Chemoreduction improves eye retention in patients with retinoblastoma: a report from the German Retinoblastoma Reference Centre.

BACKGROUND: Retinoblastoma is the most common intraocular childhood tumour. Although mortality is low in Western countries, long-term sequelae, including secondary tumours, compromised vision or loss of one or both eyes are common. Chemoreduction combined with focal treatment is currently the leading conservative treatment for retinoblastoma, with success rates of 50-75% reported. We assessed a new chemoreduction protocol using intravenous cyclophosphamide with reduced dose of carboplatin on eye retention in patients with retinoblastoma. PROCEDURE: The 40 patients with retinoblastomas in 56 eyes were treated between 1995 and 2004 at the German Retinoblastoma Reference Centre Essen. The 6-cycle chemotherapy used vincristine (days 1, 22, 43, 64, 85, 106), etoposide (days 22, 43, 85, 106), carboplatin (days 1, 43, 64, 106), and cyclophosphamide (days 1, 22, 64, 85). Mean follow-up was 101 months. Most patients received additional hyperthermia, some received local treatment with laser coagulation, cryotherapy and/or ß-ray brachytherapy. Therapy failure was defined as progression requiring enucleation or external beam radiotherapy (EBRT). RESULTS: Primary chemotherapy was successful in 42 of 56 eyes (75%). Therapy success and visual acuity at age 6 years correlated with the International Classification of Retinoblastoma (ICRB) group. Age at diagnosis (> or <6 months) correlated with relapse, but not with therapy failure or visual acuity at 6 years of age. ICRB group did not correlate with occurrence of relapse. CONCLUSIONS: In this retrospective single-centre study, chemoreduction, including cyclophosphamide, with or without focal treatment, effectively controlled retinoblastoma progression without requiring enucleation or EBRT. Addition of cyclophosphamide is safe, and allows reduction of carboplatin.

Beta-ray brachytherapy with 106Ru plaques for retinoblastoma.

PURPOSE: A retrospective analysis of 134 patients who received (106)Ru brachytherapy for retinoblastomas (175 tumors in 140 eyes). Treatment and follow-up were analyzed with special emphasis on tumor control organ, preservation, and late complications. RESULTS: Treated tumors had a mean height and diameter of 3.7+/-1.4 mm and 5.0+/-2.8 disk diameters, respectively. The radiation dose values were recalculated according to the calibration standard recently introduced by the National Institute of Standards and Technology. The recalculation revealed a mean applied dose of 419 Gy at the sclera (SD, 207 Gy) and 138 Gy (SD, 67 Gy) at the tumor apex. The 5-year tumor control rate was 94.4%. Tumor recurrence was more frequent in eyes with vitreous tumor cell seeding or fish-flesh regression. The estimated 5-year eye preservation rate was 86.5%. Previous treatment by brachytherapy or external beam radiotherapy, as well as a large tumor diameter, were significant factors for enucleation. The radiotherapy-induced complications after 5 years of follow-up were retinopathy (22%), optic neuropathy (21%), and cataract (17%). These complications were significantly more frequent after prior brachytherapy or external beam radiotherapy. CONCLUSION: Brachytherapy using (106)Ru plaques is a highly efficient therapy with excellent local tumor control and an acceptable incidence of side effects.

De novo intraocular retinoblastoma development after chemotherapy in patients with hereditary retinoblastoma.

