Content uploaded by Atsushi Makimoto
Author content
All content in this area was uploaded by Atsushi Makimoto on Jul 14, 2022
Content may be subject to copyright.
Secondary Neoplasms After Retinoblastoma Treatment:
Retrospective Cohort Study of 754 Patients in Japan
Yuko Araki1,2,*, Yutaka Matsuyama3, Yasuki Kobayashi1, Satoshi Toyokawa1, Kazuo Inoue4,
Shigenobu Suzuki5and Atsushi Makimoto2
1
Department of Public Health, Graduate School of Medicine, the University of Tokyo,
2
Department of Pediatrics, the
National Cancer Center Hospital,
3
Department of Biostatistics, Graduate School of Medicine, the University of
Tokyo,
4
Department of Community Medicine, Chiba Medical Center, Teikyo University School of Medicine and
5
Department of Ophthalmology, the National Cancer Center Hospital, Tokyo, Japan
*For reprints and all correspondence: Yuko Araki, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan.
E-mail: arakiyu-tky@umin.ac.jp
Received June 3, 2010; accepted September 30, 2010
Objective: Little is known about the incidence of secondary neoplasms among survivors of
retinoblastoma in Japan. The objective of our study was to analyze the cumulative incidence
rate of secondary neoplasms following retinoblastoma and to investigate the risk factors of
developing secondary neoplasms.
Methods: We conducted a retrospective cohort study of 754 retinoblastoma patients who
visited the National Cancer Center Hospital in Tokyo between 1964 and 2007. The cumulative
incidence rate curves were drawn using the competing risk method and compared with the
Gray’s test. Using competing risk regression analysis, multivariate analysis estimated the sub-
distribution hazard ratio of factors related to the development of secondary neoplasms.
Results: The median length of follow-up was 108 months (0–594 months). Twenty-one
(2.8%) patients developed 23 secondary neoplasms in total. The cumulative incidence rates
of secondary neoplasms after retinoblastoma treatment were 2.4% at 10 years after diagno-
sis, 4.3% at 20 years, 6.4% at 30 years and 19.1% at 40 years. Ten patients (1.3%) died and
723 (95.9%) were alive without developing secondary neoplasms. The subdistribution hazard
ratios of hereditary retinoblastoma and external beam irradiation were 4.85 (95% confidence
interval ¼0.74–31.85) and 4.76 (95% confidence interval ¼0.69 – 33.09), respectively.
Conclusions: We demonstrated the cumulative incidence rate of secondary neoplasms fol-
lowing retinoblastoma in Japan. The subdistribution hazards ratios of hereditary retinoblas-
toma and external beam irradiation were high but not significant because of statistical power.
The long-term follow-up of retinoblastoma survivors is warranted to understand secondary
neoplasm risk.
Key words: retinoblastoma – second neoplasm – incidence study – heredity – radiotherapy
INTRODUCTION
Retinoblastoma (RBL) is the most common intraocular
malignancy in children, accounting for approximately 3% of
all malignancies in children younger than 15 years of age in
the USA (1). The national registry of RBL in Japan has
reported about 70 – 90 new cases of RBL per year (2). About
25% of RBL is hereditary and bilateral, 15% is hereditary
and unilateral, and the remaining (60%) is non-hereditary
and unilateral (3,4). Patients with the hereditary form are at
risk of developing secondary neoplasms (SNs) (5,6),
especially osteosarcomas and soft tissue sarcomas (7).
Over the last 30 years, the 5-year survival of the RBL
cases has been improving. For example, the 5-year survival
rate was 96.5% in the USA (1995 – 2004) (8). Therefore, the
recent challenge in treating RBL is to improve the quality of
life of the patients by preserving the affected eyes and pre-
venting serious side effects from the treatment of the malig-
nancy. SN is one of the most critical issues to confront
#The Author (2010). Published by Oxford University Press. All rights reserved.
