Moyamoya following cranial irradiation for primary brain tumors in children.

OBJECTIVE: To study the risk factors for the development of moyamoya syndrome after cranial irradiation for primary brain tumors in children. METHODS: We reviewed neuroimaging studies and dosimetry data for 456 children who were treated with radiation for a primary brain tumor and who were prospectively evaluated with serial neuroimaging studies and neurologic evaluations. A total of 345 patients had both adequate neuroimaging and radiation dosimetry data for further analysis. We used survival analysis techniques to examine the relationship of clinically important variables as risk factors for the development of moyamoya over time. RESULTS: Overall, 12 patients (3.5%) developed evidence of moyamoya. The onset of moyamoya was more rapid for patients with neurofibromatosis type 1 (NF1) (median of 38 vs 55 months) and for patients who received >5,000 cGy of radiation (median of 42 vs 67 months). In a multiple Cox proportional hazards regression analysis controlling for age at start of radiation, each 100-cGy increase in radiation dose increased the rate of moyamoya by 7% (hazard ratio [HR] = 1.07, 95% CI: 1.02 to 1.13, p = 0.01) and the presence of NF1 increased the rate of moyamoya threefold (HR = 3.07, 95% CI: 0.90 to 10.46, p = 0.07). CONCLUSIONS: Moyamoya syndrome is a potentially serious complication of cranial irradiation in children, particularly for those patients with tumors in close proximity to the circle of Willis, such as optic pathway glioma. Patients who received higher doses of radiation to the circle of Willis and with neurofibromatosis type 1 have increased risk of the development of moyamoya syndrome.

Advantage of protons compared to conventional X-ray or IMRT in the treatment of a pediatric patient with medulloblastoma.

PURPOSE: To compare treatment plans from standard photon therapy to intensity modulated X-rays (IMRT) and protons for craniospinal axis irradiation and posterior fossa boost in a patient with medulloblastoma. METHODS: Proton planning was accomplished using an in-house 3D planning system. IMRT plans were developed using the KonRad treatment planning system with 6-MV photons. RESULTS: Substantial normal-tissue dose sparing was realized with IMRT and proton treatment of the posterior fossa and spinal column. For example, the dose to 90% of the cochlea was reduced from 101.2% of the prescribed posterior fossa boost dose from conventional X-rays to 33.4% and 2.4% from IMRT and protons, respectively. Dose to 50% of the heart volume was reduced from 72.2% for conventional X-rays to 29.5% for IMRT and 0.5% for protons. Long-term toxicity with emphasis on hearing and endocrine and cardiac function should be substantially improved secondary to nontarget tissue sparing achieved with protons. CONCLUSION: The present study clearly demonstrates the advantage of conformal radiation methods for the treatment of posterior fossa and spinal column in children with medulloblastoma, when compared to conventional X-rays. Of the two conformal treatment methods evaluated, protons were found to be superior to IMRT.

Stereotactic radiotherapy for pediatric intracranial germ cell tumors.

