4 I DISCUSSION
Our study describes the incidence of 4 ocular complications - cataract, dry eyes, retinopathy, and glaucoma, in early follow-up after cranial proton radiation. It also provides a snapshot of visual function for childhood brain tumor survivors after completion of treatment that includes proton radiation. For patients who receive whole brain radiation, as a part of CSI, the incidence of bilateral cataracts is significant, even with limited follow-up. Nearly two-thirds of patients in our cohort who received this volume of radiation, developed cataracts. In addition, nearly a quarter of these patients developed dry eyes during or shortly after CSI. Multivariable analysis identified CSI radiation volume as a significant risk factor for the development of cataract and ocular complications overall (cataract and dry eyes). In contrast, patients who did not receive whole-brain radiation, did not have ocular complications that could be attributed to radiation. Only 1 patient who received IFRT for craniopharyngioma, developed retinal scarring and bilateral cataracts, but had to be excluded due to having received both photon and proton radiation. No other patient was found to have retinal scarring, and none had glaucoma.
Infratentorial tumor location and the use of radiosensitizer chemotherapy were also found to be significant risk factors for the development of cataract and ocular complications overall. However, all patients with these 2 variables, who developed cataract or dry eyes, also received CSI radiation volume, making them likely confounding factors, although this could not be tested statistically. For whole brain radiation (CSI), the proton beam enters through the posterior aspect of the brain and ends at the eyes, to ensure adequate coverage of the planning target volume. Hence the lenses of the eyes receive some radiation. The relative biological effectiveness (RBE) for protons is postulated to be 1.1 compared to photons. However, depending on the area relative to the Bragg peak, it is possible that the RBE might be higher, with an increased linear energy transfer (LET), and potentially higher doses to tissues in the vicinity of the target volume.20 Our finding of significantly increased risk of cataracts with whole brain radiation, is consistent with the literature. A study of long-term survivors in the Childhood Cancer Survivor Study (CCSS), reported an increased risk for cataracts with radiation dose to the lens greater than, as little as 200 cGy, with a relative risk of 3.2, compared to their siblings.9Steroid treatment has been described to be significantly associated with an increased risk for cataracts in leukemia survivors who received TBI, while other literature has shown a lack of significant risk in the absence of CNS irradiation.11,21 Busulfan in the setting of conditioning for bone marrow transplant, with TBI, has also been reported to increase the risk for cataracts.14,22None of the patients in our cohort received Busulfan. However, all patients in our cohort who received CSI for embryonal tumors, also received dexamethasone as an anti-emetic prior to chemotherapy, making it a confounding factor.
At last follow-up, after completion of all treatment, nearly three in four patients in our cohort had a visual acuity in their best eye, better than 20/60. Only 3 patients had VA in both eyes, worse than 20/100. In the Childhood Cancer Survivor Study (CCSS), any radiation dose to the posterior fossa, more than 3000 cGy to the temporal lobe and more than 500 cGy to the lenses, were associated with increased risk of legal blindness at a median of 1 year from cancer diagnosis.9 However, the patients in the CCSS cohort received radiation prior to the advent of more targeted techniques like intensity modulated radiation therapy (IMRT), and the use of protons. That study also had a much larger number of patients, with more heterogeneous diagnoses and longer follow-up.
Our study had several limitations. Our study was retrospective and relied on data recorded in medical records, with missing or limited details. All patients received radiation and oncology care at a single institution, limiting the generalizability of our findings. Most patients in our cohort, did not have ophthalmology exam data from prior to radiation. Most patients’ ocular complications were detected at their first formal eye exam post-radiation. Hence the true time to develop these complications could not be accurately determined, and they were presumed to have occurred after radiation. While our study provides a glimpse of ocular function in childhood brain tumor survivors after treatment, the absence of pre-treatment visual acuity data for most patients precluded the analysis of the change of visual function with treatment. Very few patients had eye exam data during or shortly after radiation, limiting the analysis of the true incidence of dry eyes, as an acute adverse effect of radiation. Parsons et al. reported a significant increase in the risk of severe dry eyes at ocular radiation doses higher than 40 Gy.23,24 In the CCSS study, the median time to the development of dry eyes was 7.2 years, with a cumulative incidence of over 5% for survivors of primary CNS malignancies.9 The long-term incidence of dry eyes after protons could not be elucidated in our study due to the limited duration of follow-up.
Nearly half the patients in our cohort of consecutive patients who received protons for brain tumors, did not have ophthalmology data available, which creates the potential for inclusion bias. Our statistical analysis, however, suggests the excluded patients were similar to study patients in gender, age at radiation, tumor location and volume of radiation. The ophthalmology follow-up for our cohort was relatively short, and hence our results do not provide insight into the incidence of more delayed ocular complications after radiation. In the CCSS cohort, patients who received more than 500 cGy to the lens, the incidence of cataracts continued to increase up to 20 years after diagnosis.9 The lack of long-term follow-up also limited our analysis of the functional outcomes for patients who developed ocular complications.
Our study findings highlight the need for larger prospective, multi-institutional pediatric CNS tumor trials, to include regular, formal ophthalmology exams prior to and after radiation, with longer follow-up, to determine the true incidence of these complications and their impact on visual function for these children. This would help inform ophthalmology screening guidelines for these patients, and also radiation planning to mitigate the risk of these adverse effects when possible.