- Email: [email protected]
Second primary tumors in hereditary retinoblastoma: a register-based study, 1945–1997
Second Primary Tumors in Hereditary Retinoblastoma: A Register-based Study, 1945–1997 Is There an Age Effect on Radiation-related Risk? Annette C. Moll, MD, PhD,1,2 Saskia M. Imhof, MD, PhD,1 Antoinette Y. N. Schouten-Van Meeteren, MD,3 Dirk J. Kuik,2 Pieter Hofman, MD, PhD,4 Maarten Boers, MD, PhD2 Objective: The aim of this study is to evaluate the influence of age at external beam irradiation (EBRT) on the occurrence of second primary tumors (SPTs) inside and outside the irradiation field in hereditary retinoblastoma patients. Design: Cross-sectional study. Participants: The study included 263 hereditary retinoblastoma patients born in The Netherlands between 1945 and 1997. Methods: A national register-based follow-up cohort study was performed on hereditary retinoblastoma patients. Information on therapy, age at irradiation, and location of SPT was obtained from the register. The Kaplan-Meier method calculated cumulative incidences of SPT in three subgroups: irradiation before (early EBRT) and after 12 months of age (late EBRT), and no irradiation. The Mantel-Cox method determined the statistical significance of differences between the cumulative incidence curves. Main Outcome Measures: Development of SPT inside and outside a precisely defined irradiation field in relation to age at irradiation. Our definition excluded pineoblastoma as SPT, because they constitute part of a trilateral retinoblastoma; in addition, they lie outside the field of irradiation. Results: The cumulative incidence of SPT at the age of 25 years was 22% (95% confidence intervals 13%–34%) in the early EBRT group, 3% (0%–14%) in the late EBRT group, and 5% (1%–16%) in the nonirradiated group (Mantel-Cox overall: P ⫽ 0.001; between early and late EBRT, P ⫽ 0.04). However, in early irradiated patients, the incidence of SPTs inside and outside the irradiation field was similar (11%), and the difference between early and late EBRT in incidence of SPT inside the field of irradiation was less prominent than overall (11% vs. 3%: P ⫽ 0.37). Sensitivity analysis showed the results depended on the way SPT, irradiation field, and, especially, pineoblastomas are defined. Conclusions: Hereditary retinoblastoma confers an increased risk for the development of SPT, especially in patients treated with EBRT before the age of 12 months. However, the presence of similar numbers of SPTs inside and outside the irradiation field suggests that irradiation is not the cause. In other words, this study does not show an age effect on radiation-related risk. Rather, early EBRT is probably a marker for other risk factors of SPT. Ophthalmology 2001;108:1109 –1114 © 2001 by the American Academy of Ophthalmology. It is well recognized that hereditary retinoblastoma confers an increased risk for the development of a second primary tumor (SPT). Initial reports described SPTs within the field of irradiation, and investigators assumed all SPTs were induced by irradiation.1–3 A subsequent study demonstrated that SPTs developed in 14% of subjects with bilateral retinoblastoma who underwent enucleation without irradiaOriginally received: April 29, 1999. Accepted: January 24, 2001. Manuscript no. 99203. 1 Department of Ophthalmology, Vrije Universiteit, Amsterdam, The Netherlands. 2 Department of Clinical Epidemiology & Biostatistics, Vrije Universiteit, Amsterdam, The Netherlands. 3 Department of Pediatrics, Vrije Universiteit, Amsterdam, The Netherlands. © 2001 by the American Academy of Ophthalmology Published by Elsevier Science Inc.
tion.4 Since 1949, numerous articles have reported on the incidence of SPTs in retinoblastoma subjects. Because these studies differ substantially in design and population size, it is no surprise that our review5 reported a wide range in the incidence of SPTs: we found cumulative incidences of SPTs of 8.4% 18 years after diagnosis,6 15.7% at the age of 20 years,7 19% at the age of 35 years,8 and a relative risk of 15.4 for SPT,9 respectively. 4
Department of Radiotherapy, University Hospital Utrecht, Utrecht, The Netherlands. None of the authors have commercial interests in this study. Reprint requests to Dr. Annette C. Moll, Department of Ophthalmology, Academic Hospital Vrije Universiteit, De Boelelaan 1117, 1081 HV Amsterdam, The Netherlands. E-mail: [email protected]. ISSN 0161-6420/01/$–see front matter PII S0161-6420(01)00562-0
1109
Ophthalmology Volume 108, Number 6, June 2001 Abramson and Frank10 recently suggested in this journal that age at irradiation might explain some of the variability of the reported risk of SPT. This suggestion has intensified the discussion on alternative modalities of therapy. “The implications of this work, if it is replicated by others, are extremely important for the treatment of children with retinoblastoma” as Char commented in the discussion of that article. As a result of the increased incidence of SPT among irradiated patients and other complications such as orbital growth retardation,11 current attention is focused on the use of chemotherapy in combination with local treatment such as brachytherapy, diode laser, and cryotherapy.12,13 We have previously reported on the incidence, survival, and cumulative incidence of SPT and pineoblastomas in retinoblastoma patients born in The Netherlands.14,15 Our register-based follow-up cohort study can also contribute to the discussion of a possible age effect on radiation-related risk of SPT. Therefore, the aim of this study is to evaluate the influence of age at irradiation on the occurrence of SPT inside and outside the irradiation field in hereditary retinoblastoma.
Material and Methods Study Population and Follow-up Procedures A follow-up study was performed on retinoblastoma patients born in The Netherlands between 1945 and 1997. A previous report contains a complete description of the methods used to locate survivors and document deaths.14 In short, the Dutch Retinoblastoma Register was completed with the assistance of all University Eye Clinics and Cancer Registration Centers. Hospital records supplied data about demography, tumor laterality, family history of retinoblastoma, treatment, occurrence of SPT, and date and cause of death. Every effort was made to determine the survival for each retinoblastoma patient. The most recent address of each patient was determined by means of the municipal registries and the Central Bureau of Genealogy, and the patient’s status (alive or dead) was recorded. Hereditary retinoblastoma patients were followed for the occurrence of SPT. Different methods were used: patients with hereditary retinoblastoma born between 1945 and 1970 (n ⫽ 146) had previously been visited at home in 1985 (n ⫽ 141) to investigate the incidence of SPT.8 In alive patients (n ⫽ 101) new information on health, SPT, and therapy was obtained by a questionnaire, if necessary in Braille. In deceased patients information on the cause of death was obtained by means of the most recent treating physician (n ⫽ 6). Patients born between 1970 and 1997 (n ⫽ 121) and their parents were approached through their own ophthalmologist. A detailed history was obtained by interview at home or in the hospital. Reported SPTs were confirmed by clinical and histopathologic reports. We considered retinoblastoma hereditary if one or more of the following criteria were met: a bilateral retinoblastoma, a positive family history for retinoblastoma (then the related parent will be the carrier of the defect in the retinoblastoma gene), or a defect in the retinoblastoma gene found in chromosomal/DNA analysis of the subject. The criteria of Warren and Gates16 defined SPT: each of the tumors must present a definite picture of malignancy; each tumor must be distinct (different histologic appearance); and the probability that the SPT is a metastatic lesion from the primary tumor must be excluded. Because a pineoblastoma is histologically iden-
1110
tical to a retinoblastoma and therefore also called a trilateral retinoblastoma, we and others17,18 do not consider pineoblastoma an SPT.