Jpn J Clin Oncol 2011;41(3)373 – 379
doi:10.1093/jjco/hyq201
Advance Access Publication 4 November 2010
Downloaded from https://academic.oup.com/jjco/article/41/3/373/809123 by guest on 14 July 2022
because it is the major cause of early deaths among RBL
survivors (7).
Systemic chemotherapy (chemoreduction) with focal
therapies (9) has become an alternative to external beam
irradiation (EBI), a treatment that enhances the risk of devel-
oping SNs in hereditary RBL cases (5,6,10). However, little
is known about the long-term effect of chemotherapeutic
agents on SN development after RBL treatment.
The epidemiology of SNs after RBL treatment has not been
thoroughly explored in Japan. Although there are research
reports on affected Japanese patients with SNs after RBL treat-
ment, most of them are case reports. No cumulative incidence
rate (CIR) of SNs has been reported since 1992 (2). The
purpose of our study is to present the CIR of SNs with longer
follow-up time compared with the previous report (2)andto
examine relative risks of various factors on SN development.
PATIENTS AND METHODS
We conducted a retrospective cohort study of RBL patients at
the National Cancer Center Hospital (NCCH) in Tokyo. We
reviewed the medical records of 754 RBL patients who
initially visited NCCH between September 1964 and April
2007. The patients were followed until 31 May 2007. The
endpoint of the study was defined as the development of an
SN. SN was defined as a primary neoplasm, except RBL,
which developed after initial RBL treatment. Therefore, RBL
relapse and trilateral RBL (primary neoplasm of the pineal
and parasellar sites) were excluded from the category of SN.
The following factors (patient characteristics) were
extracted from the records: (i) basic characteristics at the
time of diagnosis (age, sex, laterality of RBL, family history
of RBL, the Reese– Ellsworth classification (R– E classifi-
cation) for the affected eye, metastasis) and (ii) treatments
(focal treatments, focal chemotherapy, systemic chemother-
apy, EBI, brachytherapy). The R – E classification has been
widely used as a guideline for predicting prognosis for eye
preservation (11). Focal treatments consist of photocoagula-
tion, cryotherapy and hyperthermia. Focal chemotherapy
refers to the infusion of chemotherapeutic agents either via
the ophthalmic artery or directly into the vitreous.
CIR of SNs was calculated using the competing risk
method and then it was compared with the Gray’s test (12).
The subdistribution hazard ratio of the factors related to SN
development was calculated with multivariate analysis using
the competing risk regression model (13). In these analyses,
the event of interest was defined as SN development and a
competing risk event as death prior to the SN development.
The R statistical software (version 2.10.0) was employed to
conduct the estimation. A Pvalue of ,0.05 (two-sided) was
considered statistically significant when choosing the vari-
ables in the final multivariate model. A 95% confidence
interval (CI) was calculated for the subdistribution hazard
ratio. SPSS 16.0J for Windows (SPSS Japan Inc., Tokyo,
Japan) was used for descriptive analysis.
RESULTS
Among the 754 RBL patients whose medical records were
reviewed, 21 (2.8%) individuals developed SNs. Ten
patients (1.3%) died prior to developing SNs, and 723
patients (95.9%) survived without developing SNs. Table 1
describes the characteristics of the RBL patients. Table 2
shows the regimens of systemic chemotherapy. According
to the previous studies (2,12), we defined the hereditary
RBL as ‘bilateral RBL and/or having family history of
RBL’ (Table 3). Median lengths of follow-up were deter-
mined as 102 months (the first and third quartiles, 62 –186;
range 14– 438) for the patients who developed SNs; 46
(13– 66, 5 – 133) for the patients who died prior to develop-
ing SNs and 110 (52 – 200, 0 – 594) for the patients who
survived without developing SN.
The 21 affected patients developed 23 SNs in total.
Table 4displays pathological diagnoses of the SNs. Note
that one patient developed two Meibomian carcinomas (one
for each eyelid), and one patient developed double neo-
plasms (rhabdomyosarcoma and osteosarcoma). Of all ident-
ified SNs, 15 out of 23 (65.2%) were osseous or soft tissue
sarcomas (7 were osteosarcomas, 7 were rhabdomyosarco-
mas and 1 was a myxofibrosarcoma). Others were diagnosed
as two Meibomian carcinomas, one acoustic neuroma, one
meningioma, one acute myelogenous leukemia and one
neuroendocrine tumor. Pathologic diagnoses were not
obtained for the remaining two neoplasms because no sur-
geries were performed on these tumors.
Figure 1shows the CIR of SNs in the presence of a com-
peting risk for the whole series of RBL. The CIRs of SNs
every 10 years after RBL treatment were 2.4% at 10 years
after diagnosis, 4.3% at 20 years, 6.4% at 30 years and
19.1% at 40 years.
The following six factors were independently associated
with SN development in the univariate analysis using the
Gray’s test (12): age at diagnosis (0 – 11 months: P¼0.035);
heredity (hereditary RBL: P,0.001); focal therapy (yes:
P¼0.003); focal chemotherapy (yes: P¼0.005); systemic
chemotherapy (yes: P¼0.028) and EBI (yes: P¼0.004).
Table 5demonstrates the CIRs of SNs by heredity.
The relative risks (subdistribution hazards ratio) of SN
development were estimated with the six factors above
by fitting a proportional subdistribution hazards regression
model (13). Table 6summarizes the results of this analy-
sis. Hereditary RBL and EBI showed high although
not statistically significant relative risks of 4.85 (95%
CI ¼0.74– 31.85) for hereditary RBL and 4.76 (95%CI ¼
0.69– 33.09) for EBI.
DISCUSSION
In our study, the competing risk regression analysis demon-
strated that the risks of developing SNs on the two factors,
hereditary RBL and EBI, increases more than four times,
374 Secondary neoplasms following retinoblastoma
Downloaded from https://academic.oup.com/jjco/article/41/3/373/809123 by guest on 14 July 2022
although the Pvalues were not statistically significant. Focal
chemotherapy and systemic chemotherapy had moderate but
not significant associations with SN development (SHR 2.24
and 1.92, respectively).
Previous studies documented the strong association
between hereditary RBL and SN development (5,6).
Abramson reviewed 711 hereditary RBL patients and
observed that the CIRs of SNs of their cohort were 20% at
10 years from diagnosis, 50% at 20 years and 90% at 30
years for the irradiated patients, and 10% at 10 years, 30% at
20 years and 68% at 32 years for the non-irradiated patients
(5). These values are much higher compared with those
found for our cohort. Roatry examined 215 patients with
bilateral RBL where the CIRs of SNs of the cohort were
4.44 % at 10 years, 18.3% at 20 years and 26.1 % at 30
years (6).
Loss of heterozygosity (LOH) on 13q14 of the chromo-
some has been detected among patients with hereditary RBL
(14). The locus contains the RB1 RBL gene, a tumor sup-
pressor gene originally isolated by Friend et al.(15). LOH is
caused by either deletion of the chromosomal locus (16)or
mutation of the RB1 gene (17), and is detected not only in
RBL but also in other malignancies such as osteosarcoma
(18), breast cancer (19) and small cell lung cancer (20).
Hereditary RBL patients are at an increased risk of develop-
ing osteosarcomas (7,21).
EBI is another risk factor for developing SNs, especially
among hereditary RBL patients (5,6,10,22–25). The pre-
vious study in Japan did not show a significant difference
in the CIRs of SNs by irradiation history (2), perhaps
because of the small sampling size and the short follow-up
period.
