We describe the histologic variants of papillary and follicular carcinomas associated with 109 spindle and giant cell carcinomas (SGCC) of the thyroid and determine the incidence of rhabdoid and thyroglobulin inclusions in these tumors. In addition, we searched for rhabdoid and thyroglobulin inclusions in 120 papillary carcinomas (PC) (all 15 variants included), 23 differentiated follicular carcinomas (DFC), (6 with insular pattern), 6 poorly differentiated follicular carcinomas (PDFC) and 34 follicular adenomas (FA). The following differentiated thyroid carcinomas coexisted with SGCC: 51 (46.8%) PC, (34 conventional type, 14 tall cell variant and 3 follicular variant), 6 (5.5%) DFC, 1 follicular carcinoma with insular pattern (0.9%), and 3 oncocytic carcinomas (2.8%). Eleven SGCC (10%) and 2 (33%) PDFC showed rhabdoid features, but lacked thyroglobulin inclusions. Thyroglobulin inclusions were found in 10 FA (29%), 8 (17%) follicular variants of PC and in 7 (30.4%) DFC. There were no rhabdoid inclusions in any of these differentiated thyroid tumors. Our findings support the hypothesis that most SGCC result from dedifferentiation or anaplastic transformation although the mechanisms that underlie this transformation remain unknown. The finding that only 1 (0.9%) SGCC was associated with follicular carcinoma with insular pattern contradicts the opinion that this tumor occupies an intermediate position between differentiated and anaplastic carcinomas. Rhabdoid features are markers of PDFC and SGCC while thyroglobulin inclusions are markers of FA and differentiated thyroid carcinomas with follicular phenotype.

Pseudohyperplastic prostatic adenocarcinoma is a recently described variety of adenocarcinoma that has been studied in core-needle biopsies and prostatectomy specimens. It is characterized by malignant glands that simulate benign hyperplastic glands with complex, medium to large-sized glands with papillary infoldings, luminal undulations, branching or cystic dilatations, and columnar cells with macronucleoli and nuclear enlargement. Our aim was to define frequency, tumor volume, and histologic features of pseudohyperplastic prostatic adenocarcinoma in transurethral resections of prostate. We studied 250 specimens from transurethral resections; 150 specimens were originally diagnosed as benign glandular hyperplasia, and 100 as conventional prostate adenocarcinomas. Of the 150 biopsies originally diagnosed as benign glandular hyperplasia, two (1.3%) had areas of pseudohyperplastic carcinoma. In both cases the neoplasm was limited to two chips and measured 3 and 4 mm in diameter, respectively. Both patients were asymptomatic 2 and 4 years after diagnosis. Of the 100 biopsies with adenocarcinoma, areas of pseudohyperplastic carcinoma were found in three cases. In the first two these areas were found in two fragments, and in the other case they were found in three chips, and measured 3, 4, and 6 mm, respectively. The clinical course in these cases was unfavorable, and two patients had metastasis. Main histologic findings included crowded glands (5/5), papillary projections (5/5), nuclear enlargement (5/5) macronucleoli (4/5) cystic glandular dilatation (4/5) straight luminal borders (4/5), pink amorphous secretions (4/5) nuclear hyperchromasia (3/5) and transition to small acinar pattern of adenocarcinoma (3/5). In conclusion, pseudohyperplastic prostate carcinoma is rare in transurethral resection specimens and is found in scarce chips. Frequency of false negative results in biopsies originally diagnosed as benign glandular hyperplasia was 1.3%. In biopsies diagnosed as carcinoma, this frequency was 3%. These patients had an adverse clinical course, apparently due to association with areas of conventional adenocarcinoma.