Guo-Xia Tong

PAX2: a reliable marker for nephrogenic adenoma

Nephrogenic adenoma is a rare lesion of the urinary tract. The diagnosis usually is straightforward when characteristic microscopic and clinical findings are present, and the entity is familiar. However, misdiagnosis, in particular of adenocarcinoma of the prostate gland, may occur. Immunohistochemical stains often are needed to make such a distinction, but currently available markers offered only partial help. It recently was demonstrated that nephrogenic adenoma in renal transplant patients originated from the renal tubular epithelium. This newly proved, but long sought information may be helpful in the differential diagnosis of neophrogenic adenoma. In this study, we investigated the expression of a renal transcription factor, PAX2, in 39 nonrenal transplant-related nephrogenic adenomas, 100 adenocarcinomas of the prostate gland, and 47 urothelial carcinomas of the urinary tract. A strong and distinct nuclear staining of PAX2 was found in all 39 cases of nephrogenic adenoma (100%), but not in normal prostate tissue, normal urothelium, adenocarcinomas of the prostate gland, and invasive urothelial carcinomas. Focal CD10 was detected in six of 13 nephrogenic adenomas in the superficial papillary component and in normal prostate epithelium, normal urothelium, lymphocytes, adenocarcinoma of the prostate gland, and urothelial carcinoma. There was no uroplakins detected in nephrogenic adenoma. Therefore, these findings are suggesting that nephrogenic adenoma in nonrenal transplant patients may also arise from the renal epithelium, as did the comparable lesions after transplantation. PAX2 is a specific and sensitive immunohistochemical marker in identification and differential diagnosis of nephrogenic adenoma.

Expression of PAX2 in papillary serous carcinoma of the ovary : immunohistochemical evidence of fallopian tube or secondary müllerian system origin?

PAX2 is a urogenital developmental transcription factor expressed in the Wolffian ducts, developing kidneys, and Müllerian ducts during embryonic stage. Its function in renal development is well documented and its clinical application in the diagnosis of lesions of renal origin has been reported recently. However, information on its role in the Müllerian-derived genital tract is sparse. In this study, we investigated the expression of PAX2 in human female genital tract using immunohistochemistry. We demonstrated that PAX2 was expressed specifically in the epithelial cells of fallopian tube, endometrial and endocervical glands, but not in the stromal tissues in these areas. PAX2 was detected in secondary Müllerian structures in the ovary, such as endometriotic and endosalpingiotic glands and rete ovarii, but not in ovarian surface epithelium, surface epithelium-derived inclusion cysts, stroma, or sex-cord-derived structures such as follicles, oocytes, and corpus luteum. In addition, PAX2 was detected in 67% of ovarian papillary serous carcinomas (N=36) but rarely in peritoneal malignant mesotheliomas, with two exceptions (N=54). Interestingly, the two PAX2-positive ‘peritoneal malignant mesotheliomas’ were from female patients and were positive for estrogen receptor. The significance of expression of PAX2 and estrogen receptor in these cases is under investigation. Taken together, we suggest that PAX2 is a novel Müllerian-specific epithelial marker when used in proper clinical settings. Identification of PAX2 in the majority of papillary serous carcinomas of the ovary but not in the ovarian surface epithelium or epithelium-derived inclusion cysts suggests that this malignant epithelial tumor may be directly derived from the primary or secondary Müllerian epithelium in or surrounding the ovary, rather than from the surface epithelium or its derivatives.

Expression of PAX8 in nephrogenic adenoma and clear cell adenocarcinoma of the lower urinary tract: evidence of related histogenesis?