PURPOSE: Intracranial germ cell tumors are rare, radiosensitive tumors seen most commonly in the second and third decades of life. Radiotherapy alone has been the primary treatment modality for germinomas, and is used with chemotherapy for nongerminomatous tumors. Stereotactic radiotherapy techniques minimize the volume of surrounding normal tissue irradiated and, hence, the late radiation morbidity. This study reports our experience with stereotactic radiotherapy in this group of tumors. METHODS AND MATERIALS: Between December 1992 and December 1998, 18 patients with intracranial germ cell tumors were treated with stereotactic radiotherapy. A total of 23 histologically proven tumors were treated. Thirteen patients had a histologic diagnosis of germinoma, and 5 patients had germinoma with nongerminomatous elements. Of those patients with a histologic diagnosis of germinoma, 5 had multiple midline tumors. The median age of the patients was 12.9 years (range, 5.6-17.5 years). RESULTS: A boost using stereotactic radiotherapy was delivered to 19 tumors following whole-brain radiation in 8 cases and craniospinal radiation in 11 cases. Three tumors were treated with stereotactic radiotherapy to the tumor volume alone following chemotherapy, and 1 tumor received a boost using stereotactic radiosurgery following craniospinal radiation. A median dose of 2520 cGy (range, 1500-3600) cGy was given to the whole brain, and a median dose of 2160 (range, 2100-2600) cGy was given to the spinal field. The median boost dose to the tumor was 2600 (range, 2160-3600) cGy, given by stereotactic radiotherapy delivered to the 95% isodose line. At a median follow-up time of 40 (range, 12-73) months, no local or marginal recurrences were reported in patients with germinoma. Two patients with nongerminomatous tumors have relapsed. One had elevation of tumor markers only at 37 months following treatment, and the other had persistent disease following chemotherapy and radiation therapy. Eight patients documented pituitary-hypothalamic dysfunction; in 7 (87.5%) of these patients, the dysfunction was present before commencing radiotherapy. Four patients (22%) developed newly diagnosed diabetes insipidus following surgery. Three patients (17%) received antidepressant medication at follow-up. CONCLUSION: Our series shows that stereotactic radiotherapy is achievable and well tolerated in this group of patients. Longer follow-up is required to fully assess the impact on long-term toxicity. Psychologic assessment of mood and affect should be performed as part of routine follow-up in this group of adolescent children.

Excellent therapeutic efficacy and minimal late neurotoxicity in children treated with 18 grays of cranial radiation therapy for high-risk acute lymphoblastic leukemia: a 7-year follow-up study of the Dana-Farber Cancer Institute Consortium Protocol 87-01.

BACKGROUND: In the current study, the authors evaluated late neuropsychologic effects 7 years after diagnosis and the long-term survival in a cohort of patients treated for high-risk childhood acute lymphoblastic leukemia (ALL) with cranial radiation therapy. Efficacy and toxicity were evaluated in relation to patient age at diagnosis (age < or > or = 36 months). METHODS: Two hundred and one patients treated for high-risk ALL on the Dana-Farber Cancer Institute Consortium Protocol 87-01 were included, 147 of whom were in continuous complete disease remission and were eligible for cognitive testing. Sixty-one patients consented to undergo testing. All patients received 18 grays (Gy) of cranial radiation as a component of central nervous system treatment. RESULTS: For all 201 patients, the 5-year overall survival (% +/- the standard error) was 82% +/- 2 and the 5-year event-free survival (% +/- the standard error) was 75% +/- 3. Only two patients developed a central nervous system recurrence. Intelligence quotient (IQ) and memory were at the expected mean for age, but performance on a complex figure drawing task was found to be reduced. Children who were age < 36 months at the time of diagnosis were found to have an IQ in the average range, but showed verbal deficits. CONCLUSIONS: The results of the current study demonstrate excellent efficacy of therapy and relatively limited late neurotoxicity on a childhood ALL therapy protocol in which all evaluated patients had received 18 Gy of cranial radiation. Efficacious therapy that includes cranial radiation does not appear to necessarily incur a heightened risk for significant cognitive impairment.

Radiosurgery in the management of pediatric brain tumors.

OBJECTIVE: To describe the outcome of pediatric brain tumor patients following stereotactic radiosurgery (SRS), and factors associated with progression-free survival. METHODS: We reviewed the outcome of 90 children treated with SRS for recurrent (n = 62) or residual (n = 28) brain tumors over a 10-year period. Median follow-up from SRS was 24 months for all patients and 55.5 months for the 34 patients currently alive. RESULTS: The median progression-free survival (PFS) for all patients was 13 months. Median PFS according to tumor histology was medulloblastoma = 11 months, ependymoma = 8.5 months, glioblastoma and anaplastic astrocytoma = 12 months. Median PFS in patients treated to a single lesion was 15.4 months. No patient undergoing SRS to more than 1 lesion survived disease free beyond 2 years. After adjusting for histology and other clinical factors, SRS for tumor recurrence (RR = 2.49) and the presence of > 1 lesion (RR = 2.3) were associated with a significantly increased rate of progression (p < 0.05). Three-year actuarial local control (LC) was as follows: medulloblastoma = 57%, ependymoma = 29%, anaplastic astrocytoma/glioblastoma = 60%, other histologies = 56%. Nineteen patients with radionecrosis and progressive neurologic symptoms underwent reoperation after an interval of 0.6-62 months following SRS. Pathology revealed necrosis with no evidence of tumor in 9 of these cases. CONCLUSION: SRS can be given safely to selected children with brain tumors. SRS appears to reduce the proportion of first failures occurring locally and is associated with better outcome when given as a part of initial management. Some patients with unresectable relapsed disease can be salvaged with SRS. SRS to multiple lesions does not appear to be curative. Serious neurologic symptoms requiring reoperation is infrequently caused by radionecrosis alone.