Table 1. Characteristics of the RBL patients (n¼754)
Outcome (no. of patients) No death, no SN at
follow-up (n¼723)
Death at follow-up
(n¼10)
SN at follow-up
(n¼21)
Age at diagnosis 0–11 months 344 (94.2%) 6 (1.6%) 15 (4.1%)
12 months or older 379 (97.4%) 4 (1.0%) 6 (1.5%)
Sex Male 365 (95.1%) 8 (2.1%) 11 (2.9%)
Female 358 (96.8%) 2 (0.5%) 10 (2.7%)
Laterality of RBL Bilateral 336 (93.6%) 4 (1.1%) 19 (5.3%)
Unilateral 387 (98.0%) 6 (1.5%) 2 (0.5%)
Family history of RBL Yes 55 (88.7%) 0 (0%) 7 (11.3%)
No 668 (96.5%) 10 (1.4%) 14 (2.0%)
R–E classification (Rt) Stages I–III 195 (94.7%) 2 (1.0%) 9 (4.4%)
Stages IV and V 238 (95.6%) 5 (2.0%) 6 (2.4%)
NA 290 (97.0%) 3 (1.0%) 6 (2.0%)
R–E classification (Lt) Stages I – III 214 (94.7%) 2 (0.9%) 10 (4.4%)
Stages IV and V 256 (95.9%) 4 (1.5%) 7 (2.6%)
NA 253 (97.0%) 4 (1.5%) 4 (1.5%)
Metastasis Yes 2 (66.7%) 1 (33.3%) 0 (0%)
No 721 (96.0%) 9 (1.2%) 21 (2.8%)
Focal treatments
a
Yes 426 (95.3%) 5 (1.1%) 16 (3.6%)
No 297 (96.7%) 5 (1.6%) 5 (1.6%)
Focal chemotherapy
b
Yes 316 (94.6%) 6 (1.8%) 12 (3.6%)
No 407 (96.9%) 4 (1.0%) 9 (2.1%)
Systemic chemotherapy Yes 186 (93.9%) 5 (2.5%) 7 (3.5%)
No 537 (96.6%) 5 (0.9%) 14 (2.5%)
External beam irradiation Yes 394 (92.9%) 10 (2.4%) 20 (4.7%)
No 329 (99.7%) 0 (0%) 1 (0.3%)
Brachytherapy Yes 148 (95.5%) 2 (1.3%) 5 (3.2%)
No 575 (96.0%) 8 (1.3%) 16 (2.7%)
RBL, retinoblastoma; SN, secondary neoplasm; R– E classification, Reese–Ellsworth classification; NA, not available.
a
Photocoagulation, cryotherapy and hyperthermia.
b
Infusion of chemotherapeutic agents via ophthalmic artery, or direct injection into the vitreous body.
Jpn J Clin Oncol 2011;41(3) 375
Downloaded from https://academic.oup.com/jjco/article/41/3/373/809123 by guest on 14 July 2022
The results of our study are consistent with the above
studies regarding the risk of hereditary RBL and EBI on SN
development. These two factors were statistically significant
in the univariate analysis, though not significant in the multi-
variate analysis probably due to the small number of events.
Marees et al.(25) reported 74 SNs among 668 RBL survi-
vors. Their Cox regression analysis showed an elevated risk
of SN development among hereditary patients after EBI
when compared with that among hereditary patients who
received surgery alone (hazard ratio ¼2.81, 95% CI ¼
1.28– 6.19). Other studies (5,6,10,22 –24) reported higher
CIRs of SNs among hereditary or irradiated patients using
the Kaplan– Meier method, though they did not report P
values or carry out a log-rank test. Because we took a com-
peting risk method (death prior to developing SN), the result
might be more conservative than the one estimated by Cox’s
regression analysis.
