College of American Pathologists
Printable Version

  Anatomic Abstracts





cap today

June 2001

Diagnostic criteria for distinguishing between invasive peritoneal implants and noninvasive implants
Criteria for distinguishing invasive peritoneal implants from noninvasive implants vary among investigators and may be difficult to apply in individual cases. Further, criteria for invasive and noninvasive implants need to be accurately established to conduct randomized clinical trials comparing the efficacy of therapy for tumors in those with and without such implants. In this study, the authors examined 148 implants from 60 patients to identify histologic features that correlate with adverse clinical outcome. They used the following criteria to separate invasive from noninvasive implants: invasion of underlying normal tissue, micropapillary architecture, and solid epithelial nests surrounded by clefts. If any of these features were present, the implants were classified as invasive; those lacking all three features were classified as noninvasive. Among the 31 patients with invasive implants, six were dead of disease, 13 were alive with progressive disease, and 12 were alive with no evidence of disease. Among the 29 patients with noninvasive implants, two were dead of disease, one was dead of uncertain causes, one was alive with progressive disease, and 25 were alive with no evidence of disease. Using these criteria, 49 implants classified as “invasive” had underlying normal tissue. Only seven of these, however, invaded the underlying tissue, and 42 were “on the surface.” The authors suggested that using the term “invasive” to classify the implants associated with an adverse clinical outcome is inaccurate since other histologic features (micropapillary architecture, solid epithelial nests surrounded by clefts) appear to be more sensitive than “invasion into underlying normal tissue.” The authors proposed that such implants, as defined by their criteria, be designated "well differentiated serous carcinoma." In contrast to other studies, this investigation did not find cytologic atypia and mitotic activity to be useful in classifying implants.

Bell KA, Sehdev AES, Kurman RJ. Refined diagnostic criteria for implants associated with ovarian atypical proliferative serous tumors (borderline) and micropapillary serous carcinomas. Am J Surg Pathol. 2001;25:419-432.

Reprints: Dr. Robert J. Kurman, The Johns Hopkins Hospital, Dept. of Pathology, 401 N. Broadway, Pathology 2242, Baltimore, MD 21231

Features of benign and malignant phyllodes tumors of the breast
The authors attempted to correlate histological, immunohistochemical, and molecular features of 20 phyllodes tumors (seven benign, seven low-grade malignant, and six high-grade malignant) with their clinical behavior. Immunohistochemistry for Ki-67 and p53 and PCR-based loss of heterozygosity studies were performed to determine whether deletions at FHIT and hMLH1 loci on chromosome 3p play a role in clinical behavior. The patients ranged in age from 13 to 71 years (median, 51 years). No metastases were identified in any group after a mean followup of eight years. Three tumors recurred locally and all were in the histologically benign group. Although the Ki-67-labeling index correlated with the histological appearance of malignancy and p53 was more likely to be positive in low-grade malignant and high-grade malignant tumors than benign tumors, neither of these features predicted subsequent behavior. 3p deletions were also found not to be a factor. Margins of resection were greater than 1 cm in all but four cases, including the three cases that recurred locally. Histological features and the various biological markers addressed in this study did not predict biological behavior but margin status did. This study highlights the importance of achieving adequate margins in the resection of phyllodes tumors of the breast, regardless of histological appearance.

Kleer CG, Giordano TJ, Braun T, et al. Pathology, immunohistochemical, and molecular features of benign and malignant phyllodes tumors of the breast. Mod Pathol. 2001;14;185-190.

Reprints: Dr. Celina G. Kleer, Dept. of Pathology, University of Michigan, 1500 E. Medical Center Drive, Room 2G332/Box 0054, Ann Arbor, MI 48109-0054

Overexpression of HER-2/neu in solid tumors
The authors conducted a study in which 575 primary neoplasms were assessed for HER-2/neu overexpression using immunohistochemistry on paraffin-embedded material. Overexpression was defined as a score of 2+ or greater using the HERcepTest. By these criteria, 22 percent of infiltrating ductal carcinomas of the breast were positive, as were 28 percent of pulmonary adenocarcinomas, 17 percent of colorectal adenocarcinomas, 13 percent of transitional cell carcinomas of the urinary bladder, and 11 percent each of pulmonary squamous carcinomas and gastric adenocarcinomas. Other tumor types had a lower percentage of positive cases. When 3+ staining was considered, the percentage of breast tumors showing this feature remained the highest of any tumor (12 percent), and the majority of other tumor types showed considerably lower percentages. One exception was adenocarcinoma of the stomach, in which eight percent of tumors showed 3+ staining. The authors noted that their results for prostate and pancreas were lower than those identified in the literature. They also pointed out that granular cytoplasmic staining may be seen in cells of neuroendocrine differentiation, a finding they believe represents a HER-2-related epitope rather than HER-2 itself. The authors stressed the necessity for interpreting only cell membrane staining as positive. Because specific therapy directed at HER-2/neu overexpressing breast cancers has shown efficacy, similar strategies may prove useful in some over-expressing nonbreast malignancies. But such studies are still inconclusive, and the role of Herceptin therapy for nonmammary malignancies has yet to be defined.

Koeppen HKW, Wright BD, Burt AD, et al. Overexpression of HER-2/neu in solid tumors: an immunohistochemical survey. Histopathology. 2001;38:96-104.

Reprints: Dr. Hartmut Koeppen, Genentech Inc., 1 DNA Way, Mail Stop #72B, South San Francisco, CA 94080

Specific therapy for malignant GI stromal tumors may be directed by immunohistochemistry for c-kit
The authors described how a patient with metastatic gastrointestinal stromal tumor was treated with the tyrosine kinase inhibitor ST1571. This drug has shown activity against the BCR-ABL-induced overexpression of tyrosine kinase in patients with advanced and blast-phase chronic myelogenous leukemia. It also apparently is active against the tyrosine kinase constitutively expressed in GI stromal tumors due to mutated c-kit (CD117), a growth-factor receptor on the surface of GI stromal tumor cells. CD117 is easily detected by immunohistochemistry using commercially available antibodies. The majority of GI stromal tumors are found in the stomach. Once metastatic, malignant GI stromal tumors have a dismal prognosis. In this case report, however, ST1571 caused objective evidence of tumor regression, including regression of liver metastasis, and the response continued for more than 11 months. Toxicity was considered minimal. This is an example of an effective therapy directed against a specific molecular target that can be identified by the pathologist using conventional immunohistochemical techniques. It is another example of how pathologists can provide information, beyond traditional tumor diagnosis and staging, that benefits patient care.

Joensuu H, Roberts PJ, Sarlomo-Rikala M, et al. Effect of the tyrosine inhibitor ST1571 in a patient with a metastatic gastrointestinal stromal tumor. N Engl J Med. 2001;344:1052-1056.

Reprints: Dr. H. Joensuu, Dept. of Oncology, Helsinki University Central Hospital, Haartmaninkatu 4, P.O. Box 180, FIN-00029, Helsinki, Finland