Proton magnetic spectroscopic imaging of the child's brain: the response of tumors to treatment.

Our aim was to determine and/or predict response to treatment of brain tumors in children using proton magnetic resonance spectro-scopic imaging (MRSI). We studied 24 patients aged 10 months to 24 years, using MRI and point-resolved spectroscopy (PRESS; TR 2000 TE 65 ms) with volume preselection and phase-encoding in two dimensions on a 1.5 T imager. Multiple logistic regression was used to establish independent predictors of active tumor growth. Biologically vital cell metabolites, such as N-acetyl aspartate and choline-containing compounds (Cho), were significantly different between tumor and control tissues (P < 0.001). The eight brain tumors which responded to radiation or chemotherapy, exhibited lower Cho (P = 0.05), higher total creatine (tCr) (P = 0.02) and lower lactate and lipid (L) (P = 0.04) than 16 tumors which were not treated (except by surgery) or did not respond to treatment. The only significant independent predictor of active tumor growth was tCr (P < 0.01). We suggest that tCr is useful in assessing response of brain tumors to treatment.

Results of a prospective trial of mantle irradiation alone for selected patients with early-stage Hodgkin's disease.

PURPOSE: To determine the efficacy of mantle radiation therapy alone in selected patients with early-stage Hodgkin's disease. PATIENTS AND METHODS: Between October 1988 and June 2000, 87 selected patients with pathologic stage (PS) IA to IIA or clinical stage (CS) IA Hodgkin's disease were entered onto a single-arm prospective trial of treatment with mantle irradiation alone. Eighty-three of 87 patients had > or = 1 year of follow-up after completion of mantle irradiation and were included for analysis in this study. Thirty-seven patients had PS IA, 40 had PS IIA, and six had CS IA disease. Histologic distribution was as follows: nodular sclerosis (n = 64), lymphocyte predominant (n = 15), mixed cellularity (n = 3), and unclassified (n = 1). Median follow-up time was 61 months. RESULTS: The 5-year actuarial rates of freedom from treatment failure (FFTF) and overall survival were 86% and 100%, respectively. Eleven of 83 patients relapsed at a median time of 27 months. Nine of the 11 relapses contained at least a component below the diaphragm. All 11 patients who developed recurrent disease were alive without evidence of Hodgkin's disease at the time of last follow-up. The 5-year FFTF in the 43 stage I patients was 92% compared with 78% in the 40 stage II patients (P =.04). Significant differences in FFTF were not seen by histology (P =.26) or by European Organization for Research and Treatment of Cancer H-5F eligibility (P =.25). CONCLUSION: Mantle irradiation alone in selected patients with early-stage Hodgkin's disease is associated with disease control rates comparable to those seen with extended field irradiation. The FFTF is especially favorable among stage I patients.

Memory deficits among children with craniopharyngiomas.

OBJECTIVE: To describe neuropsychological functioning (with a specific focus on cognition and memory) after surgical treatment of craniopharyngiomas. METHODS: Sixteen patients who were between 6 and 15 years of age at the time of surgery comprised the sample. Each child had been treated for a craniopharyngioma with surgery only, on Dana-Farber Cancer Institute Protocol 92-077. RESULTS: The overall level of cognitive functioning was well within the average range, with both language and visuospatial functioning being generally intact; however, specific memory problems, in both the language and visuospatial domains, were evident. CONCLUSION: Although general cognitive functioning was intact after the surgical treatment of craniopharyngiomas, difficulties in the retrieval of learned information were observed. Neuropsychological assessments, with a focus on memory recall, should be a component of the medical management plan for each child.

