kim JW, et al:
Published International Classification of Retinoblastoma (ICRB) definitions contain inconsistencies: An
analysis of impact. Ophthalmic Genet 30:40-44,
2009
JOURNAL OF CLINICAL ONCOLOGY

Downloaded from jco.ascopubs.org on March 27, 2013. For personal use only. No other uses without permission.
Copyright 2011 American Society of Clinical Oncology. All rights reserved.

Increased Mortality From Retinoblastoma

21. Mallipatna A, Dimaras H, Heon E, et al: Published International Classification of Retinoblastoma

(Icrb) Definitions Contain Inconsistencies: An analysis of impact. Evidence-Based Ophthalmology 10:
183-185, 2009
22. National Retinoblastoma Strategy Canadian
Guidelines for Care [in French]. Can J Ophthalmol
44:S1-S88, 2009

23. McDaid C, Hartley S, Bagnall AM, et al: Systematic review of effectiveness of different treatments for childhood retinoblastoma. Health Technol
Assess 9:iii, ix-x, 1-145, 2005
24. Reese AB, Ellsworth RM: The evaluation
and current concept of retinoblastoma therapy.
Trans Am Acad Ophthalmol-Otolaryngol 164-172,
1963

25. Eng C, Li FP, Abramson DH, et al: Mortality

from second tumors among long-term survivors of
retinoblastoma. J Natl Cancer Inst 85:1121-1128,
1993
26. Abramson DH, Dunkel IJ, Brodie SE, et al:
Superselective ophthalmic artery chemotherapy as
primary treatment for retinoblastoma (chemosurgery). Ophthalmology 2010

Shape Your Future and Ours: Apply for ASCOs Leadership Development Program
If you have completed your nal subspecialty training between 2001 and 2006 and are interested in becoming a future
leader in both ASCO and the prac
ce of oncology, ASCOs Leadership Development Program is for you. If selected, you
will:

Network with and receive mentorship from ASCO leadership

Gain exposure to ASCO commiees and government research agencies
Enhance leadership skills through interac
ve sessions
Receive rst-hand advocacy experience on Capitol Hill

To apply or learn about the eligibility requirements for the Leadership Development Program, please visit
www.asco.org/leadership

www.jco.org

2011 by American Society of Clinical Oncology

Downloaded from jco.ascopubs.org on March 27, 2013. For personal use only. No other uses without permission.
Copyright 2011 American Society of Clinical Oncology. All rights reserved.

851