Our study showed that systemic chemotherapy had moder-
ate association with SN development. So far, we could not
conclude whether systemic chemotherapy is a risk factor for
SN or not for two reasons. First, there was a discrepancy in
the follow-up time between the different treatments in the
Table 2. Regimens of systemic chemotherapy
Chemotherapy regimens No. of
cases
a
Neo-adjuvant
therapy
VEC Day 0, CBDCA 18.6 mg/kg 78
Days 0 and 1, VP-16 5 mg/kg
Day 0, VCR 0.05 mg/kg
Repeat every 28 days 6
Adjuvant therapy VCA Day 1, VCR 1.5 mg/m 23
Days 1 and 2, CPA 150 mg/m
2
Day 3, ADR 30 mg/m
2
VCA, VP/CDDP Weeks 1, 3, 5 Weeks 2, 4, 6 24
Day 1, VCR 1.5 mg/m
2
Days 1–5, VP-16 100 mg/m
2
5days
Days 2 and 3, CPA 800 mg/m
2
Days 1–5, CDDP 20 mg/m
2
5days
Day 4, ADR 40 mg/m
2
5 days Repeat every 28 days 6
VCA, VP/CBDCA Weeks 1, 3, 5 Weeks 2, 4, 6 19
Day 1, VCR 1.5 mg/m
2
Days 1–5, VP-16 100 mg/m
2
5days
Days 2 and 3, CPA 800 mg/m
2
Days 1–5, CBDCA 100 mg/m
2
5days
Day 4, ADR 40 mg/m
2
Repeat every 28 days 6
98-New A1 Day 1, VCR 1.5 mg/m
2
12
Day 1, CPA 1200 mg/m
2
Days 1–5, CDDP 18 mg/m
2
5days
Day 3, ADR 40 mg/m
2
Repeat every 28 days 6
James VCR/CPA alternatively, every week. 11
Day 1, VCR 1.5 mg/m
2
or
CPA 300 mg/m
2
PBSCT Megatherapy prior to
PBSCT
Days 1 and 2, CPA 60 mg/kg 2days Or 9
Days 3 and 4 L-PAM80 mg/m
2
2 days Days 1–4, VP-16 200 mg/m
2
4days
Days 5 and 6, TESPA 250 mg/m
2
2 days (day
8, PBSCT)
Days 1–4, CBDCA 400 mg/m
2
4days
Days 5–6 L-PAM 90 mg/m
2
2 days (day
8, PBSCT)
Other regimens 54
NA 19
CBDCA, carboplatin; VP-16, etoposide; VCR, vincristine; CPA, cyclophosphamide; ADR, adriamycin; CDDP, cisplatin; L-PAM, melphalan; TESPA,
triethylene thio-phosphoramide; VEC, VCR/VP-16/CBDCA; VCA, VCR/CPA/ADR; PBSCT, peripheral blood stem cell transplantation; NA, not available.
a
The total number of cases does not match that in Table 1, because some patients underwent multiple courses/regiments of chemotherapy.
376 Secondary neoplasms following retinoblastoma
Downloaded from https://academic.oup.com/jjco/article/41/3/373/809123 by guest on 14 July 2022
cohort; i.e. the length of follow-up was shorter for the
patients who received systemic chemotherapy (median 48
months, range 0–401) compared with those who received
EBI (median 131 months, range 5–500), and this might
have underestimated the effect of systemic chemotherapy.
Second, we did not stratify the cohort according to the che-
motherapeutic regimens because they were rich in variety
and the stratification of the regimen would much further
decrease the statistical power.
There were moderate associations between regional treat-
ments (focal therapy and focal chemotherapy) and SN devel-
opment. However, the authors did not interpret these
treatments as risk factors for SN development. The reason is
that the effects of the treatments are confined to the regional
RBL; thus, theoretically speaking, they would not be
expected to induce extra-ocular neoplasms. The association
might have been due to a confounding effect because those
treatments were selectively applied to hereditary RBL cases.
Regarding the risk of chemotherapy, several studies have
indicated that treatment with alkylating agents and
topoisomerase II inhibitors (epipodophyllotoxins and anthra-
cyclines) increased the probability of secondary acute
myeloid leukemia (sAML) (26–29). Gombos et al. reported
in their retrospective study that there was a high incidence of
sAML in RBL patients after chemotherapy with topoisome-
rase II inhibitors, where 12 out of 15 sAML cases of their
cohort had received topoisomerase II inhibitors (30).