Neurological dysfunction associated with postoperative cerebellar mutism.

BACKGROUND AND OBJECTIVES: The postoperative cerebellar mutism syndrome (CMS) is an unique acute postoperative complication characterized by transient decrease in speech output (often mutism), apathy, irritability as well as global cerebellar dysfunction. As much as 25% of patients undergoing a resection of a cerebellar or IV ventricular tumor may develop such a syndrome. In this retrospective study we characterize the clinical features of the CMS and explore potential etiologic mechanisms. METHODS: We conducted a retrospective analysis of medical records and imaging tests of 8 consecutive patients with the CMS identified through the database of the Children's Hospital and Dana-Farber Cancer Institute, Boston, and compared with a control group of 8 unaffected children undergoing a comparable tumor resection. RESULTS: In contrast to the control group, children in the affected group had marked decrease in speech output and comprehension, apathy and lack of initiative, inattention, persistent eye closure, flaccid hemiparesis and a severe global cerebellar dysfunction. Swallowing difficulties and bowel and bladder dysfunction were also observed. The median duration of the syndrome as judged by the persistence of the communication abnormalities was 4 weeks. The recovery was near complete with exception for a persistent global cerebellar dysfunction. A comparison of CT and MRI scans of children in both groups failed to identify distinguishing features. CONCLUSION: A surgical lesion of the midline cerebellum can cause a complex neurological dysfunction such as the CMS. Thus, we postulate that the cerebellum and its connections function as a 'modulatory system' in control of both motor and non-motor functions, including attention and language.

Adult medulloblastoma: prognostic factors and patterns of relapse.

OBJECTIVE: To determine the patterns of relapse and the prognostic factors for adult medulloblastomas treated in the magnetic resonance imaging era. METHODS: Between 1986 and 1996, 32 adult patients (age, > or =16 yr) with medulloblastomas confined to the craniospinal axis were treated in our institutions. Twenty cases involved classic histological features and 12 involved the desmoplastic variant. The Chang staging distribution was as follows: T1, 2; T2, 17; T3, 10; T4, 3; M0, 24; M1, 1; M2, 4; M3, 3. Brainstem invasion was present in nine patients. Lesions were midline in 13 cases and lateral in 19. Resection was complete in 17 cases, subtotal in 6, and partial in 5, with biopsy only in 4 cases. All patients received postoperative radiotherapy, with median doses of 36 Gy to the entire craniospinal axis and 55 Gy to the posterior fossa. Twenty-four patients received chemotherapy (20 before radiotherapy, 3 after radiotherapy, and 1 before and after radiotherapy). RESULTS: With a median follow-up period of 5.4 years, 17 patients experienced recurrences. At 5 and 8 years, overall survival rates were 83 and 45% and disease-free survival rates were 57 and 40%, respectively. The 5- and 8-year posterior fossa control rates were 67 and 59%, respectively. Twenty-nine percent of all relapses occurred more than 5 years after treatment. The posterior fossa was the most common site of relapses. In univariate analyses, factors adversely affecting posterior fossa control were less than complete resection (P<0.001), the presence of brainstem invasion (P = 0.02), and the use of chemotherapy (P = 0.03). The overall radiotherapy duration was marginally significant in predicting posterior fossa control, with 5-year posterior fossa control rates of 81 and 49% for durations of less than 48 days and 48 days or more, respectively (P = 0.06). In a multivariate analysis, complete resection was predictive of improved posterior fossa control (P = 0.02) and disease-free survival (P = 0.02) rates. Of the eight low-risk patients who received radiotherapy alone, three experienced recurrences in the bone as the only site of relapse. CONCLUSION: Late relapse is common among adult patients with medulloblastomas, and long-term follow-up monitoring is important. Because of the high risk of systemic failure among the low-risk patients treated with radiotherapy alone, the role of chemotherapy for this group of patients needs to be further investigated. Complete resection, the absence of brainstem invasion, and an overall radiotherapy duration of less than 48 days are important prognostic factors.