Field Definition and Radiotherapy Technique Tumors were classified as inside the field of irradiation if they originated in the lids, orbits, periocular sinuses, temporal bones, or skin overlying the temporal bone region. Other locations, such as thyroid gland, neck, and brain, were defined as outside the field of irradiation19,20 on the basis of the following considerations. The radiation data were assumed to be reliable for patients treated from 1971 onward. Beginning in 1971, all patients were irradiated by a 6- or 8-MV photon beam in a specially collimated field: 20 ⫻ 26 mm and D-shaped in the sagittal plane. The dosage was 45 Gy in 15 fractions, 3 fractions a week. The field covers the entire retina, the straight edge of the D lying close to the posterior pole of the lens, and the rounded part facing posteriorly. The width corresponds with the central axis of the eye. The field extends from the posterior pole of the lens in the direction of the chiasma over 20 mm. In our patients, the mean distance between the posterior pole of the lens and the internal side of the retina (12 MHz ultrasonic A-scan) was 15 mm (standard deviation, 0.3). In a few patients, the thickness of the posterior side of the sclera could be determined at 3 mm. The typical distance from the posterior pole of the lens to the entrance of the optic nerve is therefore 18 mm. The field size in that direction was 20 mm. As stated previously, the field is D-shaped on sagittal cross-section, with the rounded part facing posteriorly. This boundary follows the posterior scleral contour at a distance of 2 mm. It is now the penumbra of this specially collimated field that determines the dosage delivered to the surrounding structures, both the lens and the beginning of the optic nerve. The penumbra width in which the dosage decreases from 50% to 10% of the central field maximum dosage is only 3 mm at a depth of 2 cm under the skin. In unilateral retinoblastoma, the external irradiation was performed with one single temporal lateral field of 20 ⫻ 26 mm; in bilateral retinoblastoma, two opposing identical temporal fields were applied. This technique, developed by Schipper21 in the late 1960s offers a reproduction accuracy of at least 1 mm. From these statistics we can estimate the irradiation dosage to the sella region. If the distance from the posterior pole of the eye to the chiasma is assumed to be 2 cm, the irradiation dosage to the chiasma will be less than 1% of the maximum dosage of 45 Gy in 15 fractions at 3 fractions a week. Because the radiotherapy data before 1970 are incomplete, we did not perform statistical analysis on radiation dosage. No patients with an SPT were treated with brachytherapy.
Analysis and Statistical Methods The Kaplan-Meier (product limit) method calculated cumulative incidences of SPTs. Incidences were examined in the three subgroups: subjected to external beam irradiation (EBRT) before 12 months of age (early EBRT), after 12 months of age (late EBRT), and no irradiation. In sensitivity analyses, the definition of SPT was changed: in one analysis the definition of Abramson and Frank was used (pineoblastoma is an SPT inside the irradiation field). In another, pineoblastoma was considered an SPT outside the field. The Mantel-Cox test (log rank) determined the statistical significance of differences in the cumulative incidence curves among groups of interest. The end of the period at risk was defined as either the date of diagnosis of the SPT, the date of death from other causes, or the last date that the patient was known to be alive. The cumulative incidence figures are expressed with reference to years of age. The reason for this is that in hereditary retinoblastoma, the age at diagnosis of the SPT is not related to the age at diagnosis of
Moll et al 䡠 Age at Irradiation: Effect on Development of Second Primary Tumors? Table 1. Number of Second Primary Tumors in Irradiated and Nonirradiated Patients before and after the Age of 12 Months Inside and Outside the Irradiation Field in 263 Patients with Hereditary Retinoblastoma
Irradiation ⱕ12 mos Irradiation ⬎12 mos No irradiation Total
Total
Second Primary Tumors
Inside Field
Outside Field
128 55 80 263
24 7 3 34
10 4 0 14
14 3 3* 20
*Hypothetical irradiation field.
the retinoblastoma: subjects are assumed at risk for the development of SPT from birth, because they have a retinoblastoma gene in all somatic cells. BMDP Statistical Software (1992) was used to perform the statistical analyses.
Results The Dutch Retinoblastoma Register listed 686 patients born between 1945 and 1997, comprising 267 (38.9%) hereditary and 419 (61.1%) nonhereditary retinoblastoma patients. Four hereditary patients with retinomas (a benign form of retinoblastoma with spontaneous growth arrest) were excluded. They were diagnosed only after their offspring had a retinoblastoma develop (and none developed an SPT). The mean duration of follow-up of the 263 hereditary retinoblastoma patients was 20 years (median, 18 years; range, 1 month– 48 years). Four of the 263 patients were lost to follow-up. Sixty-three of the 263 patients had died.
Primary Analysis: Pineoblastoma Not an SPT and Outside the Irradiation Field A total of 40 SPTs was diagnosed in 34 hereditary retinoblastoma patients. Patients with multiple SPTs were counted as one case. Thirty-one patients received EBRT; 9 in combination with chemotherapy; a total of 14 tumors were located inside the radiation field (Table 1). At the age of 25, cumulative incidences of SPT were 22% (95% confidence interval, 13%–34%) in the early EBRT group, 3% (0%–14%) in the late EBRT group, and 5% (1%–16%) in the nonirradiated group (overall, P ⫽ 0.001; difference between early and late EBRT, P ⫽ 0.04; Fig 1). Tables 2 and 3 list details of the different SPTs inside and outside the irradiation field and other patient characteristics. In subgroup analyses (Table 4), the cumulative incidences of SPTs inside the irradiation field were 11% (6%–22%) in the early EBRT group, 3% (0%–13%) in the late EBRT group, and 0% (0%– 8%) in the nonirradiated group (given a hypothetical irradiation field; overall, P ⫽ 0.03; Fig 2). In this subgroup, the difference in cumulative incidence of SPT between early and late EBRT was not significant (P ⫽ 0.37). The incidences outside the irradiation field were 11% (6%–22%) in the early EBRT group, 0% (0%–9%) in the late EBRT group, and 5% (1%–16%) in the nonirradiated group (overall, P ⫽ 0.03; Fig 3). In this subgroup, the difference between early and late EBRT was nearly significant (P ⫽ 0.06). We found seven pineoblastomas, five in patients with the familial hereditary form. The mean age at diagnosis of retinoblastoma was 3.2 months (standard deviation, 3.6; median, 1.5; range, 0.5–10 months). Therefore, all pineoblastoma patients were treated for their retinoblastoma before the age of 12 months: 6 patients were irradiated and the seventh received no irradiation but underwent enucleation and xenon laser therapy.
Sensitivity Analysis When pineoblastoma10 is considered an SPT inside the irradiation field, our results resemble those of Abramson and Frank10 (Table 4): a significant difference in SPT incidence inside the field between the early and late EBRT groups (P ⫽ 0.04). When, as a “compromise,” pineoblastoma is considered an SPT, but outside the irradiation field (Table 4), only the incidences of SPT outside the irradiation field differ significantly (inside the irradiation field, P ⫽ 0.18; outside, P ⫽ 0.02).