Chemoreduction without topoisomerase II inhibitors had
Table 4. Pathological diagnoses of SNs following RBL
Pathological diagnoses No. of neoplasms
Osteosarcoma 7
a
Rhabdomyosarcoma 7
a
Meibomian carcinoma 2
b
Myxofibrosarcoma 1
Acoustic neuroma 1
Meningioma 1
Acute myelogenous leukemia 1
Neuroendocrine tumor 1
Unknown 2
a
One patient developed double neoplasms; a rhabdomyosarcoma and an
osteosarcoma.
b
One patient developed two Meibomian carcinomas (one for each eyelid).
Table 5. CIR of SNs according to heredity of RBL
Time from RBL diagnosis (years) Non-hereditary Hereditary
CIR (%) CIR (%)
10 0.3 4.3
20 1 7.3
30 1 11.3
40 NA 33.8
CIR, cumulative incidence rate; NA, not applicable.
Table 3. Definition of hereditary RBL in the study
Family history Total
Yes No
Laterality
Unilateral 13 (3.3%) 382 (96.7%) 395 (100%)
Bilateral 49 (13.6%) 310 (86.4%) 359 (100%)
Total 62 (8.2%) 692 (91.8%) 754 (100%)
Bold, hereditary RBL (372 cases, 49.3%); Italics, non-hereditary RBL (382
cases, 50.7%).
Figure 1. Cumulative incidence rate of secondary neoplasms after treating
retinoblastoma: the whole case. SN, secondary neoplasm.
Table 6. Subdistribution hazard ratio of the factors for developing SNs
following RBL (analyzed with competing risks model) (n¼754)
Factors Subdistribution
hazard ratio
95% CI Pvalue
Age at diagnosis (0–11
months/12 months or older)
1.33 0.43 — 4.17 0.620
Heredity of RBL (hereditary/
non-hereditary)
4.85 0.74 — 31.85 0.100
Focal treatments (yes/no) 1.54 0.53 — 4.53 0.430
Focal chemotherapy (yes/no) 2.24 0.82 — 6.13 0.120
Systemic chemotherapy (yes/
no)
1.92 0.71 — 5.23 0.200
External beam irradiation
(yes/no)
4.76 0.69 — 33.09 0.110
95% CI, 95% confidence interval.
Jpn J Clin Oncol 2011;41(3) 377
Downloaded from https://academic.oup.com/jjco/article/41/3/373/809123 by guest on 14 July 2022
been attempted in a recent clinical trial on early-stage RBL
patients to avoid the possible risk of developing SNs (31).
There are two limitations of this study. First, we could not
include R – E classification in the analysis because the classi-
fication is the feature per eye compared with other indepen-
dent variables, which are the features per individual. The
mixture of variously staged patients might have pulled the
result toward null or could have inflated the true findings.
Second, the shorter follow-up period of the patients after sys-
temic chemotherapy could have underestimated the potential
treatment effect. Further follow-up of these patients should
improve the accuracy of the study.
Our study has three advantages. First, this is the first report
of the CIR of SNs among RBL patients in Japan since 1992.
Second, no study has ever used an analytical method consid-
ering ‘competing risks’ on the long-term follow-up of pedi-
atric cancer patients in Japan. Competing risks are events for
which the occurrence of some other event does not allow the
event to occur (32). They may preclude the onset of the event
of interest or may modify the probability of the onset of the
event of interest (33). (In our study, a death prior to SN devel-
opment is defined as a competing risk, which prevents the
event of interest, namely SN development.) Analytical
methods that take competing risks into consideration (12)
have an advantage over the Kaplan – Meier estimation pro-
cedure because the latter assumes that any subject who does
not experience the event of interest as censored could thus
miss information from the competing risks. Third, although
our cohort is limited to a single institution, the size of the
cohort is relatively large compared with the entire population
of Japanese RBL patients; an estimated 80 children per year
develop RBL in Japan (2). In addition, none of our patients
were missed from the follow-up.