Cognitive sequelae in children treated for acute lymphoblastic leukemia with dexamethasone or prednisone.

PURPOSE: The cognitive sequelae of treatment for childhood acute lymphoblastic leukemia (ALL) were compared in a group of patients who received dexamethasone during the intensification and maintenance phases of therapy with those in a historical control group for whom antileukemia therapy was similar, except that the corticosteroid component of therapy was prednisone. METHODS: Patients treated for ALL on Dana-Farber Cancer Institute protocols 87-01 (n = 44) and 91-01 (n = 23) were evaluated by standard cognitive and achievement tests. Corticosteroid therapy was delivered in 5-day pulses given every 3 weeks during intensification and continuation phases of therapy for a total of 2 years. RESULTS: Children treated on protocol 87-01 received prednisone at a dose of 40 mg/m2/d (standard risk, SR) or 120 mg/ m2/d (high risk, HR); those treated on protocol 91-01 received dexamethasone at a dose of 6 mg/m2 per day (SR) or 18 mg/m2 per day (HR). Children treated on protocol 91-01 performed less well on cognitive testing. Subsample analysis indicated that cranial radiation therapy and methotrexate dose did not account for differences in cognitive outcomes. CONCLUSIONS: The findings of this preliminary study are consistent with the hypothesis that dexamethasone therapy can increase risk for neurocognitive late effects in children treated for ALL and indicate that further investigation of this question is warranted.

Ewing sarcoma of the rib: results of an intergroup study with analysis of outcome by timing of resection.

OBJECTIVE: We sought to establish the outcome and optimal therapeutic sequence for patients with nonmetastatic Ewing sarcoma/primitive neuroectodermal tumor of the chest wall. METHODS: Patients 30 years of age or younger with nonmetastatic Ewing sarcoma/primitive neuroectodermal tumor of the bone were randomly assigned to receive vincristine, doxorubicin, cyclophosphamide, and dactinomycin or those drugs alternating with ifosfamide and etoposide. Local control was obtained with an operation, radiotherapy, or both. RESULTS: Fifty-three (13.4%) of 393 patients had primary tumors of the chest wall (all rib). Event-free survival at 5 years was 57% for the chest wall compared with 61% for other sites (P >.2). Ifosfamide and etoposide improved outcome in the overall group (5-year event-free survival, 68% vs 54%; P =.002), and a similar trend occurred in chest wall lesions (5-year event-free survival, 64% vs 51%). Patients with chest wall lesions had more attempts at initial surgical resection (30%) than those with other primary tumor sites (8%, P <.01). The attempt at initial resection for chest wall lesions did not correlate with size. Initial resections at other sites were restricted to smaller tumors. Initial resection resulted in negative pathologic margins in 6 of 16 patients, whereas the delayed resection resulted in negative margins in 17 of 24 patients (P =.05). Although there was no difference in survival by timing of the operation in rib lesions, a higher percentage of patients with initial surgical resection received radiation than those with resection after initial chemotherapy (P =. 13). CONCLUSIONS: Although rib primary tumors are significantly larger than tumors found in other sites, their outcome is similar. We favor delayed resection whenever possible to minimize the number of patients requiring radiation therapy.

Heterozygous germline ATM mutations do not contribute to radiation-associated malignancies after Hodgkin's disease.