Discussion This study confirms previous observations of an increased risk of SPT in irradiated hereditary retinoblastoma patients.6,8,14,19,20,22–24 The risk was greater in children diagnosed and irradiated before the age of 1 year. However, the presence of equal numbers of SPTs inside and outside the irradiation field suggests that irradiation is not the cause. Given our irradiation protocol, our restricted but biologically sound definitions of the irradiation field and SPT prevented confirmation of Abramson and Frank’s hypothesis10 that the increase in SPTs in retinoblastoma patients is due to early irradiation before the age of 1 year. We and others6,20 considered only SPTs occurring within or immediately surrounding the orbit of the irradiated eye to be
Figure 1. Kaplan-Meier curve showing the effects of early (before the age of 12 months; n ⫽ 128; stippled line), late (after the age of 12 months; n ⫽ 55; continuous line), and no radiotherapy (n ⫽ 80; dashed line) on second primary tumor incidence.
1111
Ophthalmology Volume 108, Number 6, June 2001 Table 2. Fourteen Second Primary Tumors Inside the Irradiation Field among the 263 Hereditary Retinoblastoma Patients Born, gender Irradiation ⱕ12 mos 1948, M 1951, F# 1954, M# 1961, F 1962, F# 1971, F# 1973, M 1984, F 1987, F# 1988, F# Irradiation ⬎12 mos 1945, F 1950, F# 1956, M 1986, M
Retinoblastoma Year X-ray/Chemotherapy
Second Primary Tumor
Location
Second Primary Tumor Year
Status
1948 EBRT/no 1951 EBRT/no 1954 EBRT/no 1961 EBRT/yes 1963 EBRT/yes 1971 EBRT/no 1973 EBRT/yes 1985 EBRT/no 1987 EBRT/no 1988 EBRT/no
Osteosarcoma Basal cell carcinoma Sarcoma Sarcoma Sarcoma Osteosarcoma Osteosarcoma Rhabdomyosarcoma Rhabdomyosarcoma Osteosarcoma
Orbita Eyelid Face Sinus Face Orbita Zygoma Zygoma Os lacrimalis Nose
1954 1993 1972 1997 1973 1976 1991 1992 1997 1997
Dead Alive Dead Alive Dead Dead Dead Alive Alive Dead
1947 EBRT/no 1952 EBRT/no 1957 EBRT/no 1987 EBRT/no
Fibrosarcoma Sarcoma Melanoma Rhabdomyosarcoma
Face Face Canthus Orbita
1985 1981 1991 1989
Dead Dead Alive Dead
Familial hereditary retinoblastoma; ERBT ⫽ external beam radiation therapy. F ⫽ female; M ⫽ male.
#
radiation associated. In the study of Abramson and Frank,10 the definition of the irradiation field is much wider (i.e., the tissues of the head and neck, including the thyroid gland and pineal gland and other parts of the brain). Furthermore, they consider pineoblastomas an SPT and inside the irradiation field. The original discussion by Jacobiec et al17 of trilateral retinoblastoma makes it clear that these can be distinguished
from metastatic lesions. They defined the pineal tumor as a second intracranial primary (an SPT) but also as an ectopic nonmetastatic retinoblastoma. Zimmerman18 also considered tumors in the pineal and suprasellar areas occurring in survivors of retinoblastoma as ectopic retinoblastomas.25 Some studies have included26 and others have excluded pineoblastomas8,14,19,25,27 as SPTs. If we follow Abramson
Table 3. Seventeen Second Primary Tumors Outside the Irradiation Field and Second Primary Tumors in Nonirradiated Patients among the 263 Hereditary Retinoblastoma Patients Born, Gender Irradiation ⱕ12 mos 1947, M 1948, M 1952, F# 1955, F 1956, M 1959, M# 1959, M 1960, F
1961, F# 1962, M 1968, M 1968, F 1970, F# 1978, F Irradiation ⬎12 mos 1952, F 1957, M 1967, M# No irradiation 1947, M 1950, F 1954, M#
Retinoblastoma Year X-ray/Chemotherapy 1948 EBRT/no 1949 EBRT/no 1953 EBRT/no 1956 EBRT/yes 1957 EBRT/yes 1959 EBRT/yes 1960 EBRT/no 1960 EBRT/no
Second Primary Tumor
Location
1112
Status
Leg Urethra Breast Stomach Knee Femur Skull Knee Parotid Femur Ethmoid Back Shoulder Face Knee Neck Mandible
1985 1985 1995 1997 1973 1971 1979 1979 1991 1992 1996 1983 1991 1977 1994 1976 1981
Alive Dead Alive Dead Dead Dead Dead
1961 EBRT/no 1963 EBRT/yes 1969 EBRT/yes 1968 EBRT/yes 1970 EBRT/no 1978 EBRT/no
Melanoma Transitional cell carcinoma Mammary carcinoma Leiomyosarcoma Osteosarcoma Osteosarcoma Sarcoma Ewing’s sarcoma Mucoepidermoid carcinoma Sarcoma Not specified Melanoma Melanoma Rhabdomyosarcoma Chondrosarcoma Lymphosarcoma Histiocytoma
1956 EBRT/no 1960 EBRT/no 1971 EBRT/no
Fibrosarcoma Melanoma Melanoma
Thigh Back Upper arm
1980 1989 1992
Dead Alive Alive
1948 None/no
Bowen’s disease Leiomyosarcoma Osteosarcoma Melanoma
Limb Not specified Knee Forearm
1979 1980 1962 1977
Dead Dead Alive
1951 None/no 1955 None/no
Familial hereditary retinoblastoma; EBRT ⫽ external beam radiation therapy. F ⫽ female; M ⫽ male.
#
Second Primary Tumor Year
Dead Dead Alive Dead Alive Dead Alive
Moll et al 䡠 Age at Irradiation: Effect on Development of Second Primary Tumors? Table 4. Cumulative Incidences of Second Primary Tumors in Percentage at 25 Years at Age (95% Confidence Interval) According to the Different Definitions of SPT, Age at Irradiation and the Extent of the Irradiation Field
Total
Inside the field
Outside the field
ⱕ12 months ⬎12 months No irradiation Overall P ⱕ12 months ⬎12 months No irradiation Overall P ⱕ12 months ⬎12 months No irradiation Overall P
Our definition: pineoblastoma ⫽ SPT
Abramson & Frank: pineoblastoma ⴝ SPT inside the field
“Compromise”: pineoblastoma ⴝ SPT; outside the field
22 (13–34) P ⫽ 0.04 3 (0–14) 5 (1–16) P ⫽ 0.001 11 (6–22) P ⫽ 0.37 3 (0–13) 0 (0–8) P ⫽ 0.03 11 (6–22) P ⫽ 0.06 0 (0–9) 5 (1–16) P ⫽ 0.03
26 (16–38) P ⫽ 0.01 3 (0–14) 6 (2–18) P ⫽ 0.003 20 (12–32) P ⫽ 0.04 3 (0–14) 2 (0–13) P ⫽ 0.004 7 (3–17) P ⫽ 0.12 0 (0–9) 5 (1–16) P ⫽ 0.06
15 (8–26) P ⫽ 0.18 3 (0–14) 0 (0–8) P ⫽ 0.009 13 (6–23) P ⫽ 0.02 0 (0–9) 6 (2–18) P ⫽ 0.01
SPT ⫽ secondary primary tumors.
and Frank’s definitions for our study population, five SPTs and the six pineoblastomas move inside the irradiation field in favor of their hypothesis. And, indeed, if patients are irradiated before the age of 12 months, more SPTs (including pineoblastomas) are found in the irradiation field than if patients were treated with irradiation after the age of 12 months. Five of the seven pineoblastoma patients had the familial hereditary form of retinoblastoma. The five familial hereditary patients were screened from birth; therefore, their age at diagnosis and treatment was before the age of 12 months. Others have also observed that pineoblastoma tends to occur in children who have retinoblastoma develop early9 and therefore are treated early. Abramson and Frank10 included three pineoblastomas and nine other intracranial tumors as SPTs. It is possible that they included retinoblastoma metastases of the optic nerve and meninges as SPT of the brain and inside the irradiation field, because no biopsy specimens were taken, and no other brain tumors are considered usual for retinoblastoma patients. Our study concerned hereditary unilateral and bilateral retinoblastoma in
contrast to Abramson and Frank who included only bilateral retinoblastoma patients. Because unilateral retinoblastoma patients more often undergo enucleation than irradiation, this could influence the incidence of SPT. It is unclear why Abramson and Frank chose to include only 1-year retinoblastoma survivors and how many of these patients were treated before the age of 1 year; this could introduce a selection bias. Our analyses revealed a significant difference in the rate of SPT between children irradiated before and after 1 year of age, but this difference is not associated with the occurrence of SPT inside the irradiation field. Mohney et al20 recently presented similar results. Four of their 16 SPTs (in 82 hereditary retinoblastoma patients) were considered to be radiation associated. Like us, they concluded that the incidence of SPT (16% at 25 years) occurred predominantly outside the irradiation field. When we defined pineoblastoma as an SPT inside the irradiation field, the increased incidence of SPT inside the irradiation field was significant. The importance of pineoblastoma is also shown in the results of our third analysis: if pineoblastoma is defined as
Figure 2. Second primary tumors inside the irradiation field. Line marking, see Fig 1.
Figure 3. Second primary tumors outside the irradiation field. Line marking, see Figure 1.
1113
Ophthalmology Volume 108, Number 6, June 2001 SPT outside the irradiation field, the associated P values are strongly affected. A limitation of our study is the relatively small number (14) of SPTs inside the irradiation field. Therefore, the power could be too low to detect important differences inside the irradiation field. Another explanation for our findings is the postulation that these could be related to early diagnosis and therefore early treatment/radiation. Also, hereditary retinoblastoma patients diagnosed before the age of 1 year could have an association with SPT because of a specific Rb gene mutation. This Rb gene mutation could be associated with both early retinoblastoma and SPT. Finally, it is possible that in the series of Abramson and Frank10 early diagnosis is associated with a more advanced or aggressive form of retinoblastoma that requires a higher irradiation dosage. Higher dosages of radiation have a direct correlation with the risk of bone cancer developing.23,24 Studies that report relatively small numbers of SPTs have found a lower mean dosage of radiation and fewer SPTs inside the field of irradiation.20 At the moment, fewer patients are treated with EBRT than in the past and more local treatment modalities are available such as brachytherapy, hyperthermia, laser therapy, and cryotherapy, sometimes in combination with chemotherapy. The future will show whether this will lead to fewer SPTs in hereditary retinoblastoma patients. Abramson and Frank10 suggested that the risk of SPT declines with age and may, in some cases, compare favorably with that of systemic chemotherapy. We prefer to avoid EBRT altogether rather than postpone it beyond the age of 1 year. We conclude that patients with hereditary retinoblastoma have an increased risk of SPT developing, especially if they are irradiated early in life. Because SPTs develop with similar incidence inside and outside the field of irradiation, timing of irradiation (diagnosis) may be a marker rather than a cause of this risk.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
References 1. Reese AB, Merriam GR, Martin HE. Treatment of bilateral retinoblastoma by irradiation and surgery: report on 15-year results. Am J Ophthalmol 1949;32:175–90. 2. Forrest AW. Tumors following radiation about the eye. Trans Am Acad Ophthalmol Otolaryngol 1961;65:694 –717. 3. Soloway HB. Radiation-induced neoplasms following curative therapy for retinoblastoma. Cancer 1966;19:1984 – 8. 4. Abramson DH, Ronner HJ, Ellsworth RM. Second tumors in nonirradiated bilateral retinoblastoma. Am J Ophthalmol 1979;87:624 –7. 5. Moll AC, Imhof SM, Bouter LM, Tan KEWP. Second primary tumors in patients with retinoblastoma. A review of the literature. Ophthalmic Genet 1997;18:27–34. 6. Draper GJ, Sanders BM, Kingston JE. Second primary neoplasms in patients with retinoblastoma. Br J Cancer 1986;53: 661–71. 7. 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. 8. DerKinderen DJ, Koten JW, Nagelkerke NJD, et al. Non-
1114
21. 22.
23.
24.
25.
26.
27.
ocular cancer in patients with hereditary retinoblastoma and their relatives. Int J Cancer 1988;41:499 –504. Eng C, Li FP, Abramson DH, et al. Mortality from second tumors among long-term survivors of retinoblastoma. J Natl Cancer Inst 1993;85:1121– 8. Abramson DH, Frank CM. Second nonocular tumors in survivors of bilateral retinoblastoma: a possible age effect on radiation-related risk. Ophthalmology 1998;105:573–9; discussion 579 – 80. Imhof SM, Mourits MP, Hofman P, et al. Quantification of orbital and mid-facial growth retardation after megavoltage external beam irradiation in children with retinoblastoma. Ophthalmology 1996;103:263– 8. Murphree AL, Villablanca JG, Deegan WF 3rd, et al. Chemotherapy plus local treatment in the management of intraocular retinoblastoma. Arch Ophthalmol 1996;114:1348 –56. Gallie BL, Budning A, DeBoer G, et al. Chemotherapy with focal therapy can cure intraocular retinoblastoma without radiotherapy [published erratum appears in Arch Ophthalmol 1997;115:525]. Arch Ophthalmol 1996;114:1321– 8. Moll AC, Imhof SM, Bouter LM, et al. Second primary tumors in patients with hereditary retinoblastoma: a registerbased follow-up study, 1945–1994. Int J Cancer 1996;67: 515–9. Moll AC, Kuik DJ, Bouter LM, et al. Incidence and survival of retinoblastoma in The Netherlands: a register-based study 1862–1995. Br J Ophthalmol 1997;81:559 – 62. Warren S, Gates O. Multiple primary malignant tumors: a survey of the literature and a statistical study. Am J Cancer 1932;16:1358 – 414. Jacobiec JA, Tso MOM, Zimmerman LE, Danis P. Retinoblastoma and intracranial malignancy. Cancer 1977;39:2048 – 58. Zimmerman LE. Retinoblastoma and retinocytoma. In: Spencer WH, ed. Ophthalmic Pathology: An Atlas and Textbook, 3rd ed. Philadelphia: WB Saunders, 1985;Vol 2, 1292–351. Roarty JD, McLean IW, Zimmerman LE. Incidence of second neoplasms in patients with bilateral retinoblastoma. Ophthalmology 1988;95:1583–7. Mohney BG, Robertson DM, Schomberg PJ, Hodge DO. Second nonocular tumors in survivors of heritable retinoblastoma and prior radiation therapy. Am J Ophthalmol 1998;126: 269 –77. Schipper J. Retinoblastoma: a medical and experimental study [thesis]. Rotterdam: 1980. Kony SJ, de Vathaire F, Chompret A, et al. Radiation and genetic factors in the risk of second malignant neoplasms after a first cancer in childhood. Lancet 1997;350:91–5. Tucker MA, D’Angio GJ, Boice JD Jr, et al. Bone sarcomas linked to radiotherapy and chemotherapy in children. N Engl J Med 1987;317:588 –93. Hawkins MM, Wilson LMK, Burton HS, et al. Radiotherapy, alkylating agents, and risk of bone cancer after childhood cancer. J Natl Cancer Inst 1996;88:270 – 8. Abramson DH, Ellsworth RM, Kitchin FD, Tung G. Second nonocular tumors in retinoblastoma survivors. Are they radiation-induced? Ophthalmology 1984;91:1351–5. Lueder GT, Judisch GF, O’Gorman TW. Second nonocular tumors in survivors of heritable retinoblastoma. Arch Ophthalmol 1986;104:372–3. Blach LE, McCormick B, Abramson DH, Ellsworth RM. Trilateral retinoblastoma—incidence and outcome: a decade of experience. Int J Radiat Oncol Biol Phys 1994;29:729 –33.