Recent evidence has showed that nephrogenic adenoma is a true “nephrogenic” lesion derived from the proliferation of exfoliated and implanted renal tubular cells in the urinary tract, a process that closely resembles the formation of endometriosis. This new concept has led to the identification of renal transcription factor PAX2 as a diagnostic marker for nephrogenic adenoma. PAX8 is another transcription factor structurally and functionally related to PAX2. Both are cell lineage restricted transcription factors expressed in normal and neoplastic tissues of related origin, including renal tubular cells in both fetal and adult kidneys. In this study, we investigated the expression of PAX8 in nephrogenic adenoma and its mimics. We report here that PAX8 was detected in all nephrogenic adenomas (N=35) and clear cell adenocarcinoma of the lower urinary tract (N=7), but not in prostate adenocarcinoma (N=100), adenocarcinoma (N=9), squamous cell carcinoma (N=5), or urothelial carcinoma (N=48) of the urinary bladder and its variants. PAX8 was neither detected in normal urothelium of the urinary bladder nor in prostate glands and stroma. PAX2 was also detected in 2 of the 7 clear cell adenocarcinomas of the lower urinary tract. We suggest that PAX8 is an additional marker for identifying nephrogenic adenoma. Expression of PAX8 or PAX2 in both nephrogenic adenoma and clear cell adenocarcinoma of the lower urinary tract may indicate a possible related tissue origin for these 2 lesions; both may be derived from proliferating renal tubular cells in the urinary tract. In addition, detection of PAX8 or PAX2 in clear cell adenocarcinoma of the lower urinary tract is helpful in differentiating it from urothelial carcinoma and its variants and adenocarcinomas of the urinary bladder or of the prostate.

Pax8: A marker for carcinoma of Müllerian origin in serous effusions

PAX8 is a nuclear transcription factor with limited expression in normal and neoplastic tissues in a cell lineage‐dependent manner. PAX8 has been detected in embryonic Müllerian ducts, human fallopian tubes, and ovarian carcinomas. However, little is known about its expression in other areas of the female genital tract. In this study, we used immunohistochemistry (IHC) to examine PAX8 expression in the normal uterine corpus and cervix, malignant tumors arising from these sites, and malignant effusions. We reported here that PAX8 was also detected in endometrial epithelial cells and endocervical glands, with a lower expression level in the latter, but not in the stromal cells of these areas. All endometrial carcinomas (N = 52) were positive for PAX8, whereas endocervical adenocarcinomas (N = 5) and uterine leiomyosarcomas (N = 3) were negative for PAX8. PAX8 was detected in 70% (22/31) and 68.8% (11/16) of metastatic carcinomas of the ovary and endometrium in serous effusions, respectively. No PAX8 was detected in carcinomas of nongynecologic origin or noncarcinomas (N = 71) in serous effusions except in one renal‐cell carcinoma and one carcinoma of unknown primary in a woman. In addition, papillary serous carcinomas of the peritoneum (N = 10) were diffusely positive for PAX8, implying a Müllerian origin similar to malignant tumors in the female genital tract. Our findings suggest that PAX8 is an additional IHC marker for carcinomas of Müllerian origin hence we recommend including PAX8 for evaluation of malignant serous effusions in women, especially when tumors of Müllerian origin are in the differential diagnosis. Diagn. Cytopathol. 2011.

Fine-needle aspiration biopsy of primary osteosarcoma of the thyroid: Report of a case and review of the literature

Primary osteosarcoma of the thyroid is an extremely rare tumor, with only 27 well‐documented cases reported in the literature, including only one in the cytology literature. We describe here an additional case with fine‐needle aspiration biopsy findings. A 60‐year‐old woman presented with a 1‐month history of progressive midline neck swelling. CT and ultrasound demonstrated a large thyroid mass with tracheal compression. Fine‐needle aspiration biopsies were performed and showed pleomorphic spindle and epithelioid neoplastic cells, multinucleated giant cells, and scant metachromatic extracellular matrix material. Cell block sections contained minute tissue fragments with neoplastic spindle cells. Immunohistochemical stains showed the tumor cells to be positive for vimentin and negative for cytokeratins, TTF‐1, calcitonin, synatophysin, chromogranin, and S‐100 protein, suggesting a sarcoma; however, the differential diagnosis also included anaplastic thyroid carcinoma and medullary thyroid carcinoma. Tissue biopsy revealed a high‐grade spindle cell neoplasm with osteoid production, consistent with osteosarcoma of the thyroid. The patient developed a large pulmonary embolus and superior vena cava syndrome and no further surgical intervention was performed. She died 5 weeks after the initial diagnosis. Upon retrospective review, the cytologic features resemble osteosarcoma in other areas. Although cytologic features on fine‐needle aspiration biopsy may suggest a diagnosis of this rare entity, definitive diagnosis should be deferred to histologic examination. Diagn. Cytopathol. 2008; 36: 589–594.