PURPOSE: The successful treatment of Hodgkin's disease has been associated with an increased incidence of secondary malignancies. To investigate whether genetic factors contribute to the development of secondary tumors, we collected family cancer histories and performed mutational analysis of the ataxia-telangiectasia (AT) gene, ATM, in a cohort of Hodgkin's disease survivors with secondary malignancies. ATM was chosen for evaluation because of the increased radiosensitivity of cells derived from AT patients and obligate heterozygotes and the epidemiologic observation that AT carriers are at increased risk for radiation-induced breast cancer. PATIENTS AND METHODS: Fifty-two patients who developed one or more neoplasms after treatment for Hodgkin's disease participated in this study. Personal and family histories of cancer were obtained through patient interviews and review of medical records. ATM mutational analysis was performed using a yeast-based protein truncation assay. RESULTS: Seventy-six secondary neoplasms were observed in this cohort of 52 Hodgkin's disease survivors, with 18 patients (35%) developing more than one secondary neoplasm. Positive family histories of cancer were present in 11 (21%) of 52 patients, compared with three (4%) of 68 Hodgkin's disease patients in a comparison cohort who did not develop secondary neoplasms (P =.008; Fisher's exact test). No germline ATM mutations were identified, resulting in an estimated AT carrier frequency in this population of 0% (90% confidence interval, 0% to 4%). CONCLUSION: Analysis of the number of tumors per individual and the family history of cancer in our cohort suggests that genetic factors may contribute to development of secondary neoplasms in a subset of Hodgkin's disease survivors. Mutational analysis, however, does not support a significant role for heterozygous truncating ATM mutations. Future studies evaluating other genes involved in DNA damage response pathways are warranted.

Cervicomedullary astrocytomas of childhood: clinical and imaging follow-up.

UNLABELLED: BACKGROUND. Cervicomedullary astrocytomas are a unique subset of brainstem tumors in children because they have a good prognosis when compared to the pontine subset of brainstem gliomas. Objective. To review the clinical and imaging findings in a series of children with cervicomedullary astrocytomas as to diagnosis and management. MATERIALS AND METHODS: A retrospective review of eleven children (six females, five males, age range: 10 days-18 years; mean = 7 years) with cervicomedullary tumors was done including the clinical presentation, imaging studies (MR: eleven, CT and MR: four), surgical findings, pathological results, and follow-up clinical and imaging findings (range: 0.2-11 years; mean = 5.2 years). RESULTS: Symptoms and signs were delayed and protracted, often occurring over months to years (mean = 2.3 years, range 0.5-7 years). The tumors expanded the dorsal medulla and involved the upper cervical spinal cord (mean maximum tumor diameter = 4.4 cm). Only three patients had hydrocephalus. In three of four cases the tumor was not seen on CT. On MR, the majority of the tumors were T1 hypointense and T2 hyperintense. Treatment consisted of surgery only in six patients, surgery and radiation therapy in four, and surgery, chemotherapy, and radiation in one. There was recurrent local disease in four patients and on follow-up metastatic disease in the brain in one. On follow-up the majority of the patients are alive and stable (mean = 5.2 years, range 0.2-11 years). There has been one death. The majority of tumors were pilocytic astrocytomas. CONCLUSION: Cervicomedullary tumors are a unique subset of brainstem gliomas in childhood that present with a long duration of symptoms and a greater long-term survival than pontine gliomas.

Activation of neurotrophin-3 receptor TrkC induces apoptosis in medulloblastomas.

Elevated expression of the neurotrophin-3 (NT-3) receptor TrkC by childhood medulloblastomas is associated with favorable clinical outcome. Here, we provide evidence that TrkC is more than simply a passive marker of prognosis. We demonstrate that: (a) medulloblastomas undergo apoptosis in vitro when grown in the presence of NT-3; (b) overexpression of TrkC inhibits the growth of intracerebral xenografts of a medulloblastoma cell line in nude mice; and (c) trkC expression by individual tumor cells is highly correlated with apoptosis within primary medulloblastoma biopsy specimens. TrkC-mediated NT-3 signaling promotes apoptosis by activating multiple parallel signaling pathways and by inducing immediate-early gene expression of both c-jun and c-fos. Considered collectively, these results support the conclusion that the biological actions of TrkC activation affect medulloblastoma outcome by inhibiting tumor growth through the promotion of apoptosis.

Low-dose total body irradiation and G-CSF without hematopoietic stem cell support in the treatment of relapsed or refractory acute myelogenous leukemia (AML), or AML in second or subsequent remission.

PURPOSE: Patients with relapsed acute myelogenous leukemia (AML), who are not eligible for bone marrow transplantation, have a poor prognosis when treated with chemotherapy alone. Total body irradiation (TBI) is an effective modality against AML when used in doses of 1000-1400 cGy with hematopoietic stem cell support. We undertook a phase I study of TBI with granulocyte-colony-stimulating factor (G-CSF) support, without stem cell support in patients with AML either in relapse or second or subsequent remission. METHODS AND MATERIALS: Patients with relapsed AML, or AML in second or subsequent remission were treated in a phase I study of TBI followed by G-CSF. The first dose level was 200 cGy. After the initial cohort of patients it was clear that patients with overt leukemia did not benefit from this treatment, and subsequent patients were required to be in remission at the time of TBI. RESULTS: Eleven patients were treated, 4 in overt relapse, and 7 in remission. 200 cGy was used in all, and dose escalation was not possible due to prolonged thrombocytopenia in all patients but one. Neutrophil recovery was adequate in those patients who remained in remission after TBI. Patients with overt leukemia had transient reduction in blast counts, but rapid recurrence of their leukemia. Patients treated in remission had short remissions, with the exception of one patient who is in remission 32 months after treatment. CONCLUSION: There is some antileukemic effect of TBI even at 200 cGy, though this dose appears to be too low to help a significant number of patients. If TBI is to be escalated without stem cell support, then a thrombopoietic agent will need to be used.

A stereotactic radiation therapy device for retinoblastoma using a noncircular collimator and intensity filter.

The proximity of the lens to the retina makes the treatment of retinoblastoma a challenge for external beam radiation therapy. The approximately 1 mm separation between the posterior edge of the lens and the anterior region of the retina causes a trade-off between coverage of the entire retina and excessive dose to the lens. A stereotactic, LINAC based, lens sparing technique for treating retinoblastoma is presented. The technique uses noncoplanar arcs with the lens at isocenter. A special noncircular collimator blocks the lens but it also causes the dose distribution to vary across the retina. A fluence modulation filter is used to reduce the dose inhomogeneity across the target. The resulting dose distribution is roughly hemispheric, providing both anterior coverage of the retina and lens blocking unlike conventional techniques. The method used to develop the collimator and filter assembly is presented. Dosimetry of the assembly was carried out using radiochromic film, and the results were entered in a treatment planning system. The dose distribution as measured in a phantom is provided and compared to calculations.

Second malignancies in patients treated for childhood acute lymphoblastic leukemia.

PURPOSE: Second malignant neoplasms (SMN) are devastating late complications of childhood acute lymphoblastic leukemia (ALL) and its treatment. We evaluated the incidence and type of SMN diagnosed before leukemic relapse in a large series of patients with ALL. PATIENTS AND METHODS: We reviewed the outcome of all patients treated for childhood ALL between 1972 and 1995 on Dana-Farber Cancer Institute (DFCI) and DFCI ALL Consortium protocols. The follow-up time from diagnosis of ALL to induction failure, relapse, remission death, or SMN, whichever occurred first, ranged from 0 to 24.0 years (median, 7.6 years; mean, 6.7 years). RESULTS: Thirteen SMNs were diagnosed among 1,597 patients. Eight tumors occurred in a radiation field (five in the CNS and three in the head and neck), two occurred outside of a radiation field (one adenocarcinoma of the sigmoid colon and one epithelioid sarcoma of the chest wall), and three were hematopoietic malignancies. The median time to occurrence was 6.7 years (range, 1.0 to 17.2 years) and the cumulative incidence of second malignancy before another first event was 2.7% (95% confidence interval, 0.7 to 4.7). The risk of a first event, which included induction failure, relapse, or remission death, was 31.0% (95% confidence interval, 28.5 to 33.5). CONCLUSION: We found a more than 10-fold risk of other first events when compared with SMN. Thus, we conclude that SMN before first relapse is a relatively uncommon occurrence among survivors of childhood ALL. Future therapeutic regimens must focus on reducing leukemia relapse and enhancing quality of life, as well as preventing SMNs.