tion.4 Since 1949, numerous articles have reported on the incidence of SPTs in retinoblastoma subjects. Because these studies differ substantially in design and population size, it is no surprise that our review5 reported a wide range in the incidence of SPTs: we found cumulative incidences of SPTs of 8.4% 18 years after diagnosis,6 15.7% at the age of 20 years,7 19% at the age of 35 years,8 and a relative risk of 15.4 for SPT,9 respectively. 4
Department of Radiotherapy, University Hospital Utrecht, Utrecht, The Netherlands. None of the authors have commercial interests in this study. Reprint requests to Dr. Annette C. Moll, Department of Ophthalmology, Academic Hospital Vrije Universiteit, De Boelelaan 1117, 1081 HV Amsterdam, The Netherlands. E-mail: [email protected]. ISSN 0161-6420/01/$–see front matter PII S0161-6420(01)00562-0
1109
Ophthalmology Volume 108, Number 6, June 2001 Abramson and Frank10 recently suggested in this journal that age at irradiation might explain some of the variability of the reported risk of SPT. This suggestion has intensified the discussion on alternative modalities of therapy. “The implications of this work, if it is replicated by others, are extremely important for the treatment of children with retinoblastoma” as Char commented in the discussion of that article. As a result of the increased incidence of SPT among irradiated patients and other complications such as orbital growth retardation,11 current attention is focused on the use of chemotherapy in combination with local treatment such as brachytherapy, diode laser, and cryotherapy.12,13 We have previously reported on the incidence, survival, and cumulative incidence of SPT and pineoblastomas in retinoblastoma patients born in The Netherlands.14,15 Our register-based follow-up cohort study can also contribute to the discussion of a possible age effect on radiation-related risk of SPT. Therefore, the aim of this study is to evaluate the influence of age at irradiation on the occurrence of SPT inside and outside the irradiation field in hereditary retinoblastoma.
Material and Methods Study Population and Follow-up Procedures A follow-up study was performed on retinoblastoma patients born in The Netherlands between 1945 and 1997. A previous report contains a complete description of the methods used to locate survivors and document deaths.14 In short, the Dutch Retinoblastoma Register was completed with the assistance of all University Eye Clinics and Cancer Registration Centers. Hospital records supplied data about demography, tumor laterality, family history of retinoblastoma, treatment, occurrence of SPT, and date and cause of death. Every effort was made to determine the survival for each retinoblastoma patient. The most recent address of each patient was determined by means of the municipal registries and the Central Bureau of Genealogy, and the patient’s status (alive or dead) was recorded. Hereditary retinoblastoma patients were followed for the occurrence of SPT. Different methods were used: patients with hereditary retinoblastoma born between 1945 and 1970 (n ⫽ 146) had previously been visited at home in 1985 (n ⫽ 141) to investigate the incidence of SPT.8 In alive patients (n ⫽ 101) new information on health, SPT, and therapy was obtained by a questionnaire, if necessary in Braille. In deceased patients information on the cause of death was obtained by means of the most recent treating physician (n ⫽ 6). Patients born between 1970 and 1997 (n ⫽ 121) and their parents were approached through their own ophthalmologist. A detailed history was obtained by interview at home or in the hospital. Reported SPTs were confirmed by clinical and histopathologic reports. We considered retinoblastoma hereditary if one or more of the following criteria were met: a bilateral retinoblastoma, a positive family history for retinoblastoma (then the related parent will be the carrier of the defect in the retinoblastoma gene), or a defect in the retinoblastoma gene found in chromosomal/DNA analysis of the subject. The criteria of Warren and Gates16 defined SPT: each of the tumors must present a definite picture of malignancy; each tumor must be distinct (different histologic appearance); and the probability that the SPT is a metastatic lesion from the primary tumor must be excluded. Because a pineoblastoma is histologically iden-
1110
tical to a retinoblastoma and therefore also called a trilateral retinoblastoma, we and others17,18 do not consider pineoblastoma an SPT.
Field Definition and Radiotherapy Technique Tumors were classified as inside the field of irradiation if they originated in the lids, orbits, periocular sinuses, temporal bones, or skin overlying the temporal bone region. Other locations, such as thyroid gland, neck, and brain, were defined as outside the field of irradiation19,20 on the basis of the following considerations. The radiation data were assumed to be reliable for patients treated from 1971 onward. Beginning in 1971, all patients were irradiated by a 6- or 8-MV photon beam in a specially collimated field: 20 ⫻ 26 mm and D-shaped in the sagittal plane. The dosage was 45 Gy in 15 fractions, 3 fractions a week. The field covers the entire retina, the straight edge of the D lying close to the posterior pole of the lens, and the rounded part facing posteriorly. The width corresponds with the central axis of the eye. The field extends from the posterior pole of the lens in the direction of the chiasma over 20 mm. In our patients, the mean distance between the posterior pole of the lens and the internal side of the retina (12 MHz ultrasonic A-scan) was 15 mm (standard deviation, 0.3). In a few patients, the thickness of the posterior side of the sclera could be determined at 3 mm. The typical distance from the posterior pole of the lens to the entrance of the optic nerve is therefore 18 mm. The field size in that direction was 20 mm. As stated previously, the field is D-shaped on sagittal cross-section, with the rounded part facing posteriorly. This boundary follows the posterior scleral contour at a distance of 2 mm. It is now the penumbra of this specially collimated field that determines the dosage delivered to the surrounding structures, both the lens and the beginning of the optic nerve. The penumbra width in which the dosage decreases from 50% to 10% of the central field maximum dosage is only 3 mm at a depth of 2 cm under the skin. In unilateral retinoblastoma, the external irradiation was performed with one single temporal lateral field of 20 ⫻ 26 mm; in bilateral retinoblastoma, two opposing identical temporal fields were applied. This technique, developed by Schipper21 in the late 1960s offers a reproduction accuracy of at least 1 mm. From these statistics we can estimate the irradiation dosage to the sella region. If the distance from the posterior pole of the eye to the chiasma is assumed to be 2 cm, the irradiation dosage to the chiasma will be less than 1% of the maximum dosage of 45 Gy in 15 fractions at 3 fractions a week. Because the radiotherapy data before 1970 are incomplete, we did not perform statistical analysis on radiation dosage. No patients with an SPT were treated with brachytherapy.
