Article Text

Download PDFPDF

Salvage external beam radiotherapy after failed primary chemotherapy for bilateral retinoblastoma: rate of eye and vision preservation
  1. M P Y Chan1,
  2. J L Hungerford1,
  3. J E Kingston2,
  4. P N Plowman3
  1. 1
    Ocular Oncology Service, St Bartholomew’s and The Royal London Hospitals, London, UK
  2. 2
    Department of Paediatric Oncology, St Bartholomew’s and The Royal London Hospitals, London, UK
  3. 3
    Department of Clinical Oncology, St Bartholomew’s and The Royal London Hospitals, London, UK
  1. Mr J L Hungerford, Ocular Oncology Service, St Bartholomew’s Hospital, London EC1A 7BE, UK; john.hungerford{at}


Background: Radiation is implicated in the induction of second malignancies in children with bilateral retinoblastoma. There is a need to determine whether this risk can be justified by good visual outcome when external beam radiotherapy (EBRT) is used as a salvage treatment.

Aim: To study the effectiveness of EBRT as a salvage treatment after failed primary chemotherapy and focal treatment in bilateral retinoblastoma.

Methods: This is a retrospective observational case series. The outcome measures after EBRT are: rate of eye preservation, rate of tumour control, visual potential, visual acuity and radiation-induced side-effects.

Results: Thirty-six eyes (22 patients) were included. The median follow-up after EBRT was 40 months (19–165 months). Thirty-two eyes received lens-sparing radiotherapy, and four received whole-eye radiation. The rate of eye preservation was 83.3% (30/36 eyes). Twenty-four eyes (66.7%) were controlled by EBRT and required no further treatment. Of the 30 preserved eyes, 20 eyes (66.7%) had extramacular tumours without retinal detachment and therefore potential for central vision. The final visual acuity was recorded for 19 eyes. Ten eyes (52.6%) read 6/9–6/5, three eyes (15.8%) read 6/18–6/36, and six eyes (31.6%) read 6/60 or worse. Significant radiation- induced side effects were limited to cataracts and dry eyes with whole-eye radiation. There were no second cancers or deaths.

Conclusion: Salvage EBRT is highly effective in preserving eyes with useful vision in bilateral retinoblastoma after failed chemotherapy and focal treatments. These results will help the parents and ophthalmologists of such patients to reach an informed decision when weighing up the benefits of EBRT against its potential oncogenic effect.

Statistics from

Children with heritable retinoblastoma, which includes all bilateral and some unilateral cases, carry a germline mutation in the Rb1 gene. This mutation markedly increases the risk of early-onset extraocular cancers, notably osteosarcoma, soft tissue sarcoma and melanoma. It also increases the chance of an affected child developing ectopic intracranial retinoblastoma, also called pinealoblastoma or “trilateral retinoblastoma.” Cohort studies in the US have shown that the cumulative incidence of a second cancer at 50 years after diagnosis of heritable retinoblastoma was between 36%1 and 51.0%,2 compared with between 5.0%2 and 5.7%1 after non-heritable retinoblastoma. Survivors of heritable retinoblastoma also have a higher lifetime rate of mortality from common epithelial cancers such as breast and lung cancers compared with the general population.3 This trend is also demonstrated in the UK where the long-term survival rate after bilateral retinoblastoma is worse than in unilateral cases due to late deaths from second primary cancers and intracranial retinoblastoma.4

In the past, bilateral retinoblastoma was commonly treated by enucleation of the worse eye, and the remaining globe would be treated by external beam radiotherapy (EBRT). Chemotherapy was employed for metastatic disease, as an adjuvant therapy for high-risk histology in an enucleated eye, or in conjunction with radiotherapy for advanced intraocular retinoblastoma.5 Over the past decade, chemotherapy has largely supplanted EBRT as the primary treatment of intraocular retinoblastoma, including early-stage disease. Chemotherapy is now the first-line treatment of choice in most cases of bilateral retinoblastoma, often in conjunction with more focused consolidation treatments—such as cryotherapy, radioactive scleral plaque brachytherapy and thermotherapy—to enhance tumour control where necessary. This combination has eliminated the need for EBRT in most patients. However, EBRT is still a viable salvage option for tumours that are refractory to combinations of primary chemotherapy and focal treatment, or tumours that are too large, too numerous or too close to the optic disc or fovea for focal therapy. EBRT also has a special role in treating eyes with vitreous seeds.