Oncogenic PIK3CA Mutation and Dysregulation in Human Salivary Duct Carcinoma

Salivary duct carcinoma (SDC) is an aggressive malignant tumor with a high mortality, which resembles high-grade breast ductal carcinoma in morphology. The parotid gland is the most common location. Its molecular genetic characteristics remain largely unknown. We have previously reported high incidence of PIK3CA somatic mutations in head and neck squamous cell carcinoma, particularly in pharyngeal cancers. Here we examined the PIK3CA gene expression status and hotspot mutations in six cases of SDC by immunohistochemistry and genomic DNA sequencing. Immunohistochemistry showed that PIK3CA expression was elevated in all six patients with SDC. By DNA sequencing, two hotspot mutations of the PIK3CA gene, E545K (exon 9) and H1047R (exon 20), were identified in two of the six cases. Our results support that oncogenic PIK3CA is upregulated and frequently mutated in human SDC, adding evidence that PIK3CA oncogenic pathway is critical in the tumorigenesis of SDC, and may be a plausible drug target for this rare disease.

Mutations of TSHR and TP53 Genes in an Aggressive Clear Cell Follicular Carcinoma of the Thyroid

Clear cell follicular carcinoma is a rare type of thyroid cancer and some with aggressive biological behavior. The cytoplasmic clearing of the neoplastic cells has been attributed to the accumulation of various substances, such as glycogen, lipid, mucin, and thyroglobulin, or distension of mitochondria or endoplasmic reticulum. However, the molecular mechanisms responsible for the characteristic appearance of the cell cytoplasm and the biological behavior remain unknown. We report here a case of aggressive clear cell follicular carcinoma of the thyroid with molecular profile using targeted next generation sequencing (NGS) that presented as a metastatic tumor in a woman with a history of breast carcinoma. The NGS data revealed the coexisting of a well-characterized loss-of-function TP53 R248Q mutation and a putative gain-of-function mutation of TSHR L272V, which was suggested by the overexpression of thyroglobulin and SLC5A5 (NIS) genes in this tumor. TP53 mutations are usually related with dedifferentiation, progression, and metastasis of thyroid carcinomas. Identification of TP53 R248Q in this tumor correlated with its aggressive clinical behavior. Gain-of-function mutation of TSHR can overstimulate the thyroid follicular cells as the elevated level of TSH does and might have contributed to the development of clear cell morphology in this tumor. This report represents the first case of clear cell follicular carcinoma of the thyroid with NGS analysis and more molecular characterization is needed to elucidate the pathogenesis and provide more prognosis-relevant information for this uncommon variant of thyroid carcinomas.

Validity and Reliability of Using a Self-Lavaging Device for Cytology and HPV Testing for Cervical Cancer Screening: Findings from a Pilot Study

Self-sampling could increase cervical cancer screening uptake. While methods have been identified for human papillomavirus (HPV) testing, to date, self-sampling has not provided adequate specimens for cytology. We piloted the validity and reliability of using a self-lavaging device for cervical cytology and HPV testing. We enrolled 198 women in New York City in 2008–2009 from three ambulatory clinics where they received cervical cancer screening. All were asked to use the Delphi Screener™ to self-lavage 1–3 months after clinician-collected index cytological smear (100 normal; 98 abnormal). Women with abnormal cytology results from either specimen underwent colposcopy; 10 women with normal results from both specimens also underwent colposcopy. We calculated sensitivity of self-collected cytology to detect histologically confirmed high grade lesions (cervical intraepithelial neoplasia, CIN, 2+); specificity for histology-negative (CIN 1 or lower), paired cytology negative, or a third cytology negative; and kappa for paired results. One hundred and ninety-seven (99.5%) women self-collected a lavage. Seventy-five percent had moderate to excellent cellularity, two specimens were unsatisfactory for cytology. Seven of 167 (4%) women with definitive results had CIN2+; one had normal and six abnormal cytology results with the self-lavage (sensitivity = 86%, 95% Confidence Interval, CI: 42, 100). The kappa for paired cytology was low (0.36; 95% CI: 0.25, 0.47) primarily due to clinician specimens with atypical squamous cells of undetermined significance (ASC-US) and low grade squamous intraepithelial lesion (LSIL) coded as normal using Screener specimens. However, three cases of HSIL were coded as ASC-US and one as normal using Screener specimens. Seventy-three women had paired high-risk HPV tests with a kappa of 0.66 (95% CI: 0.49, 0.84). Based on these preliminary findings, a larger study to estimate the performance of the Screener for co-testing cytology and HPV or for HPV testing with cytology triage is warranted. Trial Registration ClinicalTrials.gov NCT00702208

The differential expression of PAX2 and PAX8 in the ovarian surface epithelium and fallopian tubal epithelium is an important issue.