Analysis and Statistical Methods The Kaplan-Meier (product limit) method calculated cumulative incidences of SPTs. Incidences were examined in the three subgroups: subjected to external beam irradiation (EBRT) before 12 months of age (early EBRT), after 12 months of age (late EBRT), and no irradiation. In sensitivity analyses, the definition of SPT was changed: in one analysis the definition of Abramson and Frank was used (pineoblastoma is an SPT inside the irradiation field). In another, pineoblastoma was considered an SPT outside the field. The Mantel-Cox test (log rank) determined the statistical significance of differences in the cumulative incidence curves among groups of interest. The end of the period at risk was defined as either the date of diagnosis of the SPT, the date of death from other causes, or the last date that the patient was known to be alive. The cumulative incidence figures are expressed with reference to years of age. The reason for this is that in hereditary retinoblastoma, the age at diagnosis of the SPT is not related to the age at diagnosis of
Moll et al 䡠 Age at Irradiation: Effect on Development of Second Primary Tumors? Table 1. Number of Second Primary Tumors in Irradiated and Nonirradiated Patients before and after the Age of 12 Months Inside and Outside the Irradiation Field in 263 Patients with Hereditary Retinoblastoma
Irradiation ⱕ12 mos Irradiation ⬎12 mos No irradiation Total
Total
Second Primary Tumors
Inside Field
Outside Field
128 55 80 263
24 7 3 34
10 4 0 14
14 3 3* 20
*Hypothetical irradiation field.
the retinoblastoma: subjects are assumed at risk for the development of SPT from birth, because they have a retinoblastoma gene in all somatic cells. BMDP Statistical Software (1992) was used to perform the statistical analyses.
Results The Dutch Retinoblastoma Register listed 686 patients born between 1945 and 1997, comprising 267 (38.9%) hereditary and 419 (61.1%) nonhereditary retinoblastoma patients. Four hereditary patients with retinomas (a benign form of retinoblastoma with spontaneous growth arrest) were excluded. They were diagnosed only after their offspring had a retinoblastoma develop (and none developed an SPT). The mean duration of follow-up of the 263 hereditary retinoblastoma patients was 20 years (median, 18 years; range, 1 month– 48 years). Four of the 263 patients were lost to follow-up. Sixty-three of the 263 patients had died.
Primary Analysis: Pineoblastoma Not an SPT and Outside the Irradiation Field A total of 40 SPTs was diagnosed in 34 hereditary retinoblastoma patients. Patients with multiple SPTs were counted as one case. Thirty-one patients received EBRT; 9 in combination with chemotherapy; a total of 14 tumors were located inside the radiation field (Table 1). At the age of 25, cumulative incidences of SPT were 22% (95% confidence interval, 13%–34%) in the early EBRT group, 3% (0%–14%) in the late EBRT group, and 5% (1%–16%) in the nonirradiated group (overall, P ⫽ 0.001; difference between early and late EBRT, P ⫽ 0.04; Fig 1). Tables 2 and 3 list details of the different SPTs inside and outside the irradiation field and other patient characteristics. In subgroup analyses (Table 4), the cumulative incidences of SPTs inside the irradiation field were 11% (6%–22%) in the early EBRT group, 3% (0%–13%) in the late EBRT group, and 0% (0%– 8%) in the nonirradiated group (given a hypothetical irradiation field; overall, P ⫽ 0.03; Fig 2). In this subgroup, the difference in cumulative incidence of SPT between early and late EBRT was not significant (P ⫽ 0.37). The incidences outside the irradiation field were 11% (6%–22%) in the early EBRT group, 0% (0%–9%) in the late EBRT group, and 5% (1%–16%) in the nonirradiated group (overall, P ⫽ 0.03; Fig 3). In this subgroup, the difference between early and late EBRT was nearly significant (P ⫽ 0.06). We found seven pineoblastomas, five in patients with the familial hereditary form. The mean age at diagnosis of retinoblastoma was 3.2 months (standard deviation, 3.6; median, 1.5; range, 0.5–10 months). Therefore, all pineoblastoma patients were treated for their retinoblastoma before the age of 12 months: 6 patients were irradiated and the seventh received no irradiation but underwent enucleation and xenon laser therapy.
Sensitivity Analysis When pineoblastoma10 is considered an SPT inside the irradiation field, our results resemble those of Abramson and Frank10 (Table 4): a significant difference in SPT incidence inside the field between the early and late EBRT groups (P ⫽ 0.04). When, as a “compromise,” pineoblastoma is considered an SPT, but outside the irradiation field (Table 4), only the incidences of SPT outside the irradiation field differ significantly (inside the irradiation field, P ⫽ 0.18; outside, P ⫽ 0.02).
Discussion This study confirms previous observations of an increased risk of SPT in irradiated hereditary retinoblastoma patients.6,8,14,19,20,22–24 The risk was greater in children diagnosed and irradiated before the age of 1 year. However, the presence of equal numbers of SPTs inside and outside the irradiation field suggests that irradiation is not the cause. Given our irradiation protocol, our restricted but biologically sound definitions of the irradiation field and SPT prevented confirmation of Abramson and Frank’s hypothesis10 that the increase in SPTs in retinoblastoma patients is due to early irradiation before the age of 1 year. We and others6,20 considered only SPTs occurring within or immediately surrounding the orbit of the irradiated eye to be
Figure 1. Kaplan-Meier curve showing the effects of early (before the age of 12 months; n ⫽ 128; stippled line), late (after the age of 12 months; n ⫽ 55; continuous line), and no radiotherapy (n ⫽ 80; dashed line) on second primary tumor incidence.
1111
Ophthalmology Volume 108, Number 6, June 2001 Table 2. Fourteen Second Primary Tumors Inside the Irradiation Field among the 263 Hereditary Retinoblastoma Patients Born, gender Irradiation ⱕ12 mos 1948, M 1951, F# 1954, M# 1961, F 1962, F# 1971, F# 1973, M 1984, F 1987, F# 1988, F# Irradiation ⬎12 mos 1945, F 1950, F# 1956, M 1986, M
Retinoblastoma Year X-ray/Chemotherapy
Second Primary Tumor
Location
Second Primary Tumor Year
Status
1948 EBRT/no 1951 EBRT/no 1954 EBRT/no 1961 EBRT/yes 1963 EBRT/yes 1971 EBRT/no 1973 EBRT/yes 1985 EBRT/no 1987 EBRT/no 1988 EBRT/no
Osteosarcoma Basal cell carcinoma Sarcoma Sarcoma Sarcoma Osteosarcoma Osteosarcoma Rhabdomyosarcoma Rhabdomyosarcoma Osteosarcoma
Orbita Eyelid Face Sinus Face Orbita Zygoma Zygoma Os lacrimalis Nose
1954 1993 1972 1997 1973 1976 1991 1992 1997 1997
Dead Alive Dead Alive Dead Dead Dead Alive Alive Dead
1947 EBRT/no 1952 EBRT/no 1957 EBRT/no 1987 EBRT/no
Fibrosarcoma Sarcoma Melanoma Rhabdomyosarcoma
Face Face Canthus Orbita
1985 1981 1991 1989
Dead Dead Alive Dead
Familial hereditary retinoblastoma; ERBT ⫽ external beam radiation therapy. F ⫽ female; M ⫽ male.
#
radiation associated. In the study of Abramson and Frank,10 the definition of the irradiation field is much wider (i.e., the tissues of the head and neck, including the thyroid gland and pineal gland and other parts of the brain). Furthermore, they consider pineoblastomas an SPT and inside the irradiation field. The original discussion by Jacobiec et al17 of trilateral retinoblastoma makes it clear that these can be distinguished
from metastatic lesions. They defined the pineal tumor as a second intracranial primary (an SPT) but also as an ectopic nonmetastatic retinoblastoma. Zimmerman18 also considered tumors in the pineal and suprasellar areas occurring in survivors of retinoblastoma as ectopic retinoblastomas.25 Some studies have included26 and others have excluded pineoblastomas8,14,19,25,27 as SPTs. If we follow Abramson
Table 3. Seventeen Second Primary Tumors Outside the Irradiation Field and Second Primary Tumors in Nonirradiated Patients among the 263 Hereditary Retinoblastoma Patients Born, Gender Irradiation ⱕ12 mos 1947, M 1948, M 1952, F# 1955, F 1956, M 1959, M# 1959, M 1960, F
1961, F# 1962, M 1968, M 1968, F 1970, F# 1978, F Irradiation ⬎12 mos 1952, F 1957, M 1967, M# No irradiation 1947, M 1950, F 1954, M#
Retinoblastoma Year X-ray/Chemotherapy 1948 EBRT/no 1949 EBRT/no 1953 EBRT/no 1956 EBRT/yes 1957 EBRT/yes 1959 EBRT/yes 1960 EBRT/no 1960 EBRT/no
Second Primary Tumor
Location
1112
Status
Leg Urethra Breast Stomach Knee Femur Skull Knee Parotid Femur Ethmoid Back Shoulder Face Knee Neck Mandible
1985 1985 1995 1997 1973 1971 1979 1979 1991 1992 1996 1983 1991 1977 1994 1976 1981
Alive Dead Alive Dead Dead Dead Dead
1961 EBRT/no 1963 EBRT/yes 1969 EBRT/yes 1968 EBRT/yes 1970 EBRT/no 1978 EBRT/no
Melanoma Transitional cell carcinoma Mammary carcinoma Leiomyosarcoma Osteosarcoma Osteosarcoma Sarcoma Ewing’s sarcoma Mucoepidermoid carcinoma Sarcoma Not specified Melanoma Melanoma Rhabdomyosarcoma Chondrosarcoma Lymphosarcoma Histiocytoma
1956 EBRT/no 1960 EBRT/no 1971 EBRT/no
Fibrosarcoma Melanoma Melanoma
Thigh Back Upper arm
1980 1989 1992
Dead Alive Alive
1948 None/no
Bowen’s disease Leiomyosarcoma Osteosarcoma Melanoma
Limb Not specified Knee Forearm
1979 1980 1962 1977
Dead Dead Alive
1951 None/no 1955 None/no
Familial hereditary retinoblastoma; EBRT ⫽ external beam radiation therapy. F ⫽ female; M ⫽ male.
#
Second Primary Tumor Year
Dead Dead Alive Dead Alive Dead Alive
Moll et al 䡠 Age at Irradiation: Effect on Development of Second Primary Tumors? Table 4. Cumulative Incidences of Second Primary Tumors in Percentage at 25 Years at Age (95% Confidence Interval) According to the Different Definitions of SPT, Age at Irradiation and the Extent of the Irradiation Field
Total
Inside the field
Outside the field
ⱕ12 months ⬎12 months No irradiation Overall P ⱕ12 months ⬎12 months No irradiation Overall P ⱕ12 months ⬎12 months No irradiation Overall P
Our definition: pineoblastoma ⫽ SPT
Abramson & Frank: pineoblastoma ⴝ SPT inside the field
“Compromise”: pineoblastoma ⴝ SPT; outside the field
22 (13–34) P ⫽ 0.04 3 (0–14) 5 (1–16) P ⫽ 0.001 11 (6–22) P ⫽ 0.37 3 (0–13) 0 (0–8) P ⫽ 0.03 11 (6–22) P ⫽ 0.06 0 (0–9) 5 (1–16) P ⫽ 0.03
26 (16–38) P ⫽ 0.01 3 (0–14) 6 (2–18) P ⫽ 0.003 20 (12–32) P ⫽ 0.04 3 (0–14) 2 (0–13) P ⫽ 0.004 7 (3–17) P ⫽ 0.12 0 (0–9) 5 (1–16) P ⫽ 0.06
15 (8–26) P ⫽ 0.18 3 (0–14) 0 (0–8) P ⫽ 0.009 13 (6–23) P ⫽ 0.02 0 (0–9) 6 (2–18) P ⫽ 0.01
SPT ⫽ secondary primary tumors.
and Frank’s definitions for our study population, five SPTs and the six pineoblastomas move inside the irradiation field in favor of their hypothesis. And, indeed, if patients are irradiated before the age of 12 months, more SPTs (including pineoblastomas) are found in the irradiation field than if patients were treated with irradiation after the age of 12 months. Five of the seven pineoblastoma patients had the familial hereditary form of retinoblastoma. The five familial hereditary patients were screened from birth; therefore, their age at diagnosis and treatment was before the age of 12 months. Others have also observed that pineoblastoma tends to occur in children who have retinoblastoma develop early9 and therefore are treated early. Abramson and Frank10 included three pineoblastomas and nine other intracranial tumors as SPTs. It is possible that they included retinoblastoma metastases of the optic nerve and meninges as SPT of the brain and inside the irradiation field, because no biopsy specimens were taken, and no other brain tumors are considered usual for retinoblastoma patients. Our study concerned hereditary unilateral and bilateral retinoblastoma in
contrast to Abramson and Frank who included only bilateral retinoblastoma patients. Because unilateral retinoblastoma patients more often undergo enucleation than irradiation, this could influence the incidence of SPT. It is unclear why Abramson and Frank chose to include only 1-year retinoblastoma survivors and how many of these patients were treated before the age of 1 year; this could introduce a selection bias. Our analyses revealed a significant difference in the rate of SPT between children irradiated before and after 1 year of age, but this difference is not associated with the occurrence of SPT inside the irradiation field. Mohney et al20 recently presented similar results. Four of their 16 SPTs (in 82 hereditary retinoblastoma patients) were considered to be radiation associated. Like us, they concluded that the incidence of SPT (16% at 25 years) occurred predominantly outside the irradiation field. When we defined pineoblastoma as an SPT inside the irradiation field, the increased incidence of SPT inside the irradiation field was significant. The importance of pineoblastoma is also shown in the results of our third analysis: if pineoblastoma is defined as
Figure 2. Second primary tumors inside the irradiation field. Line marking, see Fig 1.