Our study suggests that the RBL patients, especially her-
editary ones and those who received EBI, should be closely
monitored for the risk of developing SNs. Since the CIR of
SNs increases along with the time course of the disease,
attention to adult survivors is particularly important.
Although the risk of systemic chemotherapy on SN devel-
opment is not completely denied, we consider it rational to
apply such a treatment to carefully selected patients whose
benefits of preserving a functioning eye outweigh the risk of
developing SN.
CONCLUSION
Our study was the first since 1992 to reveal the CIR of SNs
after RBL treatment in Japan. The study also suggested the
risk of hereditary RBL and EBI on SN development. The
long-term follow-up of RBL patients should be pursued in
order to provide further information on SNs.
Conflict of interest statement
None declared.
References
1. Hurwitz RL, Shields CL, Shields JA, Chevez-Barrios P, Hurwitz MY,
Chintagumpala MM. Retinoblastoma. In: Pizzo PA, Poplack DG,
editors. Principles and Practice of Pediatric Oncology. Philadelphia:
Lippincott Williams & Wilkins 2006;865.
2. Survival rate and risk factors for patients with retinoblastoma in Japan.
The Committee for the National Registry of Retinoblastoma. Jpn J
Ophthalmol 1992;36:121– 31.
3. Knudson AG, Jr. Mutation and cancer: statistical study of
retinoblastoma. Proc Natl Acad Sci USA 1971;68:820–3.
4. Bonaiti-Pellie C, Briard-Guillemot ML. Segregation analysis in
hereditary retinoblastoma. Hum Genet 1981;57:411–9.
5.AbramsonD,EllsworthR,KitchinF,TungG.Secondnonocular
tumors in retinoblastoma survivors. Are they radiation-induced?
Ophthalmology 1984;91:1351– 5.
6. Roarty JD, McLean IW, Zimmerman LE. Incidence of second
neoplasms in patients with bilateral retinoblastoma. Ophthalmology
1988;95:1583– 7.
7. Eng C, Li FP, Abramson DH, Ellsworth RM, Wong FL, Goldman MB,
et al. Mortality from second tumors among long-term survivors of
retinoblastoma. J Natl Cancer Inst 1993;85:1121 – 8.
8. Broaddus E, Topham A, Singh AD. Survival with retinoblastoma in the
USA: 1975– 2004. Br J Ophthalmol 2009;93:24 – 7.
9. Friedman DL, Himelstein B, Shields CL, Shields JA, Needle M,
Miller D, et al. Chemoreduction and local ophthalmic therapy for
intraocular retinoblastoma. J Clin Oncol 2000;18:12– 7.
10. Draper GJ, Sanders BM, Kingston JE. Second primary neoplasms in
patients with retinoblastoma. Br J Cancer 1986;53:661– 71.
11. Reese AB, Ellsworth RM. Management of retinoblastoma. Ann N Y
Acad Sci 1964;114:958– 62.
12. Gray RJ. A class of K-sample tests for comparing the cumulative
incidence of a competing risk. Ann Stat 1988;16:1141 – 54.
13. Fine JP, Gray RJ. A proportional hazards model for the subdistribution
of a competing risk. J Am Stat Assoc 1999;94:496 – 509.
14. Alvarez CS. Molecular biology of retinoblastoma. Clin Transl Oncol
2008;10:389– 94.
15. Friend S, Bernards R, Rogelj S, Weinberg R, Rapaport J, Albert D,
et al. A human DNA segment with properties of the gene that
predisposes to retinoblastoma and osteosarcoma. Nature 1986;
323:643– 6.
16. Francke U. Retinoblastoma and chromosome-13. Cytogenet Cell Genet
1976;16:131– 4.
17. Ward P, Packman S, Loughman W, Sparkes M, Sparkes R,
McMahon A, et al. Location of the retinoblastoma susceptibility
gene(s) and the human esterase D locus. J Med Genet 1984;21:92–5.