To the Editor: We have read with considerable interest the recent paper in the American Journal of Surgical Pathology entitled “PAX2 and PAX8 Expression in Primary and Metastatic Müllerian Epithelial Tumors: A Comprehensive Comparison” by Ozcan et al.12 The authors concluded that PAX8 was more sensitive and useful than PAX2 in the diagnosis of primary and metastatic carcinomas of Müllerian origin, which is consistent with the results reported from other groups. However, we do have some concerns about the reported expression of PAX2 and PAX8 in the ovarian surface epithelium and ovarian epithelial inclusion cysts in this paper in relation to what has been reported by other groups. In particular, we note descriptions of their findings in the text and the summary table where clarification is merited. In the text, it was stated that “the expression of PAX2 and PAX8 was virtually identical” and “the ovarian surface epithelial cells showed staining; however, the staining was focal.” However, in Table 1, it appeared that all cases (8/8) and all surface epithelial cells (100%) were positive for PAX2, and most cases (5/8) and most of the surface epithelial cells (63%) were positive for PAX8. These findings are significantly different from the reported findings in the literature, in which the ovarian surface epithelium was usually negative for PAX8 and PAX2.1,14 In addition, the findings of expression of PAX2 and PAX8 in the ovarian epithelial inclusion cysts in this paper also are different from those of others. These discrepancies have not been adequately discussed in the papers. Several studies have shown that PAX2 and PAX8 are differentially expressed in the ovarian surface epithelium and the fallopian tubal epithelium. The fallopian tubal epithelium, especially the serous epithelial cells, is always positive for PAX2 and PAX8, whereas the ovarian surface epithelium is usually negative for PAX2 and PAX8.1,8,14 However, the findings of Ozcan et al suggest that there is no difference in PAX2 and PAX8 expression in these 2 types of epithelia. Although we have occasionally detected PAX2 and PAX8 in the ovarian surface epithelium, the frequency was much lower than that described by Ozcan et al,12 who stated that 100% and 63% of the cases were positive for PAX2 and PAX8, respectively. It is important to note that in our cases the PAX2-positive and PAX8-positive areas of the ovarian surface epithelium always demonstrated histologic features of fallopian tubal epithelium, such as serous or ciliated epithelial cells, as shown in Figure 1. The Müllerian epithelium and the ovarian surface epithelium are purported to be derived from embryonic coelomic epithelium. However, the Müllerian epithelium is derived from a subset of cells on the coelomic epithelium adjacent to the mesonephros.5 These Müllerian precursor cells are defined by paracrine or cellular signals emitted from embryonic kidneys and are distinct from the rest of the coelomic epithelium, appearing

Uterine Carcinosarcomas: Clinical, Histopathologic and Immunohistochemical Characteristics

Carcinosarcomas (malignant mixed Müllerian tumors or MMMT) are rare malignant tumors in the female genital tract composed of both malignant epithelial and malignant mesenchymal components. They comprise <5% of all neoplasms in the gynecologic tract and have an aggressive clinical course. The purpose of this study is to evaluate the immunophenotype and possible histogenesis of carcinosarcomas of the uterus. Sixty-two cases of uterine carcinosarcomas diagnosed between 1995 and 2011 were retrieved from the gynecologic pathology files at Columbia University Medical Center. Representative tissue blocks containing both epithelial and mesenchymal components were selected from each case for histologic and immunohistochemical studies. Clinical data from each case were retrieved. The epithelial component was poorly differentiated adenocarcinoma in the majority (80.7%) of cases; in 17.7%, the carcinoma was moderately differentiated, and in only 1.6% the carcinoma was well differentiated. 53% of the tumors had homologous stromal elements and 47% displayed heterologous stromal elements. Immunohistochemical study revealed almost equal staining in both epithelial and mesenchymal components of carcinosarcomas for p16 and p53. PAX8 positivity was noted in 73% of epithelial components, but only 13% of stromal components, and PAX8 stromal positivity was never seen in the absence of PAX8 epithelial positivity. Expression of p16, p53, and PAX8 in both malignant components lends support to the monoclonal theory of uterine carcinosarcoma tumorigenesis. The roles of these tumor markers in the diagnosis and pathogenesis of this tumor and associations between clinical characteristics, tumor pathologic features, and prognosis are discussed.