Search

The clinical and histological features of 13 malignant melanomas in children less than 13 years of age in New South Wales, Australia, were compared with those in a control group of children with 15 Spitz nevi, 4 of which were considered atypical, and 2 unusual compound nevocellular nevi. Six of the controls had been previously diagnosed histologically as malignant melanoma. The objective observations made by one or more histopathologists experienced in reporting melanocytic lesions, and the clinical details, mainly from the Sydney Melanoma Unit files, were entered on a detailed protocol. Evaluation was assisted by the use of SPSS-X software on a mainframe VAX computer. Six of the 13 children with malignant melanoma died with their disease. The most frequent clinical features found in the malignant melanomas were bleeding, ulceration, itching, and black or variegated color. Recent enlargement and darkening were noted in the majority of both the malignant melanomas and the Spitz nevi. Histological features favoring malignancy in this series were mitoses within 0.25 mm of the dermal margin of the melanoma, a dermal mitotic rate exceeding 2/mm2, ulceration, surface exudate, large pigment granules, and clear-cell differentiation. The median thickness of the malignant melanomas was 1.3 mm but in the 4 children who died with melanoma the median thickness was 2.9 mm. Absence of mitoses, predominance of spindle cells, and diffuse maturation favored Spitz nevus. The median thickness of the Spitz nevi was 0.7 mm.

A series of 60 cases of prostatic adenocarcinoma and 34 cases of benign prostatic hyperplasia were examined quantitatively after immunoperoxidase staining for prostate-specific antigen (PSA), prostate-specific acid phosphatase (PSAP), carcino-embryonic antigen (CEA), epithelial membrane antigen (EMA), alpha fetoprotein (AFP), and human chorionic gonadotrophin (HCG). The tumors were graded I to IV according to the MDAH grading system recently proposed. Fifty-nine of the 60 tumors were positive for PSA and 58 were positive for PSAP. The one PSA and PSAP negative case was CEA negative and weakly EMA positive. Grade I to III tumors stained more tumor cells and more diffusely for PSA and PSAP than grade IV tumors. There was no significant difference in the intensity or extent of staining between grade I and grade II-III tumors for PSA and PSAP. A comparison of PSA and PSAP showed that PSA stained more intensely and more extensively than PSAP. Benign prostatic tissue and low-grade prostatic tumors did not stain for CEA but three of the 20 grade IV tumors and one of the 23 grade II-III tumors did. Staining for EMA was focal and showed no relation to tumor grade. Benign and malignant lesions failed to stain for AFP and HCG.

A retrospective review of Hodgkin' disease at a general referral center from 1952 to 1971 was carried out. Available pretreatment biopsies were classified by the modified Lukes-Butler classification into histologic types by a single pathologist, who then reviewed the same material 6 months later without the knowledge of his previous conclusions or patient identification. There was 95% concordance between original and review classifications. The frequency distribution of age of diagnosis and sex were unchanged over the time period of the study. The frequency distribution of histologic types was also unchanged when 5-year intervals were compared. However, there was no case of the lymphocytic-predominance type in the first half of the study period, which was significantly different from the incidence of this type in the second half of the study period. Survival rates showed no improvement in overall patient survival over the time period of the study but significant differences in the survivals of patients with different histologic types were noted.

Two hundred and eighty-six patients presenting with metastatic adenocarcinoma or undifferentiated carcinoma whose primary site was not identified by clinical history, physical examination and chest radiograph have been studied. Median survival from presentation was 22 weeks. Factors independently predicting improved survival were lymph node presentations, good performance status and body weight loss of less than 10 per cent. In 88 (31 per cent) patients the primary tumour site was subsequently identified, in 58 (20 per cent) during life. Lung cancer was the most frequently identified primary tumour, and in only 32 (11 per cent) of the patients was a 'treatable' primary tumour (i.e. germ cell, breast, ovarian, prostate, thyroid cancer or lymphoma) identified. Among the treatable primary tumours were those in eight out of 16 female patients presenting with axillary metastases who were subsequently shown to have primary breast cancer and four of 13 females presenting with ascites who were found to have primary ovarian cancer. Prostatic cancer was confirmed in five out of 13 men with raised serum acid phosphatase. Of 22 patients with elevated serum alphafoetoprotein (AFP) or beta-human chorionic gonadotrophin levels (beta HCG) 18 had some features of the 'atypical teratoma syndrome'. Of the total of 32 patients with treatable tumour types, 29 (90 per cent) were identified during life. Median survival for patients with treatable tumour types identified during life was 104 weeks, compared with 22 weeks for the group as a whole. Retrospective immunocytochemical staining of the original biopsy showed that prostatic specific antigen and antibodies to beta HCG and AFP were diagnostically useful, but a series of organ site non-specific markers of histogenesis or cellular differentiation (carcinoembryonic antigen, secretory component for IgA, peanut lectin binding, epithelial membrane antigen and keratin) showed no significant correlations with identified primary sites, responsiveness to empirical chemotherapy or survival. Metastatic undifferentiated carcinoma or adenocarcinoma from an unknown primary site represents 6.5 per cent of all referrals to the medical oncology unit, Royal Prince Alfred Hospital, Sydney. We offer guidelines for the rapid identification of the limited number of primary sites for which effective and specific forms of systemic treatment are available.

Many articles concerning the accuracy of diagnosis of malignant tumor cells in serous effusions have appeared in the literature, but few authors have attempted to describe the characteristics of these cells as they relate to tumors of specific primary sites. Still fewer have dealt with determination of the site of the primary tumor by examination of effusions. In this paper, 87 well-documented cases of pleural and peritoneal serous effusions containing malignant tumor cells derived from carcinoma of the breast, lung, ovary, endometrium, and stomach were examined and characterized. Sufficient cytologic similarities occurred to permit classification into three distinct patterns in 84.6% of the cases. Grouping the cases according to these patterns and combining this with knowledge of the patient’s sex and the site of effusion allowed the pathologist to determine the site of the primary tumor with as much as 96% accuracy. Criteria of classification, as well as pitfalls in evaluation, are discussed in detail.

Of 781 cervical squamous cell carcinomas, 66 or 8.4% were microinvasive cancers. Analysis indicated a progressive decrease in the incidence of outspoken cancer and an increasing frequency of microinvasive cancer. The mean age at detection for microinvasive cancer was 46.0 ± 12.7 years. Most women were asymptomatic at the time of detection and the cervical findings were either normal or nonspecific. Cellular studies contributed to the detection of microinvasive cancer. The microinvasive neoplasms did not exceed a depth of 5.0 mm. and cervical findings were either normal or nonspecific. Microinvasion originated predominantly from carcinoma in situ and/or dysplasia in the transitional zone of the anterior lip of the cervix. The extent of microinvasion increased with advancing age. The distribution and histologic features varied, depending upon the extent of microinvasion. Carcinoma in situ and coexisting carcinoma in situ and dysplasia were the predominant epitheliums of origin for microinvasive cancer. Dysplasia alone was infrequent. Certain abnormal surface and glandular epithelial features not commonly associated with carcinoma in situ and dysplasia were observed. The severity of these changes was related to the extent of the microinvasion. The 5-year survival for microinvasive cancer was 95.8%, No residual neoplasm was identified at autopsy on those who died. The remaining women are living and are free of neoplasm after surveillance of 1 to 21 years.