The side effects associated with EBRT have been a major issue. Early whole-eye radiation caused blinding and painful anterior segment complications, but they have now been overcome with treatments that optimise radiation doses and fractionation. Cataract development can be effectively avoided using lens-sparing radiation, although whole-eye radiation is still required occasionally in treating anterior tumours. The greatest concern that still remains is radiation-induced malignancies. In heritable retinoblastoma, radiation further increases the risk of second cancers by 3.1-fold compared with non-irradiated heritable retinoblastoma patients.1 The effect is greatest in the field of irradiation1 and the penumbra of the beam, if exposure occurs before age 12 months;5 for soft-tissue sarcomas, the risk is dose-related.2

In view of these risks associated with EBRT, there is a need to determine whether they are justified by the achievement of eye preservation and good visual outcome when EBRT is used as a salvage treatment for bilateral retinoblastoma.


To study the effectiveness of salvage EBRT after failed primary chemotherapy and focal treatment for bilateral retinoblastoma


This is a retrospective observational case series from St Bartholomew’s & the Royal London Hospitals, UK. During the period of 1991–2006, our department treated 90 patients with bilateral retinoblastoma using primary chemotherapy and focal treatments. Twenty-two of these patients had failed primary chemotherapy and subsequently required salvage EBRT and were included in this study. The primary chemotherapy regimen was vincristine, carboplatin and etoposide. The majority of patients received six cycles of chemotherapy. One patient received only three cycles due to an adverse drug reaction. Two patients received eight cycles, and one received seven cycles. Focal consolidation treatment—including cryotherapy, radiation brachytherapy, xenon arc photocoagulation (in the children treated early in the study), diode laser and thermotherapy—was added during the course of treatment when necessary. All patients required subsequent salvage EBRT treatment because of recurrences (tumours that have shown a reduction in size or known regression pattern, but developed a new form of activity at the primary site) or unresponsive disease (no reduction in size or character of tumours). The total prescription dose of radiotherapy was 40–44 Gy delivered in 20–22 fractions over 4 to 5 weeks. Thirty-two eyes received lens-sparing radiotherapy, and four received whole-eye radiation. Eyes that were enucleated before EBRT were not included in the study.

The patients were examined by a consultant ocular oncologist (JLH) under general anaesthesia, using indirect ophthalmoscopy with scleral indentation, ultrasound B-scan and Retcam digital retinal imaging.

The outcome measures after EBRT were: (1) rate of eye preservation, (2) rate of tumour control, (3) potential for central vision (eyes with extramacular tumours without retinal detachment), (4) visual acuity and (5) radiation-induced side effects.


Thirty-six eyes in 22 patients were included in this study, as eight eyes were enucleated before EBRT (fig 1), but the Reese–Ellsworth classification of all eyes on presentation and their final outcome are shown in table 1. The median age at diagnosis was 3 months (range 5 days to 26 months) and the median age at which EBRT was given was 19 months (range 8–49 months). In 18 (82%) patients, it was given after age 12 months. The median duration of follow-up after diagnosis was 59 months (range 30–170 months), and the follow-up after EBRT was 40 months (19–165 months).

Figure 1

Treatment pathway of eyes in the 22 patients. EBRT, external beam radiotherapy. Eyes that were enucleated before EBRT were excluded from analysis.

Table 1 Reese–Ellsworth classification on presentation of all eyes and their final outcome*

Rate of eye preservation and visual outcome

Six eyes failed EBRT and were enucleated at a median of 19 months afterwards (range 9–27 months). The rate of eye preservation after EBRT was therefore 83.3% (30 eyes) at 40 months. The likelihood of salvage was better with low-grade retinoblastoma on presentation, although for Reese–Ellsworth group IV–V eyes, 28.6–62.5% were saved (table 1). Overall, among the 22 patients, 14 out of 44 eyes were enucleated; eight before EBRT and six after EBRT (fig 1). Three patients had bilateral enucleations.

Of the retained eyes, 20 (66.7%) were found to have extramacular tumours without retinal detachment and therefore potential for central vision. Of the eyes that were retained, 19 had their visual acuity recorded (table 2). Of these, 10 eyes (52.6%) read 6/5–6/9, one eye (5.3%) read 6/18, two eyes (10.5%) read 6/36, three eyes (15.8%) read 6/60, one eye (5.3%) had counting finger vision, and two eyes (10.5%) had hand movement vision.