OBJECTIVE: Identification of incidence and risk factors for recurrence of de novo retinoblastomas after chemotherapy treatment in patients with hereditary retinoblastoma. METHODS: A retrospective, case-control study of 32 patients (50 eyes) with sporadic or familial bilateral retinoblastomas was conducted. Patients received a systemic chemotherapy regimen applying three courses of a combination of three drugs (including vincristine, etoposide, carboplatin, or cyclophosphamide) followed by additional local therapy. The primary outcome analyzed was the development of retinoblastomas, probably arising as the cause of a new mutational event (de novo) after completion of chemotherapy treatment. RESULTS: Patients were treated with an average of 5.8 +/- 1.8 chemotherapy courses (4.6 +/- 2.4-year follow-up time). Development of de novo tumors occurred in 48% of the treated eyes. These tumors occurred during chemotherapy treatment or within 7 months of chemotherapy completion. No de novo tumors developed in patients older than 3.2 years. Children who developed de novo tumors were significantly younger at the time of diagnosis (6.7 +/- 6.3 months vs 14.4 +/- 11.4 months, P < 0.001), and had a significantly lower number of tumors per eye at treatment begin (2.6 +/- 2.3 tumors vs 4.3 +/- 6.4 tumors, P < 0.001). The difference of the total numbers of retinoblastomas that developed per eye between the patients that developed de novo retinoblastomas during or after chemotherapy and patients who did not was not statistically significant (4.9 +/- 2.7 and 4.3 +/- 6.4, respectively, P = 0.8). No eye was lost because of de novo retinoblastoma development, and 92% of the eyes were preserved. CONCLUSIONS: De novo retinoblastomas developed both during and after completion of chemotherapy treatment. Younger children were at a significantly higher risk for developing de novo intraocular retinoblastomas. Good tumor control and eye preservation rates were achieved with regular and frequent control examinations in addition to the immediate treatment of de novo retinoblastomas.

Beta-ray brachytherapy of retinoblastoma: feasibility of a new small-sized ruthenium-106 plaque.

UNLABELLED: A non-comparative case observation study estimated the feasibility of brachytherapy for retinoblastoma with a newly designed ruthenium-106 plaque (label: CXS) with an 8-mm diameter of the irradiation zone. METHODS: The new CXS plaque was used between 2001 and 2003 for brachytherapy of 13 retinoblastomas. Indications were recurrences after preceding local treatment or endophytic retinoblastoma with an impending vitreous tumour cell seeding. The prescribed radiation dose at the apex was 88 Gy (NIST-calibrated dosimetry). RESULTS: The mean age at brachytherapy was 1.2 years (standard deviation, SD: 1.1 years), and the mean follow-up was 1.7 years (SD: 0.6 years). The treated tumours had a mean diameter of 2.3 mm (SD: 0.7 mm) and a mean height of 1.5 mm (SD: 0.6 mm) with a mean distance to the optic disc of 9.9 mm (SD: 2.2 mm). The mean duration of irradiation was 29.3 h (SD: 9.9 h) with a mean dose at the sclera of 213 Gy (SD: 80 Gy). Surgery was uneventful in all cases. Complete regression developed after 3.1 months (SD: 2.8 months) in all cases without a recurrence or a progression of the vitreous tumour cell seeding. The eyes developed no further side-effects besides a temporary circumscribed intra-ocular haemorrhage that emerged from the regressive tumour remnants. CONCLUSION: Brachytherapy with the CXS plaque seems to be a safe and reliable treatment option for small-sized retinoblastoma when laser or cryocoagulation failed to control the tumour growth or for small retinoblastoma with an incipient local tumour cell seeding on the tumour surface.

T1 Gd-enhanced compared with CISS sequences in retinoblastoma: superiority of T1 sequences in evaluation of tumour extension.

BACKGROUND: As adequate therapy for retinoblastoma in young children depends on infiltration of extra-retinal structures, diagnostic modalities play an essential role. METHODS: In this widely extended study, 80 children with retinoblastoma were studied with MRI (standard fat-suppressed Gd-enhanced T1, T2 thin-slice sequences (additionally with small loop surface coil), constructive interference in steady state (CISS) sequence covering the orbita). The images were analysed by two blinded neuroradiologists. Histology was used as the gold standard. RESULTS: MRI assumed infiltration of extra-retinal structures in 13 of 80 patients of which ten were confirmed by histology. Affected extra-retinal structures were: optic nerve (five, of which two were on CISS and three on T1 with higher image resolution using the surface coil), scleral infiltration (five, of which four on CISS and T1) and ciliary body infiltration (one on CISS and T1). Another 61 enucleated patients did not have any extra-retinal infiltration in histology. The CISS sequence with multiplanar reconstruction was mainly helpful in revealing exact three-dimensional tumour extension with excellent clinical acceptance and pre-surgical planning but T1 fat-suppressed Gd-enhanced images were superior in revealing exact tumour extension. CONCLUSION: CISS sequences allow to produce excellent anatomical images and to perform multiplanar reconstruction to better demonstrate tumour extension. However, T1-weighted sequences after contrast application are more sensitive (60 versus 40%) in detecting infiltration of the optic nerve but equal in detecting scleral infiltration.