Current management techniques for retinoblastoma have improved the survival rates in retinoblastoma; of the child, the globe, and the vision. Mortality is most commonly consequent to microscopic residual disease and adjuvant chemotherapy is an effective tool to eradicate presumed micrometastases before they clinically manifest and to thus reduce mortality. However, chemotherapy is not without side effects. Common side effects include anemia, thrombocytopenia, and neutropenia, potentially requiring hospitalization and transfusions. Other significant, but controversial side effects include a risk of secondary malignancies and ototoxicity. Although children with heritable retinoblastoma are at risk of secondary cancer due to the innate genetic predisposition, the risk is increased with the use of chemotherapy agents.
15 Malignancies observed at a mean of 11 years' follow-up were osteosarcoma, rhabdomyosarcoma, melanoma, glioma, and acute promyelocytic leukemia. There have also been reports of acute myelogenous leukemia.
16
It is thus important to identify the eyes at high risk of developing metastasis to reduce the risk of metastasis without the consequent side effects.
11 Although massive choroidal invasion, postlaminar optic nerve invasion, and scleral invasion are accepted histopathological HRFs purported to increase risk of metastasis,
17–19 anterior chamber seeding is a contentious histopathological finding. Determination of structures whose involvement constitutes AC seeds is also a gray area.
Figures 4C2a–d demonstrate invasion of various anterior chamber structures by the tumor cells.
Seeds in the anterior chamber, on corneal endothelium and iris surface and stroma are well accepted as AC seeds; however, involvement of trabecular meshwork and ciliary body is contentious. In an article by Baroni et al,
20 AC seeds were not considered an HRF and chemotherapy was administered only if concomitant HRFs were present. However, AC seeds alone were not analyzed independently. In an article by Khelfaoui et al,
7 AC seeds, although identified as a risk factor for metastasis in the multivariate analysis, did not remain risk factors once eyes with extra scleral or resection line involvement were excluded. Presence of AC seeds has been considered an HRF and treated with six cycles of chemotherapy in a few studies,
18 considered to be of intermediate risk for development of metastasis and treated with four cycles of chemotherapy in a few studies
13,21 and not considered a risk factor for metastasis and monitored for metastasis without chemotherapy in a few studies
15 (
Table 4). In most of the studies, AC seeds have not been independently analyzed for effect of adjuvant chemotherapy, and when analyzed, the number of eyes was too small to comment on the effect of chemotherapy.
In this study, we compared the survival rates of eyes with no HRFs, only AC seeds, only HRFs, and AC seeds with HRFs to demonstrate the good survival rate of eyes with AC seeds alone. A study by Kaliki et al
12 reported AC seeds in 36 (23%) of 157 eyes with anterior chamber seeding in 6% and iris infiltration in 3%. This was comparable to 16.9% eyes with AC seeds seen in our case series with iris infiltration being observed in 5.7% and anterior chamber seeds in 7.1%. In a study by Honavar et al.,
18 21% of cases showed AC tumor seeding, although only unilateral sporadic retinoblastomas were included in that series.
We observed a higher age (mean and median) at time of clinical diagnosis in those eyes that had only AC seeds. Children with sporadic unilateral retinoblastoma are known to present at an older age than those with bilateral retinoblastoma, and in this group, with the exception of one, all had unilateral retinoblastoma. In this group, there were two eyes that had the tumor in the anterior chamber only, with no tumor in the posterior chamber, representing diffuse anterior retinoblastoma, whereas one eye had the diagnosis of diffuse infiltrating retinoblastoma. This presentation of retinoblastoma is known to occur in a relatively older group of children. There was no significant difference in the distribution of other demographic variables among the groups. Presence of AC seeds alone was associated with a reduced duration of symptom to enucleation period as compared with when HRFs were also present, although the difference did not reach statistical significance. There was a statistically significant difference in the distribution of group D and group E tumors among the groups (P = 0.0015). The percentage of group E tumors was higher than group D tumors in all four groups. However, the difference was statistically significant in three groups: those with AC seeds, AC seeds with HRFs, and those with HRFs. Massive choroidal invasion and optic nerve invasion are appreciated on histopathology and cannot be determined clinically. They do not influence the clinical grading of the eyes directly. However, presence of AC seeds can be appreciated clinically as well as on HPE and their presence upgrades the retinoblastoma to grade E, thus explaining the higher percentage of grade E tumors in these eyes.
Survival rate of all the four groups did not show statistically significant difference, reflecting the overall good prognosis of retinoblastoma. Survival rate of eyes with no HRFs and no AC seeds was similar to that with eyes with only AC seeds representing lack of effect of AC seeds on mortality in these children. Survival rate of eyes with HRF and AC seeds was lower than that of eyes with only HRFs, although not reaching statistical significance. This could be due to asymmetrical distribution of number of HRFs. Although eight (44.4%) eyes with HRFs and AC seeds had at least three HRFs in the enucleated specimen, only nine (11.7%) eyes with only HRFs had three HRFs on HPE.
AC seeds may merely represent spillover of tumor activity into anterior chamber, especially when the original focus of retinoblastoma is anterior and the early involvement of anterior chamber is to be expected. Here anterior chamber involvement represents only the anterior location of the tumor and not the grade of the tumor or degree of anaplasia. Diffuse anterior retinoblastoma represents another entity in which the tumor involves only the anterior segment of the eye. These children are typically older (6 to 12 years of age) at diagnosis.
24 They are often sporadic, although a recent case report described a heritable form. Although Herwig et al.,
25 Longmuir et al.,
26 and Garner et al.
27 treated diffuse anterior retinoblastoma with enucleation followed by adjuvant chemotherapy and radiotherapy, Khetan et al.
28 treated with enucleation alone and did not report a recurrence at time of publication. A recent report by Shields et al.
29 described successful globe salvage in three such cases wherein only two of the three cases received systemic chemotherapy and none developed metastasis over a follow-up of 35 months.
Studies have shown that metastasis, after enucleation, usually occurs within 5 years.
2 A study had identified 6.4 months in unilateral cases and 14.2 months in bilateral cases as the median time to development of metastasis.
30 In a study by Shields et al.,
31 78.2% of recurrences occurred within the first year of diagnosis of retinoblastoma, whereas the median time to death was 10 months in that series. In our case series, 80 (37.7%) of the 212 eyes had a follow-up of 1 year, and in the group with AC seeds alone, 89.9% had a follow-up of more than a year and during this period, none developed metastasis.