Figure 3. Second primary tumors outside the irradiation field. Line marking, see Figure 1.
1113
Ophthalmology Volume 108, Number 6, June 2001 SPT outside the irradiation field, the associated P values are strongly affected. A limitation of our study is the relatively small number (14) of SPTs inside the irradiation field. Therefore, the power could be too low to detect important differences inside the irradiation field. Another explanation for our findings is the postulation that these could be related to early diagnosis and therefore early treatment/radiation. Also, hereditary retinoblastoma patients diagnosed before the age of 1 year could have an association with SPT because of a specific Rb gene mutation. This Rb gene mutation could be associated with both early retinoblastoma and SPT. Finally, it is possible that in the series of Abramson and Frank10 early diagnosis is associated with a more advanced or aggressive form of retinoblastoma that requires a higher irradiation dosage. Higher dosages of radiation have a direct correlation with the risk of bone cancer developing.23,24 Studies that report relatively small numbers of SPTs have found a lower mean dosage of radiation and fewer SPTs inside the field of irradiation.20 At the moment, fewer patients are treated with EBRT than in the past and more local treatment modalities are available such as brachytherapy, hyperthermia, laser therapy, and cryotherapy, sometimes in combination with chemotherapy. The future will show whether this will lead to fewer SPTs in hereditary retinoblastoma patients. Abramson and Frank10 suggested that the risk of SPT declines with age and may, in some cases, compare favorably with that of systemic chemotherapy. We prefer to avoid EBRT altogether rather than postpone it beyond the age of 1 year. We conclude that patients with hereditary retinoblastoma have an increased risk of SPT developing, especially if they are irradiated early in life. Because SPTs develop with similar incidence inside and outside the field of irradiation, timing of irradiation (diagnosis) may be a marker rather than a cause of this risk.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
References 1. Reese AB, Merriam GR, Martin HE. Treatment of bilateral retinoblastoma by irradiation and surgery: report on 15-year results. Am J Ophthalmol 1949;32:175–90. 2. Forrest AW. Tumors following radiation about the eye. Trans Am Acad Ophthalmol Otolaryngol 1961;65:694 –717. 3. Soloway HB. Radiation-induced neoplasms following curative therapy for retinoblastoma. Cancer 1966;19:1984 – 8. 4. Abramson DH, Ronner HJ, Ellsworth RM. Second tumors in nonirradiated bilateral retinoblastoma. Am J Ophthalmol 1979;87:624 –7. 5. Moll AC, Imhof SM, Bouter LM, Tan KEWP. Second primary tumors in patients with retinoblastoma. A review of the literature. Ophthalmic Genet 1997;18:27–34. 6. Draper GJ, Sanders BM, Kingston JE. Second primary neoplasms in patients with retinoblastoma. Br J Cancer 1986;53: 661–71. 7. 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. 8. DerKinderen DJ, Koten JW, Nagelkerke NJD, et al. Non-
1114
21. 22.
23.
24.
25.
26.
27.
ocular cancer in patients with hereditary retinoblastoma and their relatives. Int J Cancer 1988;41:499 –504. Eng C, Li FP, Abramson DH, et al. Mortality from second tumors among long-term survivors of retinoblastoma. J Natl Cancer Inst 1993;85:1121– 8. Abramson DH, Frank CM. Second nonocular tumors in survivors of bilateral retinoblastoma: a possible age effect on radiation-related risk. Ophthalmology 1998;105:573–9; discussion 579 – 80. Imhof SM, Mourits MP, Hofman P, et al. Quantification of orbital and mid-facial growth retardation after megavoltage external beam irradiation in children with retinoblastoma. Ophthalmology 1996;103:263– 8. Murphree AL, Villablanca JG, Deegan WF 3rd, et al. Chemotherapy plus local treatment in the management of intraocular retinoblastoma. Arch Ophthalmol 1996;114:1348 –56. Gallie BL, Budning A, DeBoer G, et al. Chemotherapy with focal therapy can cure intraocular retinoblastoma without radiotherapy [published erratum appears in Arch Ophthalmol 1997;115:525]. Arch Ophthalmol 1996;114:1321– 8. Moll AC, Imhof SM, Bouter LM, et al. Second primary tumors in patients with hereditary retinoblastoma: a registerbased follow-up study, 1945–1994. Int J Cancer 1996;67: 515–9. Moll AC, Kuik DJ, Bouter LM, et al. Incidence and survival of retinoblastoma in The Netherlands: a register-based study 1862–1995. Br J Ophthalmol 1997;81:559 – 62. Warren S, Gates O. Multiple primary malignant tumors: a survey of the literature and a statistical study. Am J Cancer 1932;16:1358 – 414. Jacobiec JA, Tso MOM, Zimmerman LE, Danis P. Retinoblastoma and intracranial malignancy. Cancer 1977;39:2048 – 58. Zimmerman LE. Retinoblastoma and retinocytoma. In: Spencer WH, ed. Ophthalmic Pathology: An Atlas and Textbook, 3rd ed. Philadelphia: WB Saunders, 1985;Vol 2, 1292–351. Roarty JD, McLean IW, Zimmerman LE. Incidence of second neoplasms in patients with bilateral retinoblastoma. Ophthalmology 1988;95:1583–7. Mohney BG, Robertson DM, Schomberg PJ, Hodge DO. Second nonocular tumors in survivors of heritable retinoblastoma and prior radiation therapy. Am J Ophthalmol 1998;126: 269 –77. Schipper J. Retinoblastoma: a medical and experimental study [thesis]. Rotterdam: 1980. Kony SJ, de Vathaire F, Chompret A, et al. Radiation and genetic factors in the risk of second malignant neoplasms after a first cancer in childhood. Lancet 1997;350:91–5. Tucker MA, D’Angio GJ, Boice JD Jr, et al. Bone sarcomas linked to radiotherapy and chemotherapy in children. N Engl J Med 1987;317:588 –93. Hawkins MM, Wilson LMK, Burton HS, et al. Radiotherapy, alkylating agents, and risk of bone cancer after childhood cancer. J Natl Cancer Inst 1996;88:270 – 8. Abramson DH, Ellsworth RM, Kitchin FD, Tung G. Second nonocular tumors in retinoblastoma survivors. Are they radiation-induced? Ophthalmology 1984;91:1351–5. Lueder GT, Judisch GF, O’Gorman TW. Second nonocular tumors in survivors of heritable retinoblastoma. Arch Ophthalmol 1986;104:372–3. Blach LE, McCormick B, Abramson DH, Ellsworth RM. Trilateral retinoblastoma—incidence and outcome: a decade of experience. Int J Radiat Oncol Biol Phys 1994;29:729 –33.