Table 2 Final recorded visual acuity of the salvaged eyes*

Rate of tumour control

In 24 eyes (66.7%) local tumour control was established after EBRT, and no further treatment was required. Twelve eyes (33.3%) developed recurrences and new tumours which required a combination of further treatments (fig 2). Ten eyes received cryotherapy, four had ruthenium plaque brachytherapy, four had diode laser therapy, one had subconjunctival carboplatin injection, and one had orbital floor carboplatin injection. Four patients (five eyes) in addition received either four or six courses of second line chemotherapy with vincristine, doxorubicin and ifosfamide after EBRT, and three of those eyes were eventually enucleated.

Figure 2

Response after external beam radiotherapy. All tumours in all eyes were inert at the last follow-up.

Radiation-induced side effects

There were no metastases, extraocular cancers or deaths reported in this group of patients. The four eyes that received whole-eye radiation developed keratoconjunctivitis sicca and cataracts, three underwent lensectomy, and one was eventually enucleated due to uncontrolled retinoblastoma. Middle facial growth retardation was seen after both modalities of radiation, but it was more evident after whole-eye radiation, and with younger age of the patient. Radiation retinopathy was not seen in any patient.


Since the development of effective chemotherapy for the treatment of retinoblastoma, EBRT is required infrequently, and its use has been actively avoided. The main deterrent at present is its significant tumour-inducing effect, a risk which is magnified 3.1-fold among heritable retinoblastoma patients.1

In fact, when faced with unresponsive or uncontrolled disease after primary chemotherapy and focal treatments, many ophthalmologists would opt for enucleation, or persevere with focal treatment in visually important sites in the hope of saving the eye while vision is sacrificed. These choices are justifiable in cases in which the visual prognosis is deemed poor.

In this study, however, the findings demonstrate that EBRT is a highly effective salvage treatment for bilateral retinoblastoma, and it should be considered a worthy salvage option in certain cases. After failed primary chemotherapy and focal treatment, EBRT alone achieves tumour control in 66.7% of eyes; together with additional focal treatment, 83.3% of eyes are salvaged.

Most importantly, salvage EBRT enables vision to be retained. Two-thirds of eyes preserved after treatment are deemed to have the potential for central vision, and 52.6% of eyes with recorded visual acuity have 6/5–6/9 reading and driving vision, with the remaining all retaining useful vision of at least hand movement. This compares favourably with the visual outcome in other studies with or without EBRT. Demirci et al reported that for eyes receiving primary chemotherapy and focal treatments alone, 50% read 20/40 after at least 5 years’ follow-up.6 One study employed two-cycle neo-adjuvant chemotherapy and primary whole eye radiation, but unlike our study, their patients all had advanced bilateral Reese–Ellsworth group V disease.7 Most eyes were salvaged, but less than half of the patients had vision better than 1/60. In a study of 11 eyes with macular retinoblastoma treated with EBRT, 10 eyes were retained, and eight eyes read 20/25–20/100.8

In our experience, there may be a synergistic effect between chemotherapy and EBRT to produce retinal and optic disc ischaemia, which may be evident within 2 or 3 years after completion of treatment, or it may be long delayed. We did not encounter this complication in this series despite a median follow-up of 40 months. However, we cannot exclude the possibility of late effects in these patients, and there is therefore a potential risk of visual loss due to late ischaemia in our subjects.

Possibly due to the relatively short follow-up period and the small number of patients, there are no second cancers or deaths reported in our study. For 82% of patients in this study, EBRT treatment began after age 12 months. Abramson and Frank5 demonstrated that the risk of second cancers in bilateral retinoblastoma is heavily dependent on the age at which EBRT is applied, and that there is no difference in the long-term tumour-free survival between those who did not receive EBRT and those who received it after age 12 months. This observation implies that the oncogenic risk is less when the therapy is delivered after age 1, and this may be important for future recommendations.

Based on the favourable results in this study, salvage EBRT in bilateral retinoblastoma should be considered in eyes with visual potential, when focal consolidation treatment is likely to damage vision, or when tumours are too large or numerous for further focal treatment.

Ultimately, the decision to give salvage radiotherapy rests with the ophthalmologists and the parents. The benefit of saving eyes with useful vision must be weighed against the possibility of a serious life-threatening complication. This study will help those involved reach an informed decision by providing quantitative information on the current effectiveness of this treatment.



  • Competing interests: None.

  • See Editorial, p 848

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles