The argument for a combined cancer consortium research data base

Month-of-birth and incidence of acute lymphoblastic leukemia in children

As a means of examining the virus-relatedness of acute lymphoblastic leukemia (ALL) in children, we investigated the association between month-of-birth and the occurrence of ALL in 1487 children aged 0-15 years at the time of diagnosis. Our hypothesis being that evidence of seasonal variation in births of ALL cases would suggest exposure to a transmissible etiologic agent during the perinatal period. The data were obtained from the Surveillance, Epidemiology, and End Results (SEER) Program and consisted of children diagnosed during the years 1973-1986. Aggregate monthly incidence rates of ALL stratified by month-of-birth, for each SEER site, all sites combined, and for broad geographic regions were calculated. No evidence for an association between month-of-birth and childhood ALL was found.

Role for proton beam irradiation in treatment of pediatric CNS malignancies

The ability to vary the proton energy (depth of beam penetration) and modulate the dose distribution at the end of range permits delivery of an increased dose to the designated cancer-containing volume with a reduced dose to overlying normal brain tissue. The evolution of childhood CNS malignancy following therapy is reviewed to identify radiation response variables indicating where the proton dose distribution will improve the therapeutic ratio. The review documents that of the 1262 children expected to develop CNS malignancy in 1989, only 43% will survive 5 years. About 75% of those with medulloblastoma and over 90% with astrocytoma die from persistent (in-field) disease. When the patient has been treated with radiation, it is accepted that disease persistence indicates the cancer dose was insufficient. Potentially 536 children could show an improved incidence of local control and improved survival from an increased cancer dose available from proton irradiation. As the total dose and volume of brain irradiated is increased about 1800 cGy, brain dysfunction increases, producing a spectrum of functional and intellectual deficits which are age and volume related. About 900 irradiated patients would have fewer in-field histologic and functional changes if the dose to normal brain, or the volume of brain irradiated, is reduced by an improved dose distribution. A proton beam treatment plan, delivering a cancer dose of 7400 cGy, is simulated for a thalamic astrocytoma. The dose distribution of this plan is compared with an x-ray plan used to treat a patient, in which a dose of 5400 cGy was delivered to the astrocytoma. Comparative isodose distributions and dose-volume histograms indicate a decreased integral dose to normal brain and a decreased volume of normal brain irradiated, even as the cancer dose is boosted 2000 cGy with protons.

Systematic underreporting of cutaneous malignant melanoma in Massachusetts. Possible implications for national incidence figures

An independent tabulation of incidence of cutaneous malignant melanoma in Massachusetts indicates that 12% and perhaps as many as 19% of new cases of cutaneous malignant melanoma in Massachusetts are not recorded in the Massachusetts Cancer Registry, significantly more than the expected 5% (p = 0.0001). The increasing number of nonhospital medical settings in which melanomas can be diagnosed and/or treated appears to account for this discrepancy. We suspect that these findings in Massachusetts also apply to cancer reporting systems in other regions of the United States. We suggest that the true incidence of cutaneous malignant melanoma in Massachusetts, and perhaps in the United States, may be significantly higher than reported.

Use of the SEER Cancer Registry for technology assessment

The Surveillance, Epidemiology, and End Results (SEER) cancer registry contributed to technology assessment by providing population-based samples for detailed case-control studies, by serving as the control group in comparisons with various experimental groups, by allowing an assessment of selection bias in clinical trials, and by facilitating evaluations of classification and coding systems.

A study of childhood brain tumors based on surgical biopsies from ten North American institutions: sample description. Childhood Brain Tumor Consortium

The Childhood Brain Tumor Consortium has collected an extensive amount of neuropathologic and clinical information under rigorously controlled conditions about 3,291 children with brain tumors. In this overview of the entire sample, five observations are prominent: 1) many tumors involve more anatomic sites at the time of the first surgical exploration than previously recognized; 2) one-third of infratentorial tumors involve both the brainstem and the cerebellum; 3) the spinal compartment is involved primarily, or in combination with the posterior fossa, in 11% of childhood brain tumors; 4) 43.2% of childhood brain tumors are limited to the posterior fossa; 5) only a few World Health Organization diagnoses account for most brain tumors in children, and 6) there is a male predominance over all ages for infratentorial tumors. Subsequent reports will describe observer variability of participating neuropathologists, correlates of clinical and histologic information, the search for homogeneous subtypes of tumors, and prognostic factors.

Surgical staging of cerebellar astrocytomas in childhood

Benign cerebellar astrocytomas of childhood, when removed grossly in toto, have high cure rates and thus can be staged by their surgical resectability. Currently, total tumor resection is limited by attachment to or invasion of the tumor into the brain stem or by evidence of recurrence or metastases. Tumors of the cyst-nodule form are more often amenable to total gross removal because of their discrete margins and lateral placement, which is clear of brain stem structures. However, solid astrocytomas of the cerebellum, with or without visible cysts, may also be surgically cured if there is sufficient clearance from the brain stem. Patients with malignant astrocytomas of the cerebellum or metastases have much poorer prognoses. Preoperative and postoperative staging systems have been suggested based on these assumptions.