18. Hansen M, Koufos A, Gallie B, Phillips R, Fodstad O, Brøgger A, et al.
Osteosarcoma and retinoblastoma: a shared chromosomal mechanism
revealing recessive predisposition. Proc Natl Acad Sci USA
1985;82:6216– 20.
19. Bookstein R, Lee E, Peccei A, Lee WH. Human retinoblastoma gene—
long-range mapping and analysis of its deletion in a breast-cancer
cell-line. Mol Cell Biol 1989;9:1628– 34.
20. Harbour JW, Lai SL, Whangpeng J, Gazdar AF, Minna JD, Kaye FJ.
Abnormalities in structure and expression of the human retinoblastoma
gene in SCLC. Science 1988;241:353– 7.
21. Wong FL, Boice JD, Abramson DH, Tarone RE, Kleinerman RA,
Stovall M, et al. Cancer incidence after retinoblastoma—radiation dose
and sarcoma risk. JAMA 1997;278:1262– 7.
22. Fletcher O, Easton D, Anderson K, Gilham C, Jay M, Peto J. Lifetime
risks of common cancers among retinoblastoma survivors. JNatl
Cancer Inst 2004;96:357– 63.
23. Kleinerman R, Tucker M, Tarone R, Abramson D, Seddon J, Stovall M,
et al. Risk of new cancers after radiotherapy in long-term survivors
of retinoblastoma: an extended follow-up. J Clin Oncol 2005;23:
2272–9.
24. Moll AC, Imhof SM, Schouten-Van Meeteren AY, Kuik DJ, Hofman P,
Boers M. Second primary tumors in hereditary retinoblastoma: a
register-based study, 1945–1997: is there an age effect on
radiation-related risk? Ophthalmology 2001;108:1109–14.
25. Marees T, Moll A, Imhof S, de Boer M, Ringens P, van Leeuwen F.
Risk of second malignancies in survivors of retinoblastoma: more than
40 years of follow-up. J Natl Cancer Inst 2008;100:1771– 9.
378 Secondary neoplasms following retinoblastoma
Downloaded from https://academic.oup.com/jjco/article/41/3/373/809123 by guest on 14 July 2022
26. Davies SM. Therapy-related leukemia associated with alkylating agents.
Med Pediatr Oncol 2001;36:536– 40.
27. Felix CA. Leukemias related to treatment with DNA topoisomerase II
inhibitors. Med Pediatr Oncol 2001;36:525– 35.
28. Pui CH, Relling MV. Topoisomerase II inhibitor-related acute myeloid
leukaemia. Br J Haematol 2000;109:13– 23.
29. Pui CH, Ribeiro RC, Hancock ML, Rivera GK, Evans WE,
Raimondi SC, et al. Acute myeloid leukemia in children treated with
epipodophyllotoxins for acute lymphoblastic leukemia. NEnglJMed
1991;325:1682– 7.
30. Gombos D, Hungerford J, Abramson D, Kingston J, Chantada G,
Dunkel I, et al. Secondary acute myelogenous leukemia in patients
with retinoblast oma: is chemo therapy a factor ? Ophthalmology
2007;114:1378– 83.
31. Rodriguez-Galindo C, Wilson M, Haik B, Merchant T, Billups C,
Shah N, et al. Treatment of intraocular retinoblastoma with vincristine
and carboplatin. J Clin Oncol 2003;21:2019– 25.
32. Logan B, Zhang M, Klein J. Regression models for hazard
rates versus cumulative incidence probabilities in hematopoietic
cell transplantation data. Biol Blood Marrow Transplant 2006;
12(1 Suppl 1):107–12.
33. Satagopan J, Ben-Porat L, Berwick M, Robson M, Kutler D,
Auerbach A. A note on competing risks in survival data analysis. Br J
Cancer 2004;91:1229– 35.
Jpn J Clin Oncol 2011;41(3) 379
Downloaded from https://academic.oup.com/jjco/article/41/3/373/809123 by guest on 14 July 2022
