The Long Island Breast Cancer Study Project is a federally mandated, population-based case-control study to determine whether breast cancer risk among women in the counties of Nassau and Suffolk, NY, is associated with selected environmental exposures, assessed by blood samples, self-reports, and environmental home samples. This report describes the collaborative project's background, rationale, methods, participation rates, and distributions of known risk factors for breast cancer by case-control status, by blood donation, and by availability of environmental home samples. Interview response rates among eligible cases and controls were 82.1% (n = 1,508) and 62.8% (n = 1,556), respectively. Among case and control respondents who completed the interviewer-administered questionnaire, 98.2 and 97.6% self-completed the food frequency questionnaire; 73.0 and 73.3% donated a blood sample; and 93.0 and 83.3% donated a urine sample. Among a random sample of case and control respondents who are long-term residents, samples of dust (83.6 and 83.0%); soil (93.5 and 89.7%); and water (94.3 and 93.9%) were collected. Established risk factors for breast cancer that were found to increase risk among Long Island women include lower parity, late age at first birth, little or no breast feeding, and family history of breast cancer. Factors that were found to be associated with a decreased likelihood that a respondent would donate blood include increasing age and past smoking; factors associated with an increased probability include white or other race, alcohol use, ever breastfed, ever use of hormone replacement therapy, ever use of oral contraceptives, and ever had a mammogram. Long-term residents (defined as 15+ years in the interview home) with environmental home samples did not differ from other long-term residents, although there were a number of differences in risk factor distributions between long-term residents and other participants, as anticipated.

During a period of 10 years, 51 patients were found to have metastatic cancer in the breast. There were 44 women and seven men. Eighteen patients had carcinoma, 16 had malignant lymphoma, 14 had malignant melanoma, and three had myosarcomas. I t is interesting that 16 of the 51 patients had no prior history of malignant disease-the mammary lesion presenting as the first manifestation of an occult extramammary primary. Metastatic cancer should be suspected when a multinodular neoplasm is found in the superficial tissues of the breast. Forty-one patients had a rapidly fulminating course and died of disease. Ten patients are alive, seven with and three without apparent evidence of disease. It is concluded that earlier recognition of these tumors may lead to more rational therapy and avoidance of unnecessary radical surgery. LTHOUGH PRIMARY MALIGNANT TUMORS OFA the breast are common, tumors metastatic to the breast are rare.2,6.11 The mammary manifestation of extramammary primary neoplasms-known or unknown-should be clinically and histologically distinguished from primary mammary neoplasms.6~7.DJ0The few reported cases in the literature indicate not only the rarity of metastatic mammary tumors but show the limited experience with this condition in any given institution.

Three hundred and three patients with benign solitary neurilemomas, or schwannomas, not associated with the classical type of von Recklinghausen's disease were studied. One hundred and seventy‐two were women and 131 were men. Forty‐five percent of the tumors were seen in the extremities, and no specific area could be related to the development of these tumors. Approximately 9% of the tumors occurred in the soft tissues of the trunk and 14% in various unusual sites. Most patients were seen initially because of a painless mass; only 10 complained of pain radiating along the course of a peripheral nerve. One documented case of a malignant degeneration in a benign schwannoma has been reported. The treatment of a benign solitary neurilemoma is enucleation. Excision of a segment of a major peripheral nerve is contraindicated. The histologic features of benign peripheral nerve tumors were briefly described. The confusion regarding terminology has been discussed, and it is suggested that the term schwannoma be more frequently used. The lack of neurologic deficit in schwannomas arising in major peripheral nerves has been pointed out, and further studies have been suggested. The high association of malignant tumors in these patients has been discussed, and it is proposed that a further study be undertaken to ascertain whether these patients are more prone to develop unrelated cancers.

A system for staging the clinical status of patients with soft tissue sarcomas is presented, based on the clinical characteristics of the primary tumor (size, extension), the involvement of lymph nodes, the presence of metastases, and the grade of the tumor. This represents the TNM system with grade of tumor (G) added. The system evolved was based on examination of 1215 cases of 13 types of soft tissues sarcomas, primarily in the extremities (fibrosarcoma, liposarcoma, etc.). Nine stages are described, and they are correlated with survival in the cases reviewed. The staging system now can be used for case evaluation for therapy determination and for intercomparison of series of patients as to incidence of different kinds of tumors, effects of treatment, and survival.

Background. Soft tissue sarcomas of the head and neck represent uncommon malignant neoplasms. With the exception of orbital and parameningeal sites, the treatment of sarcomas in the head and neck has not been standardized. The authors used a prospectively collected database of adult soft tissue sarcomas to identify prognostic factors for local control and survival. Methods. A prospectively collected database of adult soft tissue sarcoma from 1982 to 1989 was analyzed for the impact of prognostic factors on local control and survival. Factors examined included histologic type, tumor grade, size, and resection margins. Results. The overall and disease free survival at 5 years was 71 and 60%, respectively. Local control was 70% at 5 years. On univariate analysis, grade and margin status were predictors for local control. Analysis based on the Cox proportional hazard model revealed that margin status was the only significant factor in predicting local control. Grade and margin status were significant prognostic indicators for survival both on univariate analysis and in the Cox proportional hazard model. Conclusion. Patients with head and neck sarcomas should undergo wide excision with the removal of all gross disease and the acquisition of negative, microscopic surgical margins. Patients with positive margins should receive adjuvant radiotherapy for local control. High grade lesions place patients at risk for local recurrence and distant dissemination. Investigational regimens designed to prevent metastatic disease should be performed. Cancer 1994; 74: 697‐702.